CN114708268A - Hardware part defect detection method and system - Google Patents
Hardware part defect detection method and system Download PDFInfo
- Publication number
- CN114708268A CN114708268A CN202210637709.7A CN202210637709A CN114708268A CN 114708268 A CN114708268 A CN 114708268A CN 202210637709 A CN202210637709 A CN 202210637709A CN 114708268 A CN114708268 A CN 114708268A
- Authority
- CN
- China
- Prior art keywords
- defect
- hardware part
- picture
- detected
- hardware
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 757
- 238000001514 detection method Methods 0.000 title claims abstract description 219
- 230000002950 deficient Effects 0.000 claims abstract description 141
- 238000012545 processing Methods 0.000 claims abstract description 73
- 238000000034 method Methods 0.000 claims abstract description 34
- 230000008859 change Effects 0.000 claims abstract description 7
- 241000272168 Laridae Species 0.000 claims description 57
- 238000000605 extraction Methods 0.000 claims description 42
- 238000012795 verification Methods 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 18
- 238000002372 labelling Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 8
- 238000004519 manufacturing process Methods 0.000 description 26
- 230000009286 beneficial effect Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 6
- 230000002708 enhancing effect Effects 0.000 description 6
- 238000001914 filtration Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 239000000725 suspension Substances 0.000 description 6
- 239000000463 material Substances 0.000 description 5
- 238000003754 machining Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 208000003464 asthenopia Diseases 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Processing (AREA)
Abstract
The invention provides a hardware part defect detection method and a hardware part defect detection system, wherein the detection method comprises the steps of collecting a hardware part standard picture and a defect sample picture image corresponding to a defective hardware part, and carrying out pixel change processing and graying processing on the image; constructing a three-dimensional hardware part model, and extracting characteristic points of the three-dimensional hardware part model to obtain a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model; and determining the position and the type of the defect by using a region shape positioning method. The system comprises a module corresponding to the step.
Description
Technical Field
The invention provides a hardware part defect detection method and system, and belongs to the technical field of part defect detection.
Background
In modern society, hardware parts have wide application in chemical industry, electronics, aerospace and other industries and are ubiquitous in our lives. In an industrial environment developing at a high speed, people pay more attention to the appearance quality of hardware parts, so that detecting defects of the parts is one of indispensable links in the processing industry, and the people are forced to carry out stricter quality detection on products. The traditional detection method is human eye detection, the method is not only low in efficiency, but also has the problem of error detection or missed detection easily caused by visual fatigue of workers, even if the workers find the proof defects of hardware parts, the hardware parts can be reprocessed only by means of naked eye positioning, secondary processing defects are easily caused, and if the workpieces are directly abandoned, resource waste is caused. Therefore, the human eye cannot continuously and stably complete the repetitive quality inspection work, and finding a new product defect inspection method to replace human eye inspection becomes one of the problems that many enterprises need to solve urgently.
Disclosure of Invention
The invention provides a hardware part defect detection method and system, which are used for solving the problem of hardware part defect certification:
the invention provides a hardware part defect detection method and a hardware part defect detection system, which are characterized in that the hardware part defect detection method is characterized by comprising the following steps:
s201, collecting a standard picture of the hardware part and a defect sample picture corresponding to the defective hardware part, and obtaining images of the standard picture of the hardware part and the defect sample picture corresponding to the defective hardware part by utilizing pixel change processing and graying processing;
s202, constructing a three-dimensional hardware part model, and extracting feature points of the three-dimensional hardware part model to obtain a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model;
s203, detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, and acquiring the defect type and the defect area corresponding to the detected picture of the hardware part;
s204, recognizing the regional edge line of the defect region, utilizing the regional edge line to acquire the regional shape, utilizing the regional shape is right the defect region of the detected picture of the hardware part is positioned, the defect type and the position of the detected picture of the hardware part are obtained, and the defect type and the position are displayed through an intelligent terminal.
Further, gather hardware component standard picture and correspond defective sample picture of defective hardware component, utilize pixel change to handle and graying out to handle and acquire will acquire hardware component standard picture and the image that corresponds defective hardware component's defective sample picture include:
acquiring a complete standard picture of the hardware part and a corresponding defect sample picture of the defective hardware part through an industrial camera, labeling the defect sample picture and defining the defect type; carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and a defect sample picture of the corresponding defective hardware part; acquiring a low-pixel image of the hardware part standard picture and a defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
graying the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
improving pixels of the hardware part standard picture and the gray level image of the defect sample picture corresponding to the defective hardware part, and acquiring a high pixel image of the hardware part standard picture and the high pixel image of the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the high pixel image of the defect sample picture of the corresponding defective hardware part reach 800 multiplied by 600;
wherein, the graying image that obtains hardware component standard picture and the defect sample picture that corresponds defective hardware component includes:
acquiring the input hardware part standard picture and a low-pixel image of a defect sample picture corresponding to the defective hardware part; calculating a graying weighting coefficient of the image based on a gull algorithm; carrying out graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part by using an image graying weighting coefficient to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
calculating the grayed weighting coefficient of the image based on a seagull algorithm, and initializing the seagull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
Further, constructing a three-dimensional hardware part model, and extracting feature points of the three-dimensional hardware part model to obtain a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model, wherein the defect detection model comprises the following steps:
establishing a hardware part three-dimensional model by utilizing a characteristic modeling mode through the acquired hardware part standard picture and high-pixel image information of a defect sample picture corresponding to the defective hardware part; acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures, and determining defect information corresponding to each defect picture so as to generate a training data set and a verification data set; initializing the defect detection model by using the training data set, fixing the weight parameters of the generated comparison network, respectively carrying out discrimination processing and generation processing, and then enabling the loss function value of the generated comparison network to be within a preset threshold range to obtain the trained defect detection model;
inputting the three-dimensional model of the hardware part into the defect detection model, carrying out defect detection identification on the three-dimensional model by using the defect detection model to obtain a defect detection result corresponding to the three-dimensional model, and comparing the defect detection result with corresponding defect data in the verification data set to obtain an error value between the defect detection result and the corresponding defect data in the verification data set;
when the obtained error value is within a preset error range, the defect detection precision is proved to meet the standard, and then a defect detection model is output;
and initializing the defect detection model, acquiring the image information of the detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model.
Further, the defect detection model is utilized to detect and identify the image information of the detected picture of the hardware part, and the defect type and the defect area corresponding to the detected picture of the hardware part are acquired, wherein the defect type and the defect area comprise:
acquiring image information of a defect sample picture of the hardware part, and detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model to identify a defect type corresponding to the defect sample picture of the hardware part;
classifying the defects according to the defect types to obtain defect classification results corresponding to the defect sample pictures of the hardware parts;
performing region extraction on a defect sample picture of the hardware part by using the defect classification result to obtain a defect region in the defect sample picture;
and comparing and identifying the defect area by using the three-dimensional model of the hardware part to obtain defect range information and further obtain the defect area.
Further, the regional margin line of defect region is discerned, and utilizes regional margin line obtains regional shape, utilizes regional shape is right the defect region that hardware component detected the picture is fixed a position, obtains the defect type and the position that hardware component detected the picture to show through intelligent terminal and include:
positioning the input detected picture of the hardware part to obtain the information of the detected picture of the hardware part;
identifying a defect area of the detected picture information of the hardware part through a defect identification model, and sorting by adopting an area edge line to obtain the area shape of the area;
positioning the defect area of the detected picture of the hardware part by using the area shape to obtain the position information and the category of the defect area of the detected picture of the hardware part, and displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the detection of the edge comprises:
determining a filtering range, enhancing edge lines, detecting edge areas and positioning the positions of the edge lines;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the type and position of the detected picture template defect of the hardware part through sub-pixel precision positioning;
the method for storing and displaying the defect types and the positions of the hardware parts to be detected on the intelligent terminal comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of a picture to be detected; acquiring an image displayed by the picture to be detected, and determining a display mode of the picture to be detected in the display area of the picture to be detected according to the defect type and the position of the image to be displayed of the picture to be detected and the defect type and the position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
Further, the hardware part defect detection system comprises:
the system comprises an image acquisition module, a modeling extraction module, a defect extraction module and a defect identification module;
the image acquisition module is used for acquiring a hardware part standard picture and a defect sample picture corresponding to a defective hardware part, and acquiring an image of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part by utilizing pixel change processing and graying processing;
the modeling extraction module is used for constructing a three-dimensional model of the hardware part, extracting characteristic points of the three-dimensional model of the hardware part and obtaining a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model;
the defect extraction module is used for detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, and acquiring the defect type and the defect area corresponding to the detected picture of the hardware part;
the defect identification module identifies the regional margin line of the defect area, utilizes the regional margin line obtains the regional shape of region, utilizes regional shape is right the defect area that hardware component detected the picture is fixed a position, obtains the hardware component is detected the defect type and the position of picture to show through intelligent terminal.
Further, the image acquisition module comprises:
the low-pixel module is used for acquiring a standard picture of a good hardware part and a defect sample picture of a corresponding defective hardware part through an industrial camera, labeling the defect sample picture and defining a defect type; carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and a defect sample picture of the corresponding defective hardware part; acquiring a low-pixel image of the hardware standard picture and a defect sample picture corresponding to the defective hardware; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
the graying module is used for performing graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
the high pixel module is used for improving pixels of the hardware part standard picture and the gray level image of the defect sample picture corresponding to the defective hardware part, and acquiring high pixel images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the high pixel image of the defect sample picture of the corresponding defective hardware part reach 800 multiplied by 600;
wherein the graying module comprises:
the gull algorithm module is used for calculating a graying weighting coefficient of the image and initializing the gull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
Further, the modeling extraction module includes:
the three-dimensional module is used for establishing a three-dimensional model of the hardware part by utilizing a characteristic modeling mode through the acquired high-pixel image information of the standard hardware part picture and the defect sample picture corresponding to the defective hardware part; acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures, and determining defect information corresponding to each defect picture so as to generate a training data set and a verification data set;
the detection model module is used for initializing the defect detection model by using the training data set, fixing the weight parameters of the generated comparison network, respectively carrying out discrimination processing and generation processing, and then enabling the loss function value of the generated comparison network to be within a preset threshold range to obtain a trained defect detection model;
the data comparison module is used for inputting the three-dimensional model of the hardware part into the defect detection model, utilizing the defect detection model to carry out defect detection identification on the three-dimensional model to obtain a defect detection result corresponding to the three-dimensional model, and comparing the defect detection result with corresponding defect data in the verification data set to obtain an error value between the defect detection result and the corresponding defect data in the verification data set;
when the obtained error value is within a preset error range, the defect detection precision is proved to meet the standard, and then a defect detection model is output;
initializing the defect detection model, acquiring image information of a detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model;
wherein the feature modeling comprises:
a feature modeler, a product database and an application interface;
the characteristic modeling device is used for generating a characteristic model of the part; a product database for storing geometric and non-geometric data; the application interface is used for data exchange with other systems.
Further, the defect extraction module includes:
the defect type module is used for acquiring image information of a defect sample picture of the hardware part, detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model, and identifying a defect type corresponding to the defect sample picture of the hardware part;
the defect region module is used for classifying defects according to the defect types to obtain defect classification results corresponding to the defect sample pictures of the hardware parts;
the area extraction module is used for carrying out area extraction on a defect sample picture of the hardware part according to the defect classification result to obtain a defect area in the defect sample picture;
and comparing and identifying the defect area by using the three-dimensional model of the hardware part to obtain defect range information and further obtain the defect area.
Further, the defect identifying module includes:
the positioning module is used for positioning the input detected picture of the hardware part to obtain the information of the detected picture of the hardware part;
the shape module is used for identifying the defect area of the detected picture information of the hardware part through a defect identification model and sorting by adopting an area edge line to obtain the area shape of the area;
the display module is used for positioning the defect area of the detected picture of the hardware part according to the area shape, obtaining the position information and the type of the defect area of the detected picture of the hardware part, and displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the sorting of the region edge lines comprises:
determining a filtering range, enhancing edge lines, detecting edge areas and positioning the positions of the edge lines;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the type and position of the detected picture template defect of the hardware part through sub-pixel precision positioning;
the method comprises the following steps of displaying the defect type and the defect position of the hardware part to be detected on the intelligent terminal, wherein the method comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of a picture to be detected; acquiring an image displayed by the picture to be detected, and determining a display mode of the picture to be detected in the display area of the picture to be detected according to the defect type and the position of the image to be displayed of the picture to be detected and the defect type and the position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
The invention has the beneficial effects that:
according to the hardware part defect detection method and system, the three-dimensional modeling of the hardware part is established through the acquired standard picture of the hardware part and the image of the improved pixel of the corresponding defective sample picture, the defect detection model is established through the three-dimensional modeling, the shape, the defect type and the position of the defect area are determined and determined through the edge line, the defect type and the position of the hardware part to be detected are stored and displayed on the intelligent terminal, the automatic detection of the hardware part proof defect is realized, the detection efficiency of the hardware part proof defect is improved, and the manpower, the material resources and the financial resources are saved.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a diagram illustrating the steps of locating the types and corresponding locations of defects according to the present invention;
fig. 3 is a system block diagram of the system of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
The embodiment of the invention provides a hardware part defect detection method, which comprises the following steps of:
s201, collecting a standard picture of the hardware part and a defect sample picture corresponding to the defective hardware part, and obtaining images of the standard picture of the hardware part and the defect sample picture corresponding to the defective hardware part by utilizing pixel processing and gray processing;
s202, constructing a three-dimensional hardware part model, and extracting feature points of the three-dimensional hardware part model to obtain a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model;
s203, detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, and acquiring the defect type and the defect area corresponding to the detected picture of the hardware part;
s204, recognizing the regional edge line of the defect region, utilizing the regional edge line to acquire the regional shape, utilizing the regional shape is right the defect region of the detected picture of the hardware part is positioned, the defect type and the position of the detected picture of the hardware part are obtained, and the defect type and the position are displayed through an intelligent terminal.
The working principle of the technical scheme is as follows: collecting a standard picture of the hardware part and a defect sample picture corresponding to the defective hardware part, firstly reducing picture pixels of the obtained standard picture of the hardware part and the defect sample picture corresponding to the defective hardware part to obtain a low-pixel image of the standard picture of the hardware part and the defect sample picture corresponding to the defective hardware part, then carrying out gray processing on the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain a gray image of the hardware part standard picture and the gray image of the defect sample picture corresponding to the defective hardware part, then improving pixels of the gray images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part, and obtaining a high-pixel image of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
according to the acquired hardware part standard picture and high-pixel image information of a defect sample picture corresponding to a defective hardware part, constructing a hardware part three-dimensional model by using the hardware part standard picture and the high-pixel image information of the defect sample picture corresponding to the defective hardware part, extracting feature points of the hardware part three-dimensional model, and performing model training by using the feature points of the hardware part three-dimensional model to obtain a defect detection model of the hardware part; initializing the defect detection model, acquiring image information of a detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model;
detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, acquiring the defect type corresponding to the detected picture of the hardware part, and performing defect type according to the type to acquire a defect classification result corresponding to the detected picture of the hardware part; performing region extraction on the detected hardware part picture by using the defect classification result to obtain a defect region in the defect sample picture;
identifying the area edge line of the defect area, acquiring the area shape of the area by using the area edge line, positioning the defect area of the detected picture of the hardware part by using the area shape, and acquiring the position information of the defect area of the detected picture of the hardware part; the position information and the shape of the defect area are utilized to intercept and select the defect of the detected picture of the hardware part on the picture, the defect type and the position of the detected picture of the hardware part are obtained, and the defect type and the corresponding position information of the detected picture of the hardware part are obtained by displaying through an intelligent terminal.
The beneficial effects of the above technical scheme are: the embodiment of the invention provides a hardware part defect detection method, which can detect a hardware part according to a defect detection model, further, the defect regions of the hardware parts are positioned and classified, so that the matching between the production suspension time setting and the integral execution state of the actual production project can be effectively improved, the influence on the integral execution time of the production project can be reduced to the maximum extent under the condition of ensuring that the legal responsibility can be completed and supplemented with enough production suspension time, under the condition of ensuring that sufficient production time is provided to perfect production safety and supervision management, the problem that the whole period of finished production projects is delayed due to unreasonable production suspension time setting is effectively prevented, the working efficiency and the working speed of hardware part production can be effectively improved, and the yield of production and the accuracy of products can be improved.
In an embodiment of the present invention, acquiring a standard picture of a hardware component and a defect sample picture corresponding to the defective hardware component, and acquiring images of the standard picture of the hardware component and the defect sample picture corresponding to the defective hardware component by using pixel processing and graying processing includes:
acquiring a standard picture of a good hardware part and a defect sample picture of a corresponding defective hardware part by an industrial camera, labeling the defect sample picture and defining a defect type; carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and the corresponding defect sample picture of the defective hardware part; acquiring a low-pixel image of the hardware part standard picture and a defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
graying the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
improving pixels of the hardware part standard picture and the gray level image of the defect sample picture corresponding to the defective hardware part, and acquiring a high pixel image of the hardware part standard picture and the high pixel image of the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the high pixel image of the defect sample picture of the corresponding defective hardware part reach 800 multiplied by 600;
wherein, the graying image that obtains hardware component standard picture and the defect sample picture that corresponds defective hardware component includes:
acquiring the input hardware part standard picture and a low-pixel image of a defect sample picture corresponding to the defective hardware part; calculating a graying weighting coefficient of the image based on a seagull algorithm; carrying out graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part by using an image graying weighting coefficient to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
calculating a gray-scale weighting coefficient of the image based on a gull algorithm, and initializing the gull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
The working principle of the technical scheme is as follows: acquiring a standard picture of a good hardware part and a defect sample picture of a corresponding defective hardware part by an industrial camera, labeling the defect sample picture and defining a defect type; carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and the corresponding defect sample picture of the defective hardware part; acquiring a low-pixel image of the hardware part standard picture and a defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
graying the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
improving pixels of the hardware part standard picture and the gray level image of the defect sample picture corresponding to the defective hardware part, and acquiring a high pixel image of the hardware part standard picture and the high pixel image of the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware standard picture and the high pixel image of the defect sample picture corresponding to the defective hardware reach 800 multiplied by 600;
wherein, the graying image that obtains hardware component standard picture and the defect sample picture that corresponds defective hardware component includes:
acquiring the input hardware part standard picture and a low-pixel image of a defect sample picture corresponding to the defective hardware part; calculating a graying weighting coefficient of the image based on a gull algorithm; carrying out graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part by using an image graying weighting coefficient to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
calculating a gray-scale weighting coefficient of the image based on a gull algorithm, and initializing the gull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
The beneficial effects of the above technical scheme are: through aiming at hardware component standard picture and the defect sample picture pixel processing and the grey scale processing that correspond defective hardware component, and then gather the image information who states hardware component standard picture and correspond defective hardware component's defect sample picture, reduced hardware component standard picture and the file size that corresponds defective hardware component's defect sample picture have saved the memory space, have reduced the calculated amount to graying processing simultaneously, have alleviateed computing device's loss, the calculation error when having avoided the calculation simultaneously, and then have reduced the calculating time, have practiced thrift the work load, have reduced hardware component defect detection's cost.
According to one embodiment of the invention, a three-dimensional hardware model is constructed, and characteristic points of the three-dimensional hardware model are extracted to obtain a defect detection model of the hardware; the method for acquiring the image information of the detected picture of the hardware part and importing the image information into the defect detection model comprises the following steps:
establishing a hardware part three-dimensional model by utilizing a characteristic modeling mode through the acquired hardware part standard picture and high-pixel image information of a defect sample picture corresponding to the defective hardware part; acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures, and determining defect information corresponding to each defect picture so as to generate a training data set and a verification data set; initializing the defect detection model by using the training data set, fixing the weight parameters of the generated comparison network, respectively carrying out discrimination processing and generation processing, and then enabling the loss function value of the generated comparison network to be within a preset threshold range to obtain the trained defect detection model;
inputting the three-dimensional model of the hardware part into the defect detection model, carrying out defect detection identification on the three-dimensional model by using the defect detection model to obtain a defect detection result corresponding to the three-dimensional model, and comparing the defect detection result with corresponding defect data in the verification data set to obtain an error value between the defect detection result and the corresponding defect data in the verification data set;
when the obtained error value is within a preset error range, the defect detection precision is proved to meet the standard, and then a defect detection model is output;
and initializing the defect detection model, acquiring the image information of the detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model.
The working principle of the technical scheme is as follows: constructing a three-dimensional hardware part model, and extracting characteristic points of the three-dimensional hardware part model to obtain a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model, wherein the defect detection model comprises the following steps:
establishing a hardware part three-dimensional model by using a characteristic modeling mode through the acquired hardware part standard picture and high-pixel image information of a defect sample picture corresponding to the defective hardware part; acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures, and determining defect information corresponding to each defect picture so as to generate a training data set and a verification data set; initializing the defect detection model by using the training data set, fixing the weight parameters of the generated comparison network, respectively carrying out discrimination processing and generation processing, and then enabling the loss function value of the generated comparison network to be within a preset threshold range to obtain the trained defect detection model;
inputting the three-dimensional model of the hardware part into the defect detection model, carrying out defect detection identification on the three-dimensional model by using the defect detection model to obtain a defect detection result corresponding to the three-dimensional model, and comparing the defect detection result with corresponding defect data in the verification data set to obtain an error value between the defect detection result and the corresponding defect data in the verification data set;
when the obtained error value is within a preset error range, the defect detection precision is proved to meet the standard, and then a defect detection model is output;
and initializing the defect detection model, acquiring the image information of the detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model.
The beneficial effects of the above technical scheme are: the three-dimensional model of the hardware part is obtained through feature modeling, the requirement of a machining process is met, information such as material, machining precision, proof quality, behavior error and the like of the part is provided, useful data are provided for obtaining the defect detection model, and the detection efficiency and speed of the hardware part are improved; the obtained three-dimensional model of the hardware part has the advantages of being free from geometric invariance such as translation, rotation and scaling of the model and invariance of multiple formats of the model, having good stability, having high precision on detected data, presenting information more directly and accurately, and saving time cost of people.
On the other hand, through the construction three-dimensional model of hardware component makes the more directly perceived that hardware component's spatial information presented, and is more concrete, has more the third dimension, and then acquires the defect detection model, and according to the defect detection model is right the image information that hardware component detected the picture detects, makes the error value in the defect detection model more accurate, obtains the data of high accuracy, and then makes the testing result more accurate, has improved the accuracy degree that hardware component defect detected, simultaneously the three-dimensional model of hardware component makes information directly perceived and visual, thereby makes quick and accurate judgement is made in the detection of defect detection model, has reduced the time of artifical judgement testing result, has improved work efficiency.
In an embodiment of the present invention, detecting and identifying the image information of the detected image of the hardware component by using the defect detection model, and acquiring the defect type and the defect area corresponding to the detected image of the hardware component includes:
acquiring image information of a defect sample picture of the hardware part, and detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model to identify a defect type corresponding to the defect sample picture of the hardware part;
classifying the defects according to the defect types to obtain defect classification results corresponding to the defect sample pictures of the hardware parts;
performing region extraction on a defect sample picture of the hardware part by using the defect classification result to obtain a defect region in the defect sample picture;
and comparing and identifying the defect area by using the three-dimensional model of the hardware part to obtain defect range information and further obtain the defect area.
The working principle of the technical scheme is as follows: utilize the defect detection model is right the image information that hardware component detected the picture detects the discernment, acquires the hardware component detects the defect classification and the defect region that the picture corresponds and include:
acquiring image information of a defect sample picture of the hardware part, and detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model to identify a defect type corresponding to the defect sample picture of the hardware part;
classifying the defects according to the defect types to obtain defect classification results corresponding to the defect sample pictures of the hardware parts;
performing region extraction on a defect sample picture of the hardware part by using the defect classification result to obtain a defect region in the defect sample picture;
and comparing and identifying the defect area by using the three-dimensional model of the hardware part to obtain defect range information and further obtain the defect area.
The beneficial effects of the above technical scheme are: through the defect detection model is right the image information that hardware component detected the picture detects the discernment, acquires the hardware component is detected defect classification and defect area that the picture corresponds, can be accurate through the above-mentioned mode confirm that hardware component is detected the defect area of picture, more quick finding the defect position that hardware component detected the defect area of picture has improved work efficiency, and is right simultaneously hardware component is detected the picture and is carried out defect classification, has saved the time to and the calculation step, and then reduced the error rate that hardware component defect detected has improved the productivity ratio of product simultaneously, has increased the income.
In an embodiment of the present invention, identifying an area edge line of the defect area, obtaining an area shape of the area by using the area edge line, positioning the defect area of the detected image of the hardware component by using the area shape, obtaining a defect type and a defect position of the detected image of the hardware component, and displaying through an intelligent terminal includes:
positioning the input detected picture of the hardware part to obtain the information of the detected picture of the hardware part;
identifying a defect area of the detected picture information of the hardware part through a defect identification model, and sorting by adopting an area edge line to obtain the area shape of the area;
positioning the defect area of the detected picture of the hardware part by using the area shape to obtain the position information and the category of the defect area of the detected picture of the hardware part, and displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the detection of the edge comprises:
determining a filtering range, enhancing edge lines, detecting edge areas and positioning the positions of the edge lines;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the type and position of the detected picture template defect of the hardware part through sub-pixel precision positioning;
the method for storing and displaying the defect types and the positions of the hardware parts to be detected on the intelligent terminal comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of the picture to be detected; acquiring an image displayed by the picture to be detected, and determining a display mode of the picture to be detected in the display area of the picture to be detected according to the defect type and the position of the image to be displayed of the picture to be detected and the defect type and the position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
The working principle of the technical scheme is as follows: the discernment regional margin line of defect area, and utilize regional margin line acquires regional shape utilizes regional shape is right the hardware component detects the defect area of picture and fixes a position, obtains the hardware component detects the defect type and the position of picture to show through intelligent terminal and include:
positioning the input detected picture of the hardware part to obtain the information of the detected picture of the hardware part;
identifying a defect area of the detected picture information of the hardware part through a defect identification model, and sorting by adopting an area edge line to obtain the area shape of the area;
positioning the defect area of the detected picture of the hardware part by using the area shape to obtain the position information and the category of the defect area of the detected picture of the hardware part, and displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the detection of the edge comprises:
determining a filtering range, enhancing edge lines, detecting edge areas and positioning the positions of the edge lines;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the defect type and position of the detected picture template of the hardware part through sub-pixel precision positioning;
the method for storing and displaying the defect types and the positions of the hardware parts to be detected on the intelligent terminal comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of the picture to be detected; acquiring an image displayed by the picture to be detected, and determining a display mode of the picture to be detected in the display area of the picture to be detected according to the defect type and the position of the image to be displayed of the picture to be detected and the defect type and the position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
The beneficial effects of the above technical scheme are: the method for detecting the defects of the hardware parts, which is provided by the embodiment, effectively improves the detection efficiency of the proven defects of the hardware parts, simultaneously reduces the occurrence of false detection or missed detection of the hardware parts, realizes the automatic detection of the proven defects of the hardware parts, can ensure the production time, improves the production efficiency, simultaneously effectively reduces the detection time, saves the step processing flow, simultaneously improves the time for detecting the defects of the hardware parts, improves the production automation degree, furthest reduces the defect identification error and the error rate of the hardware parts, effectively prevents the occurrence of the problems of the defective hardware parts, effectively improves the matching property of the standard parts and the defective parts of the hardware parts, more effectively reduces the occurrence rate of defective products of the hardware parts, reduces the cost, simultaneously saves a large amount of time for manually detecting the defects of the parts, and saves the time, and manpower, material resources and financial resources are saved.
In one embodiment of the present invention, as shown in fig. 3, the hardware part defect detecting system includes:
the system comprises an image acquisition module, a modeling extraction module, a defect extraction module and a defect identification module;
the image acquisition module is used for acquiring a hardware part standard picture and a defect sample picture corresponding to a defective hardware part, and acquiring images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part by utilizing pixel processing and graying processing;
the modeling extraction module is used for constructing a three-dimensional hardware part model, extracting characteristic points of the three-dimensional hardware part model and obtaining a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model;
the defect extraction module is used for detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, and acquiring the defect type and the defect area corresponding to the detected picture of the hardware part;
and the defect identification module is used for identifying the regional edge line of the defect region and utilizing the regional edge line to acquire the regional shape of the region and utilizing the regional shape to locate the defect region of the detected picture of the hardware part to obtain the defect type and the position of the detected picture of the hardware part, and displaying the defect type and the position through the intelligent terminal.
The working principle of the technical scheme is as follows: hardware component defect detection system includes:
the system comprises an image acquisition module, a modeling extraction module, a defect extraction module and a defect identification module;
the image acquisition module is used for acquiring a standard picture of the hardware part and a defect sample picture of the hardware part corresponding to the defect, and acquiring an image of the standard picture of the hardware part and the defect sample picture of the hardware part corresponding to the defect by utilizing pixel processing and graying processing;
the modeling extraction module is used for constructing a three-dimensional hardware part model, extracting characteristic points of the three-dimensional hardware part model and obtaining a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model;
the defect extraction module is used for detecting and identifying the image information of the detected image of the hardware part by using the defect detection model and acquiring the defect type and the defect area corresponding to the detected image of the hardware part;
and the defect identification module is used for identifying the regional edge line of the defect region and utilizing the regional edge line to acquire the regional shape of the region and utilizing the regional shape to locate the defect region of the detected picture of the hardware part to obtain the defect type and the position of the detected picture of the hardware part, and displaying the defect type and the position through the intelligent terminal.
The beneficial effects of the above technical scheme are: the embodiment of the invention provides a hardware part defect detection system which can detect a hardware part according to a defect detection model, further, the defect regions of the hardware parts are positioned and classified, so that the matching between the production suspension time setting and the integral execution state of the actual production project can be effectively improved, the influence on the integral execution time of the production project can be reduced to the maximum extent under the condition of ensuring that the legal responsibility can be completed and supplemented with enough production suspension time, under the condition of ensuring that sufficient production time is provided to perfect production safety and supervision management, the problem that the whole period of finished production projects is delayed due to unreasonable production suspension time setting is effectively prevented, the working efficiency and the working speed of hardware part production can be effectively improved, and the yield of production and the accuracy of products can be improved.
In one embodiment of the present invention, the image acquisition module comprises:
the low-pixel module is used for acquiring a standard picture of a good hardware part and a defect sample picture of a corresponding defective hardware part through an industrial camera, labeling the defect sample picture and defining a defect type; carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and the corresponding defect sample picture of the defective hardware part; acquiring a low-pixel image of the hardware standard picture and a defect sample picture corresponding to the defective hardware; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
the graying module is used for performing graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
the high pixel module is used for improving pixels of the hardware part standard picture and the gray level image of the defect sample picture corresponding to the defective hardware part, and acquiring high pixel images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the high pixel image of the defect sample picture of the corresponding defective hardware part reach 800 multiplied by 600;
wherein the graying module comprises:
the gull algorithm module is used for calculating a graying weighting coefficient of the image and initializing the gull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
The working principle of the technical scheme is as follows:
the low pixel module operation process comprises the following steps:
firstly, acquiring a standard picture of a good hardware part and a defect sample picture of a corresponding defective hardware part by an industrial camera, labeling the defect sample picture and defining a defect type;
then, carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and the corresponding defect sample picture of the defective hardware part;
finally, obtaining the hardware part standard picture and a low-pixel image of a defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
the graying module operation process comprises the following steps:
firstly, acquiring a standard picture of the hardware part and a low-pixel image of a defect sample picture of the hardware part corresponding to the defect;
then, carrying out graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture of the corresponding defective hardware part;
finally, obtaining the standard hardware part picture and a gray level image of a defect sample picture corresponding to the defective hardware part;
the high pixel module operation process comprises the following steps:
firstly, acquiring a standard picture of the hardware part and a low-pixel image of a defect sample picture of the hardware part corresponding to the defect;
then, pixels of the standard hardware part picture and the gray level image of the defect sample picture corresponding to the defective hardware part are increased;
finally, acquiring the hardware part standard picture and a high-pixel image of a defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the high pixel image of the defect sample picture of the corresponding defective hardware part reach 800 multiplied by 600;
wherein the graying module comprises:
the gull algorithm module is used for calculating a graying weighting coefficient of the image and initializing the gull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
The beneficial effects of the above technical scheme are: through aiming at hardware component standard picture and the defect sample picture pixel processing and the grey scale processing that correspond defective hardware component, and then gather the image information who states hardware component standard picture and correspond defective hardware component's defect sample picture, reduced hardware component standard picture and the file size that corresponds defective hardware component's defect sample picture have saved the memory space, have reduced the calculated amount to graying processing simultaneously, have alleviateed computing device's loss, the calculation error when having avoided the calculation simultaneously, and then have reduced the calculating time, have practiced thrift the work load, have reduced hardware component defect detection's cost.
In one embodiment of the present invention, the modeling extraction module includes: constructing a three-dimensional hardware part model by using the standard hardware part picture and the high-pixel image information of the defect sample picture corresponding to the defective hardware part according to the acquired standard hardware part picture and the high-pixel image information of the defect sample picture corresponding to the defective hardware part, extracting feature points of the three-dimensional hardware part model, and performing model training by using the feature points of the three-dimensional hardware part model to obtain a defect detection model of the hardware part; the defect detection model is initialized to acquire the image information of the detected picture of the hardware part, and the image information of the detected picture of the hardware part is imported into the defect detection model, wherein the image information comprises the following steps:
the three-dimensional module is used for establishing a three-dimensional model of the hardware part by utilizing a characteristic modeling mode through the acquired high-pixel image information of the standard hardware part picture and the defect sample picture corresponding to the defective hardware part; acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures, and determining defect information corresponding to each defect picture so as to generate a training data set and a verification data set;
the detection model module is used for carrying out initialization processing on the defect detection model by using the training data set, fixing the weight parameters of the generated comparison network, respectively carrying out discrimination processing and generation processing, and then enabling the loss function value of the generated comparison network to be within a preset threshold range to obtain the trained defect detection model;
the data comparison module is used for inputting the three-dimensional model of the hardware part into the defect detection model, utilizing the defect detection model to carry out defect detection identification on the three-dimensional model to obtain a defect detection result corresponding to the three-dimensional model, and comparing the defect detection result with corresponding defect data in the verification data set to obtain an error value between the defect detection result and the corresponding defect data in the verification data set;
and if the obtained error value is within a preset error range, proving that the defect detection precision meets the standard, and then outputting a defect detection model.
Initializing the defect detection model, acquiring image information of a detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model;
wherein the feature modeling comprises:
a feature modeler, a product database and an application interface;
wherein the feature modeler is to generate a feature model of the part; a product database for storing geometric and non-geometric data; the application interface is used for data exchange with other systems.
The working principle of the technical scheme is as follows:
the three-dimensional module operation process comprises the following steps:
firstly, establishing a three-dimensional hardware part model by using a characteristic modeling mode through the acquired hardware part standard picture and high-pixel image information of a defect sample picture corresponding to the defective hardware part;
then, acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures;
finally, determining the defect information corresponding to each defect picture so as to generate a training data set and a verification data set;
the operation process of the detection model module comprises the following steps:
firstly, initializing the defect detection model by using the training data set;
then, fixing the weight parameters of the generated comparison network, and respectively carrying out discrimination processing and generation processing;
finally, the loss function value of the generated comparison network is within a preset threshold range, and a trained defect detection model is obtained;
the data comparison module operation process comprises the following steps:
firstly, inputting a three-dimensional model of the hardware part into the defect detection model, and carrying out defect detection and identification on the three-dimensional model by using the defect detection model to obtain a defect detection result corresponding to the three-dimensional model;
then, comparing the defect detection result with corresponding defect data in the verification dataset;
and finally, obtaining an error value between the defect detection result and corresponding defect data in the verification data set, and if the obtained error value is within a preset error range, proving that the defect detection precision meets the standard, and then outputting a defect detection model.
Initializing the defect detection model, acquiring image information of a detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model;
wherein the feature modeling comprises:
a feature modeler, a product database and an application interface;
wherein the feature modeler is to generate a feature model of the part; a product database for storing geometric and non-geometric data; the application interface is used for data exchange with other systems.
The beneficial effects of the above technical scheme are: the three-dimensional model of the hardware part is obtained through feature modeling, the requirement of a machining process is met, information such as material, machining precision, proof quality, behavior error and the like of the part is provided, useful data are provided for obtaining the defect detection model, and the detection efficiency and speed of the hardware part are improved; the obtained three-dimensional model of the hardware part has the advantages of being free of geometric invariance such as translation, rotation and scaling of the model, free of invariance of multiple formats of the model, good in stability, high in precision of detected data, direct and accurate in presented information and capable of saving time cost of people.
On the other hand, through the construction three-dimensional model of hardware component makes the more directly perceived that hardware component's spatial information presented, and is more concrete, has more the third dimension, and then acquires the defect detection model, and according to the defect detection model is right the image information that hardware component detected the picture detects, makes the error value in the defect detection model more accurate, obtains the data of high accuracy, and then makes the testing result more accurate, has improved the accuracy degree that hardware component defect detected, simultaneously the three-dimensional model of hardware component makes information directly perceived and visual, thereby makes quick and accurate judgement is made in the detection of defect detection model, has reduced the time of artifical judgement testing result, has improved work efficiency.
In one embodiment of the present invention, the defect extraction module includes: detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, acquiring the defect type corresponding to the detected picture of the hardware part, classifying the defects according to the type, and acquiring the defect classification result corresponding to the detected picture of the hardware part; the method comprises the following steps of utilizing the defect classification result to carry out region extraction on the detected picture of the hardware part, and obtaining the defect region in the defect sample picture comprises the following steps:
the defect type module is used for acquiring image information of a defect sample picture of the hardware part, detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model, and identifying a defect type corresponding to the defect sample picture of the hardware part;
the defect region module is used for classifying defects according to the defect types to obtain defect classification results corresponding to the defect sample pictures of the hardware parts;
performing region extraction on a defect sample picture of the hardware part by using the defect classification result to obtain a defect region in the defect sample picture;
and comparing and identifying the defect area by using the three-dimensional model of the hardware part to obtain defect range information and further obtain the defect area.
The working principle of the technical scheme is as follows:
the defect type module operation process comprises the following steps:
firstly, acquiring image information of a defect sample picture of the hardware part;
then, detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model;
finally, identifying the defect type corresponding to the defect sample picture of the hardware part;
the operation process of the defect area module comprises the following steps:
firstly, classifying the defects according to the defect types to obtain a defect classification result corresponding to a defect sample picture of the hardware part;
then, performing region extraction on a defect sample picture of the hardware part by using the defect classification result;
finally, obtaining a defect area in the defect sample picture;
and comparing and identifying the defect area by using the three-dimensional model of the hardware part to obtain defect range information and further obtain the defect area.
The beneficial effects of the above technical scheme are: through the defect detection model is right the image information that hardware component detected the picture detects the discernment, acquires the hardware component is detected defect classification and defect area that the picture corresponds, can be accurate through the above-mentioned mode confirm that hardware component is detected the defect area of picture, more quick finding the defect position that hardware component detected the defect area of picture has improved work efficiency, and is right simultaneously hardware component is detected the picture and is carried out defect classification, has saved the time to and the calculation step, and then reduced the error rate that hardware component defect detected has improved the productivity ratio of product simultaneously, has increased the income.
In one embodiment of the present invention, the defect identifying module includes: identifying the area edge line of the defect area, acquiring the area shape of the area by using the area edge line, positioning the defect area of the detected picture of the hardware part by using the area shape, acquiring the defect type and position of the detected picture of the hardware part, and displaying the defect type and position through an intelligent terminal;
the positioning module is used for positioning the input detected picture of the hardware part to obtain the information of the detected picture of the hardware part;
the shape module is used for identifying the defect area of the detected picture information of the hardware part through a defect identification model and sorting by adopting an area edge line to obtain the area shape of the area;
the display module is used for positioning the defect area of the detected picture of the hardware part by the area shape, acquiring the position information and the type of the defect area of the detected picture of the hardware part, and displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the sorting of the region edge lines comprises:
determining a filtering range, enhancing edge lines, detecting edge areas and positioning the positions of the edge lines;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the type and position of the detected picture template defect of the hardware part through sub-pixel precision positioning;
the method comprises the following steps of displaying the defect type and the defect position of the hardware part to be detected on the intelligent terminal, wherein the method comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of the picture to be detected; acquiring an image displayed by the picture to be detected, and determining the display mode of the picture to be detected in the display area of the picture to be detected according to the defect type and the position of the image to be displayed of the picture to be detected and the defect type and the position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
The working principle of the technical scheme is as follows:
the operation process of the positioning module comprises the following steps:
firstly, acquiring a detected picture of the hardware part;
then, positioning the input detected picture of the hardware part;
finally, the detected picture information of the hardware part is obtained;
the shape die operation process comprises the following steps:
firstly, identifying a defect area of the detected picture information of the hardware part through a defect identification model;
then, sorting by adopting an area edge line;
after the combination, obtaining the area shape of the area;
the operation process of the display module comprises the following steps:
firstly, positioning a defect area of the detected picture of the hardware part according to the area shape;
then, obtaining the position information and the category of the defect area of the detected picture of the hardware part;
finally, displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the sorting of the region edge lines comprises:
determining a filtering range, enhancing edge lines, detecting edge areas and positioning the positions of the edge lines;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the type and position of the detected picture template defect of the hardware part through sub-pixel precision positioning;
the method comprises the following steps of displaying the defect type and the defect position of the hardware part to be detected on the intelligent terminal, wherein the method comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of the picture to be detected; acquiring an image displayed by the picture to be detected, and determining a display mode of the picture to be detected in the display area of the picture to be detected according to the defect type and the position of the image to be displayed of the picture to be detected and the defect type and the position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
The beneficial effects of the above technical scheme are: the method for detecting the defects of the hardware parts, which is provided by the embodiment, effectively improves the detection efficiency of the proven defects of the hardware parts, simultaneously reduces the occurrence of false detection or missed detection of the hardware parts, realizes the automatic detection of the proven defects of the hardware parts, can ensure the production time, improves the production efficiency, simultaneously effectively reduces the detection time, saves the step processing flow, simultaneously improves the time for detecting the defects of the hardware parts, improves the production automation degree, furthest reduces the defect identification error and the error rate of the hardware parts, effectively prevents the occurrence of the problems of the defective hardware parts, effectively improves the matching property of the standard parts and the defective parts of the hardware parts, more effectively reduces the occurrence rate of defective products of the hardware parts, reduces the cost, simultaneously saves a large amount of time for manually detecting the defects of the parts, and saves the time, and manpower, material resources and financial resources are saved.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. The method for detecting the defects of the hardware parts is characterized by comprising the following steps:
acquiring a hardware part standard picture and a defect sample picture corresponding to a defective hardware part, and acquiring images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part by utilizing pixel change processing and graying processing;
constructing a three-dimensional hardware part model, and extracting characteristic points of the three-dimensional hardware part model to obtain a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model;
detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, and acquiring the defect type and the defect area corresponding to the detected picture of the hardware part;
the method comprises the steps of identifying the regional edge line of a defect region, utilizing the regional edge line to obtain the regional shape, utilizing the regional shape to locate the defect region of the detected picture of the hardware part, obtaining the defect type and the position of the detected picture of the hardware part, and displaying the defect type and the position through an intelligent terminal.
2. The hardware part defect detection method according to claim 1, wherein the acquiring of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part, and the acquiring of the image of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part by using pixel change processing and graying processing comprises:
acquiring a standard picture of a good hardware part and a defect sample picture of a corresponding defective hardware part by an industrial camera, labeling the defect sample picture and defining a defect type; carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and the corresponding defect sample picture of the defective hardware part; acquiring a low-pixel image of the hardware part standard picture and a defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
graying the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
improving pixels of the hardware part standard picture and the gray level image of the defect sample picture corresponding to the defective hardware part, and acquiring a high pixel image of the hardware part standard picture and the high pixel image of the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the high pixel image of the defect sample picture of the corresponding defective hardware part reach 800 multiplied by 600;
wherein, the graying image that obtains hardware component standard picture and the defect sample picture that corresponds defective hardware component includes:
acquiring the input hardware part standard picture and a low-pixel image of a defect sample picture corresponding to the defective hardware part; calculating a graying weighting coefficient of the image based on a gull algorithm; carrying out graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture of the corresponding defective hardware part by using an image graying weighting coefficient;
the gray-scale weighting coefficient of the image calculated based on the gull algorithm comprises the following steps: initializing the gull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
3. The hardware part defect detection method according to claim 1, wherein a hardware part three-dimensional model is constructed, and feature points of the hardware part three-dimensional model are extracted to obtain a hardware part defect detection model; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model, wherein the defect detection model comprises the following steps:
establishing a hardware part three-dimensional model by utilizing a characteristic modeling mode through the acquired hardware part standard picture and high-pixel image information of a defect sample picture corresponding to the defective hardware part; acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures, and determining defect information corresponding to each defect picture so as to generate a training data set and a verification data set; initializing the defect detection model by using the training data set, fixing the weight parameters of the generated comparison network, respectively carrying out discrimination processing and generation processing, and then enabling the loss function value of the generated comparison network to be within a preset threshold range to obtain the trained defect detection model;
inputting the three-dimensional model of the hardware part into the defect detection model, carrying out defect detection identification on the three-dimensional model by using the defect detection model to obtain a defect detection result corresponding to the three-dimensional model, and comparing the defect detection result with corresponding defect data in the verification data set to obtain an error value between the defect detection result and the corresponding defect data in the verification data set;
when the obtained error value is within a preset error range, the defect detection precision is proved to meet the standard, and then a defect detection model is output;
and initializing the defect detection model, acquiring the image information of the detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model.
4. The method for detecting the hardware part defect according to claim 1, wherein the step of detecting and identifying the image information of the detected hardware part picture by using the defect detection model and the step of acquiring the defect type and the defect area corresponding to the detected hardware part picture comprises the following steps:
acquiring image information of a defect sample picture of the hardware part, and detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model to identify a defect type corresponding to the defect sample picture of the hardware part;
classifying the defects according to the defect types to obtain defect classification results corresponding to the defect sample pictures of the hardware parts;
and performing region extraction on the defect sample picture of the hardware part by using the defect classification result to obtain a defect region in the defect sample picture.
5. The hardware part defect detection method according to claim 1, wherein the step of identifying the area edge line of the defect area, acquiring the area shape of the area by using the area edge line, positioning the defect area of the detected image of the hardware part by using the area shape, acquiring the defect type and position of the detected image of the hardware part, and displaying the defect type and position through an intelligent terminal comprises the steps of:
positioning the input detected picture of the hardware part to obtain the information of the detected picture of the hardware part;
identifying a defect area of the detected picture information of the hardware part through a defect identification model, and sorting by adopting an area edge line to obtain the area shape of the area;
positioning the defect area of the detected picture of the hardware part by using the area shape to obtain the position information and the category of the defect area of the detected picture of the hardware part, and displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the type and position of the detected picture template defect of the hardware part through sub-pixel precision positioning;
the method for storing and displaying the defect types and the positions of the hardware parts to be detected on the intelligent terminal comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of a picture to be detected; acquiring an image displayed by a picture to be detected, and determining the display mode of the picture to be detected in the display area of the picture to be detected according to the defect type and position of the image to be displayed of the picture to be detected and the defect type and position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
6. The utility model provides a hardware defect detecting system which characterized in that, hardware defect detecting system includes:
the system comprises an image acquisition module, a modeling extraction module, a defect extraction module and a defect identification module;
the image acquisition module is used for acquiring a hardware part standard picture and a defect sample picture corresponding to a defective hardware part, and acquiring images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part by utilizing pixel change processing and graying processing;
the modeling extraction module is used for constructing a three-dimensional hardware part model, extracting characteristic points of the three-dimensional hardware part model and obtaining a defect detection model of the hardware part; acquiring image information of a detected picture of the hardware part and importing the image information into the defect detection model;
the defect extraction module is used for detecting and identifying the image information of the detected picture of the hardware part by using the defect detection model, and acquiring the defect type and the defect area corresponding to the detected picture of the hardware part;
the defect identification module identifies the regional margin line of the defect area, utilizes the regional margin line obtains the regional shape of region, utilizes regional shape is right the defect area that hardware component detected the picture is fixed a position, obtains the hardware component is detected the defect type and the position of picture to show through intelligent terminal.
7. The hardware component defect detection system of claim 6, wherein said image acquisition module comprises:
the low-pixel module is used for collecting a complete standard picture of the hardware part and a corresponding defect sample picture of the defective hardware part through an industrial camera, labeling the defect sample picture and defining a defect type; carrying out image denoising processing and pixel reduction processing on the acquired standard picture of the hardware part and the corresponding defect sample picture of the defective hardware part; acquiring a low-pixel image of the hardware part standard picture and a defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part reach 16 multiplied by 16;
the graying module is used for performing graying processing on the hardware part standard picture and the low-pixel image of the defect sample picture corresponding to the defective hardware part to obtain grayed images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part;
the high pixel module is used for improving pixels of the hardware part standard picture and the gray level image of the defect sample picture corresponding to the defective hardware part, and acquiring high pixel images of the hardware part standard picture and the defect sample picture corresponding to the defective hardware part; wherein, the pixels of the hardware standard picture and the high pixel image of the defect sample picture corresponding to the defective hardware reach 800 x 600;
wherein the graying module comprises:
the gull algorithm module is used for calculating a graying weighting coefficient of the image and initializing the gull position within a set parameter range; iterating the gull target position; determining an image graying weighting coefficient according to the gull target position; wherein the setting parameters include: the round-trip boundary of the gull, the number of gull channels, the size of gull population and the maximum number of iterations.
8. The hardware part defect detection system of claim 6, wherein the modeling extraction module comprises:
the three-dimensional module is used for establishing a three-dimensional model of the hardware part by utilizing a characteristic modeling mode through the acquired high-pixel image information of the standard hardware part picture and the defect sample picture corresponding to the defective hardware part; acquiring a defect three-dimensional model library through a plurality of hardware part defect pictures, and determining defect information corresponding to each defect picture so as to generate a training data set and a verification data set;
the detection model module is used for carrying out initialization processing on the defect detection model by using the training data set, fixing the weight parameters of the generated comparison network, respectively carrying out discrimination processing and generation processing, and then enabling the loss function value of the generated comparison network to be within a preset threshold range to obtain the trained defect detection model;
the data comparison module is used for inputting the three-dimensional model of the hardware part into the defect detection model, utilizing the defect detection model to carry out defect detection identification on the three-dimensional model to obtain a defect detection result corresponding to the three-dimensional model, and comparing the defect detection result with corresponding defect data in the verification data set to obtain an error value between the defect detection result and the corresponding defect data in the verification data set;
when the obtained error value is within a preset error range, the defect detection precision is proved to meet the standard, and then a defect detection model is output;
and initializing the defect detection model, acquiring the image information of the detected picture of the hardware part, and importing the image information of the detected picture of the hardware part into the defect detection model.
9. The hardware component defect detection system of claim 6, wherein said defect extraction module comprises:
the defect type module is used for acquiring image information of a defect sample picture of the hardware part, detecting and identifying the image information of the defect sample picture of the hardware part by using the defect detection model, and identifying a defect type corresponding to the defect sample picture of the hardware part;
the defect region module is used for classifying defects according to the defect types to obtain defect classification results corresponding to the defect sample pictures of the hardware parts;
the area extraction module is used for carrying out area extraction on a defect sample picture of the hardware part by using the defect classification result to obtain a defect area in the defect sample picture;
and comparing and identifying the defect area by using the three-dimensional model of the hardware part to obtain defect range information and further obtain the defect area.
10. The hardware component defect detection system of claim 6, wherein said defect identification module comprises:
the positioning module is used for positioning the input detected picture of the hardware part to obtain the information of the detected picture of the hardware part;
the shape module is used for identifying the defect area of the detected picture information of the hardware part through a defect identification model and sorting by adopting an area edge line to obtain the area shape of the area;
the display module is used for positioning the defect area of the detected picture of the hardware part by the area shape, acquiring the position information and the type of the defect area of the detected picture of the hardware part, and displaying the defect type and the corresponding position of the detected picture of the hardware part on an intelligent terminal;
wherein the defect type and corresponding position location comprises the following steps:
s1, establishing a detected picture template of the hardware part, and performing a series of rotation, scaling and pyramid down-sampling on the detected picture template of the hardware part to generate a series of template images of the detected picture of the hardware part; carrying out edge extraction on a template image of the detected picture of the hardware part;
s2, generating a template image pyramid of the detected picture of the hardware part, calculating the edge direction gradient of the image from the top layer of the pyramid, and calculating the template response through an NCC algorithm;
s3, determining the defect type and position of the detected picture template of the hardware part through sub-pixel precision positioning;
the method comprises the following steps of displaying the defect type and the defect position of the hardware part to be detected on the intelligent terminal, wherein the method comprises the following steps:
after an application in the intelligent terminal is started, determining a display area of a picture to be detected; acquiring an image displayed by a picture to be detected, and determining a display mode of the picture to be detected in a display area of the picture to be detected according to the defect type and the position of the image to be displayed of the picture to be detected and the defect type and the position of the display area of the picture to be detected; and displaying and storing the picture to be detected in the display area of the picture to be detected in a determined display mode.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210637709.7A CN114708268A (en) | 2022-06-08 | 2022-06-08 | Hardware part defect detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210637709.7A CN114708268A (en) | 2022-06-08 | 2022-06-08 | Hardware part defect detection method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114708268A true CN114708268A (en) | 2022-07-05 |
Family
ID=82177877
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210637709.7A Pending CN114708268A (en) | 2022-06-08 | 2022-06-08 | Hardware part defect detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114708268A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6442301B1 (en) * | 1997-01-06 | 2002-08-27 | Applied Science Fiction, Inc. | Apparatus and method for defect channel nulling |
CN106568783A (en) * | 2016-11-08 | 2017-04-19 | 广东工业大学 | Hardware part defect detecting system and method |
CN113112497A (en) * | 2021-05-06 | 2021-07-13 | 合肥中科迪宏自动化有限公司 | Industrial appearance defect detection method based on zero sample learning, electronic device and storage medium |
CN113376183A (en) * | 2021-06-08 | 2021-09-10 | 深圳市英视自动化科技有限公司 | Glass defect detection method, system and equipment |
CN113643370A (en) * | 2021-08-24 | 2021-11-12 | 凌云光技术股份有限公司 | Image positioning method and device based on NCC algorithm |
CN114092474A (en) * | 2022-01-19 | 2022-02-25 | 深圳市杰美特科技股份有限公司 | Method and system for detecting processing defects of complex texture background of mobile phone shell |
CN114240806A (en) * | 2022-02-24 | 2022-03-25 | 北京盈通恒信电力科技有限公司 | Method and apparatus for graying image, computer device, and storage medium |
CN114460053A (en) * | 2022-01-21 | 2022-05-10 | 西安工业大学 | Optical element subsurface defect three-dimensional reconstruction method based on quantum dot fluorescence effect |
-
2022
- 2022-06-08 CN CN202210637709.7A patent/CN114708268A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6442301B1 (en) * | 1997-01-06 | 2002-08-27 | Applied Science Fiction, Inc. | Apparatus and method for defect channel nulling |
CN106568783A (en) * | 2016-11-08 | 2017-04-19 | 广东工业大学 | Hardware part defect detecting system and method |
CN113112497A (en) * | 2021-05-06 | 2021-07-13 | 合肥中科迪宏自动化有限公司 | Industrial appearance defect detection method based on zero sample learning, electronic device and storage medium |
CN113376183A (en) * | 2021-06-08 | 2021-09-10 | 深圳市英视自动化科技有限公司 | Glass defect detection method, system and equipment |
CN113643370A (en) * | 2021-08-24 | 2021-11-12 | 凌云光技术股份有限公司 | Image positioning method and device based on NCC algorithm |
CN114092474A (en) * | 2022-01-19 | 2022-02-25 | 深圳市杰美特科技股份有限公司 | Method and system for detecting processing defects of complex texture background of mobile phone shell |
CN114460053A (en) * | 2022-01-21 | 2022-05-10 | 西安工业大学 | Optical element subsurface defect three-dimensional reconstruction method based on quantum dot fluorescence effect |
CN114240806A (en) * | 2022-02-24 | 2022-03-25 | 北京盈通恒信电力科技有限公司 | Method and apparatus for graying image, computer device, and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111179251B (en) | Defect detection system and method based on twin neural network and by utilizing template comparison | |
CN107543828B (en) | Workpiece surface defect detection method and system | |
CN111080622B (en) | Neural network training method, workpiece surface defect classification and detection method and device | |
KR20190063839A (en) | Method and System for Machine Vision based Quality Inspection using Deep Learning in Manufacturing Process | |
CN116188475B (en) | Intelligent control method, system and medium for automatic optical detection of appearance defects | |
CN111242899B (en) | Image-based flaw detection method and computer-readable storage medium | |
CN110766095A (en) | Defect detection method based on image gray level features | |
WO2024002187A1 (en) | Defect detection method, defect detection device, and storage medium | |
CN115439458A (en) | Industrial image defect target detection algorithm based on depth map attention | |
CN117115147B (en) | Textile detection method and system based on machine vision | |
CN115147363A (en) | Image defect detection and classification method and system based on deep learning algorithm | |
CN114881998A (en) | Workpiece surface defect detection method and system based on deep learning | |
CN111754502A (en) | Method for detecting surface defects of magnetic core based on fast-RCNN algorithm of multi-scale feature fusion | |
CN113205511B (en) | Electronic component batch information detection method and system based on deep neural network | |
CN115830403B (en) | Automatic defect classification system and method based on deep learning | |
CN113822836A (en) | Method of marking an image | |
CN114708268A (en) | Hardware part defect detection method and system | |
CN113020428B (en) | Progressive die machining monitoring method, device, equipment and storage medium | |
CN114882009A (en) | Fatigue crack tip automatic detection method and system capable of adapting to various surface states | |
CN109559294B (en) | Method and device for detecting quality of circular hole of drop | |
Candra et al. | Computer Vision Implementation in Scratch Inspection and Color Detection on The Car Roof Surface | |
CN117911796B (en) | Intelligent data processing system and method based on image recognition | |
KR102686135B1 (en) | Method of inspecting defect using artificial intelligence and apparatus for inspecting defect using artificial intelligence | |
CN117495846B (en) | Image detection method, device, electronic equipment and storage medium | |
CN116958086B (en) | Metal surface defect detection method and system with enhanced feature fusion capability |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220705 |
|
RJ01 | Rejection of invention patent application after publication |