CN110977767B - Casting defect distribution detection method and casting polishing method - Google Patents
Casting defect distribution detection method and casting polishing method Download PDFInfo
- Publication number
- CN110977767B CN110977767B CN201911102331.5A CN201911102331A CN110977767B CN 110977767 B CN110977767 B CN 110977767B CN 201911102331 A CN201911102331 A CN 201911102331A CN 110977767 B CN110977767 B CN 110977767B
- Authority
- CN
- China
- Prior art keywords
- casting
- defect
- polishing
- flaw
- algorithm
- 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.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B24—GRINDING; POLISHING
- B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
- B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
- B24B49/12—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation involving optical means
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Constituent Portions Of Griding Lathes, Driving, Sensing And Control (AREA)
Abstract
The invention relates to a casting defect distribution detection method and a casting polishing method, wherein the casting defect distribution detection method comprises the following steps: (1) sampling a casting image; (2) processing an image; (3) detecting defects; (4) and (4) obtaining the flaw distribution condition of the whole casting surface according to the flaw monitoring and storing results of the subareas of the casting surface in the step (3). The invention can realize the acquisition of weak flaw polishing points, store flaw defects and count the flaw distribution of the casting, so that a polishing path can be designed according to the distribution statistical result, the reliability of replanning the polishing path can be improved by surreptitious planning, and suspected flaw points can not be missed while skipping over part of polishing areas. The casting polishing method comprises the steps of determining polishing points and optimizing polishing paths according to stored defect information on the basis of casting defect distribution detection, and then polishing the casting through the designed polishing paths.
Description
Technical Field
The invention belongs to the technical field of machining, and particularly relates to a casting defect distribution detection method and a casting polishing method.
Background
The polishing and cleaning of the casting is very heavy work, the working conditions are poor (high temperature, dust and dirty environment), and the labor intensity is high. At present, the clearance of foundry goods is polished and is still used a large amount of manual works in the casting trade, and the burden of casting enterprise has been aggravated to the continuous rising of cost of labor, and the environment of polishing that the dust flies upward simultaneously causes very big harm to workman's health, and occupational disease prevention, industrial injury compensation have further improved the cost of labor, and the operational environment is poor, intensity of labour also makes the enterprise sink into the dilemma that recruits workers difficultly greatly. In addition, the large castings are manually polished, a large number of fields are needed, the efficiency is low, the polishing quality cannot be guaranteed, and the production cost of enterprises is increased invisibly. Moreover, with the increase of the living standard of people and the enhancement of health consciousness, many foundries have been difficult to bring to the grinders. Therefore, in the casting industry, particularly in the casting polishing and cleaning process, the adoption of robots to replace manual work is a great trend.
In the existing casting polishing operation, classification of polishing points is not emphasized, and generally, after a workpiece is positioned, a polishing tool bit with a certain size is used for flat-pushing polishing according to a preset polishing path, so that the polishing tool bit with lower strength meets a special larger polishing area, and the polishing tool bit is easily damaged; on the other hand, the condition that the selection of the grinding tool bit and the grinding point are not matched can cause a larger characteristic electric signal value of the grinding monitoring system, so that unnecessary triggering shutdown of the grinding protection system is caused.
In addition, the flaw distribution of the surface of the casting to be cleaned is irregular, the distribution of the polishing points cannot be effectively identified by the conventional polishing method, blind polishing exists, and efficient polishing cannot be achieved.
In view of the foregoing, it is desirable to provide a detection method capable of accurately identifying defect distribution of a casting and a casting polishing method based on the detection method, which can improve polishing efficiency.
Disclosure of Invention
The invention aims to provide an automatic loading and positioning method which is high in economy, low in cost, large in measurement range and capable of accurately and quickly determining the position of a vehicle body.
The above purpose is realized by the following technical scheme: a casting defect distribution detection method comprises the following steps:
(1) sampling a casting image: the sensor senses that the casting enters a visual identification area and transmits a trigger signal to the visual identification unit, and the visual identification unit acquires images of the preset subarea of the casting and transmits the acquired real-time images to the control unit;
(2) image processing: the control unit processes the received image information;
(3) flaw detection: the control unit compares the processed collected patterns with a detection standard to detect whether defect exists or not, if so, defect information is stored, and if not, the casting is moved to the next detection area to repeat the step (1) and the step (2) until the sub-area detection of the whole casting surface is finished;
(4) and (4) obtaining the flaw distribution condition of the whole casting surface according to the flaw monitoring and storing results of the subareas of the casting surface in the step (3).
The invention can realize the acquisition of weak flaw polishing points, store flaw defects and count the flaw distribution of the casting, so that a polishing path can be designed according to the distribution statistical result, the reliability of replanning the polishing path can be improved by surreptitious planning, and suspected flaw points can not be missed while skipping over part of polishing areas.
The further technical scheme is that, in the step (3), the extraction of the ROI of the acquired image is firstly completed, wherein the extraction includes manual delineation of the ROI based on a priori knowledge and automatic setting of the ROI based on a matching algorithm.
The further technical scheme is that the step (3) comprises flaw edge characteristic detection and threshold-based flaw segmentation detection, and a local rapid Otsu segmentation tiny flaw detection algorithm and/or a wavelet transformation-based weak flaw detection algorithm are/is designed for weak flaw points which are difficult to distinguish, so that the grinding flaw defects of regions which are difficult to segment are detected.
The further technical scheme is that the control unit in the step (2) adopts an image preprocessing algorithm of a flaw point to be polished to process image information.
The further technical scheme is that in the step (2), analysis is performed according to the reason and the type of noise generation in the process of acquiring the image of the defect area of the casting to be polished, and an image preprocessing algorithm of the point to be polished of the defect is designed on the basis.
According to the actual casting grinding production process and environment, the basic principle of a machine vision detection technology is combined, and on the basis of referring to a large number of latest vision detection technical documents, an overall scheme of flaw detection and positioning in advance in a casting grinding system is designed, wherein the overall scheme comprises the composition, model selection and light source illumination scheme of a hardware system, a detection algorithm flow of a software system and the like.
The further technical scheme is that the image preprocessing algorithm of the flaw point to be polished comprises the following steps: the weighted mean filtering is improved, a self-adaptive weighted mask mean filtering algorithm is designed, then the image is sharpened and enhanced, and then the distribution of the characteristic region of the point to be polished is enhanced from two angles of gray value linear mapping and gray value dynamic stretching by adopting gray level transformation and histogram equalization.
Therefore, a foundation is laid for next defect segmentation.
The further technical scheme is that in the step (3), a feature extraction algorithm of the defects of the to-be-polished points of the castings is designed by analyzing the defect images: the extracted feature point information is combined into a one-dimensional feature vector, a defect algorithm based on a feature threshold and a defect classification recognition algorithm based on a BP neural network are set, the two detection recognition modes can be matched in height, the defect algorithm based on the feature threshold is adopted for most of points to be detected, and the defect classification recognition algorithm based on the BP neural network is used for accurately recognizing small weak feature points which are difficult to recognize.
The defect algorithm based on the characteristic threshold is characterized by simplicity and convenience. In order to further improve the accuracy of classification, a defect classification and identification algorithm based on a BP neural network is used for accurately identifying the small weak feature points which are difficult to identify.
The further technical scheme is that the defect classification and identification algorithm based on the BP neural network comprises optimization of a minimum error valueWherein the algorithm modifies the weight coefficients according to the direction of the negative gradient of the error propagation: first, an error function is defined: ep = ∑ (t)i-yi)2/2, wherein tiRepresenting the current actual output, yiIs the result obtained by forward calculation, and the total error is: eA = ∑∑(ti-yi)2And/2, setting Wsp as a connection weight between any two neurons in the neural network, wherein eta represents a learning rate, and according to a gradient descent method, the correction quantity of the weight is as follows: Δ Wsp = - η · EA/Wsp, using gradient method to gradually reduce the error until delta EAAnd = 0, when the input and the output have a nonlinear relationship and the training sample is large enough, the classification of the nonlinear relationship, that is, the classification of the defective point and the normal point can be well realized.
The BP algorithm is more conventional, and the core of the method lies in solving the optimization problem of the minimum error value, wherein the detection of weak fault based on the BP neural network is the core content of the method.
In order to achieve the purpose, the invention also provides a casting polishing method, which comprises the following steps:
(1) detecting defect distribution of the casting: detecting the defect distribution of the casting according to any one of the defect distribution detection methods of the casting;
(2) determining a polishing point according to the stored defect information and optimizing a polishing path;
(3) and (4) polishing the casting according to the polishing path designed in the step (2).
The method for pre-identifying the grinding point can optimize the grinding path, so that the area without flaws is skipped, and the time required for grinding the single casting is effectively reduced by the method.
The technical scheme is that the sensor is a photoelectric sensor, the visual recognition unit is an industrial camera, and the control unit is an industrial personal computer.
The greatest difference between the method and the prior art lies in the advancement of grinding path planning, and the advancement is based on the adoption of a computer vision system to preliminarily position the defect area of the casting to be ground. Meanwhile, the invention can also carry out preliminary evaluation on the detected point to be polished, and carry out optimization on the subsequent polishing path to a certain extent on the basis, thereby realizing breakthrough of polishing efficiency by taking the innovation point as the basis.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention.
FIG. 1 is a schematic flow chart of a casting flaw distribution detection method and a grinding path design according to an embodiment of the invention;
fig. 2 is a diagram of a neural network structure according to an embodiment of the present invention.
Detailed Description
The present invention will now be described in detail with reference to the drawings, which are given by way of illustration and explanation only and should not be construed to limit the scope of the present invention in any way. Furthermore, features from embodiments in this document and from different embodiments may be combined accordingly by a person skilled in the art from the description in this document.
The embodiment of the invention provides a method for detecting defect distribution of a casting, which comprises the following steps of:
(1) sampling a casting image: the sensor senses that the casting enters a visual identification area and transmits a trigger signal to the visual identification unit, and the visual identification unit acquires images of the preset subarea of the casting and transmits the acquired real-time images to the control unit;
(2) image processing: the control unit processes the received image information;
(3) flaw detection: the control unit compares the processed collected patterns with a detection standard to detect whether defect exists or not, if so, defect information is stored, and if not, the casting is moved to the next detection area to repeat the step (1) and the step (2) until the sub-area detection of the whole casting surface is finished;
(4) and (4) obtaining the flaw distribution condition of the whole casting surface according to the flaw monitoring and storing results of the subareas of the casting surface in the step (3).
The invention can realize the acquisition of weak flaw polishing points, store flaw defects and count the flaw distribution of the casting, so that a polishing path can be designed according to the distribution statistical result, the reliability of replanning the polishing path can be improved by surreptitious planning, and suspected flaw points can not be missed while skipping over part of polishing areas.
On the basis of the above embodiment, in another embodiment of the present invention, in the step (3), the extraction of the ROI of the acquired image is first completed, which includes manually defining the ROI based on a priori knowledge and automatically setting the ROI based on a matching algorithm.
On the basis of the above embodiment, in another embodiment of the present invention, the step (3) includes defect edge feature detection and defect segmentation detection based on a threshold, and a local fast Otsu segmentation tiny defect detection algorithm and/or a wavelet transform-based weak defect detection algorithm is designed for weak defect points that are difficult to distinguish, so as to achieve detection of grinding defect defects in regions that are difficult to segment.
On the basis of the above embodiment, in another embodiment of the present invention, as shown in fig. 1, the control unit in step (2) performs image information line processing by using an image preprocessing algorithm for the defective to-be-polished point.
On the basis of the above embodiment, in another embodiment of the present invention, in the step (2), analysis is performed according to the cause and type of noise generation in the process of obtaining the image of the defect area of the casting to be polished, and an image preprocessing algorithm for the defect point to be polished is designed on the basis.
According to the actual casting grinding production process and environment, the basic principle of a machine vision detection technology is combined, and on the basis of referring to a large number of latest vision detection technical documents, an overall scheme of flaw detection and positioning in advance in a casting grinding system is designed, wherein the overall scheme comprises the composition, model selection and light source illumination scheme of a hardware system, a detection algorithm flow of a software system and the like.
On the basis of the above embodiment, in another embodiment of the present invention, the image preprocessing algorithm for the defective to-be-polished point includes: the weighted mean filtering is improved, a self-adaptive weighted mask mean filtering algorithm is designed, then the image is sharpened and enhanced, and then the distribution of the characteristic region of the point to be polished is enhanced from two angles of gray value linear mapping and gray value dynamic stretching by adopting gray level transformation and histogram equalization.
Therefore, a foundation is laid for next defect segmentation.
On the basis of the above embodiment, in another embodiment of the present invention, in the step (3), a feature extraction algorithm for defects of the to-be-polished point of the casting is designed by analyzing the defect image: the extracted feature point information is combined into a one-dimensional feature vector, a defect algorithm based on a feature threshold and a defect classification recognition algorithm based on a BP neural network are set, the two detection recognition modes can be matched in height, the defect algorithm based on the feature threshold is adopted for most of points to be detected, and the defect classification recognition algorithm based on the BP neural network is used for accurately recognizing small weak feature points which are difficult to recognize.
The defect algorithm based on the characteristic threshold is characterized by simplicity and convenience. In order to further improve the accuracy of classification, a defect classification and identification algorithm based on a BP neural network is used for accurately identifying the small weak feature points which are difficult to identify.
On the basis of the above embodiments, in another embodiment of the present invention, as shown in fig. 1 and fig. 2, the defect classification and identification algorithm based on the BP neural network includes optimization of the minimum error value, wherein the algorithm modifies the weight coefficients according to the negative gradient direction of error propagation: first, an error function is defined: ep = ∑ (t)i-yi)2/2, wherein tiRepresenting the current actual output, yiIs the result obtained by forward calculation, and the total error is: eA = ∑∑(ti-yi)2(v 2) let Wsp be the connection between any two neurons in the neural networkReceiving the weight value, wherein eta represents the learning rate, and according to a gradient descent method, the correction quantity of the weight value is as follows: Δ Wsp = - η · EA/Wsp, using gradient method to gradually reduce the error until delta EAAnd = 0, when the input and the output have a nonlinear relationship and the training sample is large enough, the classification of the nonlinear relationship, that is, the classification of the defective point and the normal point can be well realized.
The BP algorithm is more conventional, and the core of the method lies in solving the optimization problem of the minimum error value, wherein the detection of weak fault based on the BP neural network is the core content of the method.
The invention also provides a casting grinding method, and the embodiment is as follows, as shown in figure 1, and the method comprises the following steps:
(1) detecting defect distribution of the casting: detecting the defect distribution of the casting according to any one of the defect distribution detection methods of the casting;
(2) determining a polishing point according to the stored defect information and optimizing a polishing path;
(3) and (4) polishing the casting according to the polishing path designed in the step (2).
The method for pre-identifying the grinding point can optimize the grinding path, so that the area without flaws is skipped, and the time required for grinding the single casting is effectively reduced by the method.
On the basis of the above embodiment, in another embodiment of the present invention, as shown in fig. 1, the sensor is a photoelectric sensor, the visual recognition unit is an industrial camera, and the control unit is an industrial personal computer.
The greatest difference between the method and the prior art lies in the advancement of grinding path planning, and the advancement is based on the adoption of a computer vision system to preliminarily position the defect area of the casting to be ground. Meanwhile, the invention can also carry out preliminary evaluation on the detected point to be polished, and carry out optimization on the subsequent polishing path to a certain extent on the basis, thereby realizing breakthrough of polishing efficiency by taking the innovation point as the basis.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (5)
1. A casting defect distribution detection method is characterized by comprising the following steps:
(1) sampling a casting image: the sensor senses that the casting enters a visual identification area and transmits a trigger signal to the visual identification unit, and the visual identification unit acquires images of the preset subarea of the casting and transmits the acquired real-time images to the control unit;
(2) image processing: the control unit processes the received image information, wherein the control unit processes the image information by adopting an image preprocessing algorithm of a flaw to-be-polished point, analyzes the reason and the type of noise generation in the process of acquiring the image of the flaw area of the casting to be polished, and designs an image preprocessing algorithm of the flaw to-be-polished point on the basis, and the image preprocessing algorithm of the flaw to-be-polished point comprises the following steps: the weighted mean filtering is improved, a self-adaptive weighted mask mean filtering algorithm is designed, then the image is sharpened and enhanced, and then the distribution of the characteristic region of the point to be polished is enhanced from two angles of gray value linear mapping and gray value dynamic stretching by adopting gray level transformation and histogram equalization;
(3) flaw detection: the control unit compares the collected pattern after being processed with a detection standard, detects whether a defect exists, stores defect information if the defect exists, and moves to the next detection area to repeat the step (1) and the step (2) if the defect does not exist until the detection of the subareas on the surface of the whole casting is completed, wherein, by analyzing a defect image, a characteristic extraction algorithm of the casting for treating the grinding point defect is designed: the extracted feature point information is combined into a one-dimensional feature vector, a defect algorithm based on a feature threshold value and a defect classification and identification algorithm based on a BP neural network are set, and the two detection and identification modes can be matched in heightThe defect classification and identification method based on the BP neural network comprises the following steps of adopting a defect algorithm based on a characteristic threshold value for most points to be detected, accurately identifying the tiny weak characteristic points which are difficult to identify by using a defect classification and identification algorithm based on the BP neural network, optimizing a minimum error value, and modifying a weight coefficient according to a negative gradient direction of error propagation by the algorithm: first, an error function is defined: ep = ∑ (t)i-yi)2/2, wherein tiRepresenting the current actual output, yiIs the result obtained by forward calculation, and the total error is: eA = ∑∑(ti-yi)2And/2, setting Wsp as a connection weight between any two neurons in the neural network, wherein eta represents a learning rate, and according to a gradient descent method, the correction quantity of the weight is as follows: Δ Wsp = - η · EA/Wsp, using gradient method to gradually reduce the error until delta EA= 0, when the input and output have a nonlinear relationship and the training sample is large enough, classification of the nonlinear relationship can be well realized, that is, classification of the defective point and the normal point can be realized;
(4) and (4) obtaining the flaw distribution condition of the whole casting surface according to the flaw monitoring and storing results of the subareas of the casting surface in the step (3).
2. The casting defect distribution detection method according to claim 1, wherein the step (3) is to perform the extraction of ROI of the collected image firstly, which comprises manual ROI delineation based on a priori knowledge and automatic ROI setting based on a matching algorithm.
3. The casting defect distribution detection method according to claim 2, wherein the step (3) comprises defect edge feature detection and defect segmentation detection based on a threshold, and a local fast Otsu segmentation tiny defect detection algorithm and/or a wavelet transform weak defect detection algorithm is designed for weak defect points which are difficult to distinguish, so that grinding defect defects of regions which are difficult to segment are detected.
4. A casting grinding method is characterized by comprising the following steps:
(1) detecting defect distribution of the casting: detecting the defect distribution of the casting according to the defect distribution detection method of the casting according to any one of claims 1 to 3;
(2) determining a polishing point according to the stored defect information and optimizing a polishing path;
(3) and (4) polishing the casting according to the polishing path designed in the step (2).
5. The casting grinding method of claim 4, wherein the sensor is a photoelectric sensor, the visual recognition unit is an industrial camera, and the control unit is an industrial personal computer.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911102331.5A CN110977767B (en) | 2019-11-12 | 2019-11-12 | Casting defect distribution detection method and casting polishing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911102331.5A CN110977767B (en) | 2019-11-12 | 2019-11-12 | Casting defect distribution detection method and casting polishing method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110977767A CN110977767A (en) | 2020-04-10 |
CN110977767B true CN110977767B (en) | 2021-07-02 |
Family
ID=70084177
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911102331.5A Active CN110977767B (en) | 2019-11-12 | 2019-11-12 | Casting defect distribution detection method and casting polishing method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110977767B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111203766B (en) * | 2020-04-20 | 2020-08-14 | 广东博智林机器人有限公司 | Wall surface polishing path planning method, device, equipment and medium |
CN112792619B (en) * | 2020-12-17 | 2022-12-09 | 湖南中南智能装备有限公司 | Polishing path planning method based on machine vision |
CN113554355B (en) * | 2021-09-17 | 2021-12-03 | 江苏正金建筑装饰工程有限公司 | Road engineering construction management method and system based on artificial intelligence |
CN114559348B (en) * | 2022-01-13 | 2023-05-30 | 西安理工大学 | Ballastless track slab mold cleaning system and method based on industrial robot |
CN115330803B (en) * | 2022-10-17 | 2023-01-24 | 菲特(天津)检测技术有限公司 | Surface defect data enhancement method and device, electronic equipment and storage medium |
CN117754464B (en) * | 2023-07-28 | 2024-05-17 | 嘉兴市安得特种设备科技有限公司 | Intelligent polishing control method and system for aluminum die castings |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102854191A (en) * | 2012-07-18 | 2013-01-02 | 湖南大学 | Real-time visual detection and identification method for high speed rail surface defect |
MX2012015238A (en) * | 2012-12-19 | 2013-02-20 | Mextrauma S A De C V | Polish and isnpection system assisted by computer for joint components. |
CN107052950A (en) * | 2017-05-25 | 2017-08-18 | 上海莫亭机器人科技有限公司 | A kind of complex-curved sanding and polishing system and method |
CN108000250A (en) * | 2017-12-07 | 2018-05-08 | 长沙长泰机器人有限公司 | A kind of method and system of casting grinding |
CN108226164A (en) * | 2017-12-29 | 2018-06-29 | 深圳市智能机器人研究院 | A kind of the robot polishing process and system of view-based access control model detection |
CN109483369A (en) * | 2018-12-13 | 2019-03-19 | 中国船舶重工集团公司第七六研究所 | A kind of robot polishing system and its control method with 3D vision |
CN110385616A (en) * | 2019-06-25 | 2019-10-29 | 共享智能铸造产业创新中心有限公司 | The casting automatically grinding method of view-based access control model detection |
-
2019
- 2019-11-12 CN CN201911102331.5A patent/CN110977767B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102854191A (en) * | 2012-07-18 | 2013-01-02 | 湖南大学 | Real-time visual detection and identification method for high speed rail surface defect |
MX2012015238A (en) * | 2012-12-19 | 2013-02-20 | Mextrauma S A De C V | Polish and isnpection system assisted by computer for joint components. |
CN107052950A (en) * | 2017-05-25 | 2017-08-18 | 上海莫亭机器人科技有限公司 | A kind of complex-curved sanding and polishing system and method |
CN108000250A (en) * | 2017-12-07 | 2018-05-08 | 长沙长泰机器人有限公司 | A kind of method and system of casting grinding |
CN108226164A (en) * | 2017-12-29 | 2018-06-29 | 深圳市智能机器人研究院 | A kind of the robot polishing process and system of view-based access control model detection |
CN109483369A (en) * | 2018-12-13 | 2019-03-19 | 中国船舶重工集团公司第七六研究所 | A kind of robot polishing system and its control method with 3D vision |
CN110385616A (en) * | 2019-06-25 | 2019-10-29 | 共享智能铸造产业创新中心有限公司 | The casting automatically grinding method of view-based access control model detection |
Also Published As
Publication number | Publication date |
---|---|
CN110977767A (en) | 2020-04-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110977767B (en) | Casting defect distribution detection method and casting polishing method | |
CN111292305B (en) | Improved YOLO-V3 metal processing surface defect detection method | |
CN111951237B (en) | Visual appearance detection method | |
CN109141232B (en) | Online detection method for disc castings based on machine vision | |
CN110163853B (en) | Edge defect detection method | |
CN109507192B (en) | Magnetic core surface defect detection method based on machine vision | |
CN106127148B (en) | A kind of escalator passenger's anomaly detection method based on machine vision | |
CN109685760B (en) | MATLAB-based SLM powder bed powder laying image convex hull depression defect detection method | |
CN114693610A (en) | Welding seam surface defect detection method, equipment and medium based on machine vision | |
CN111982916A (en) | Welding seam surface defect detection method and system based on machine vision | |
CN115984271B (en) | Metal burr identification method based on corner detection | |
CN105891233A (en) | Intelligent detection system for surface defects of lens based on machine vision and implementation method of intelligent detection system | |
CN114897908B (en) | Machine vision-based method and system for analyzing defects of selective laser powder spreading sintering surface | |
CN113935666B (en) | Building decoration wall tile abnormity evaluation method based on image processing | |
CN110348461A (en) | A kind of Surface Flaw feature extracting method | |
CN116309277A (en) | Steel detection method and system based on deep learning | |
CN115375686A (en) | Glass edge flaw detection method based on image processing | |
CN115063423A (en) | Mechanical casting cold and hot crack self-adaptive identification method based on computer vision | |
CN114445707A (en) | Intelligent visual fine detection method for defects of bottled water labels | |
CN112304957A (en) | Machine vision-based intelligent detection method and system for appearance defects | |
CN112258444A (en) | Elevator steel wire rope detection method | |
CN111932490A (en) | Method for extracting grabbing information of visual system of industrial robot | |
CN116228651A (en) | Cloth defect detection method, system, equipment and medium | |
CN110728286B (en) | Abrasive belt grinding material removal rate identification method based on spark image | |
CN113588656A (en) | Novel bolt appearance size detection method |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |