CN111624203B - Relay contact point alignment non-contact measurement method based on machine vision - Google Patents
Relay contact point alignment non-contact measurement method based on machine vision Download PDFInfo
- Publication number
- CN111624203B CN111624203B CN202010545565.3A CN202010545565A CN111624203B CN 111624203 B CN111624203 B CN 111624203B CN 202010545565 A CN202010545565 A CN 202010545565A CN 111624203 B CN111624203 B CN 111624203B
- Authority
- CN
- China
- Prior art keywords
- relay
- contact
- image
- point
- gradient
- 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
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Immunology (AREA)
- Chemical & Material Sciences (AREA)
- Computing Systems (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- Software Systems (AREA)
- Signal Processing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a non-contact measurement method for relay contact alignment based on machine vision, which comprises the following steps: and processing the sequence images of the relay multi-contact point motion process obtained by shooting by using image acquisition equipment such as a high-speed camera and the like by utilizing an image processing technology of computer machine vision, and measuring and obtaining the relay contact point alignment parameter in a non-contact mode. Firstly, carrying out normalization processing on the image to obtain a conversion relation between image pixels and actual length, then carrying out related pre-processing on the series of images to establish a template of a relay target joint, finally carrying out iterative computation to obtain the most similar area with the target template in the image by using a similarity matching method to obtain the coordinate position change condition of a plurality of pairs of target joints of the relay, and comparing and analyzing the action condition of the plurality of pairs of joints to obtain the coordinate errors of the simultaneous contact of all the joints, wherein the joint alignment of the relay is obtained through normalization processing. The non-contact measuring method based on machine vision provided by the invention can relatively accurately measure and obtain the contact alignment degree of the relay, and has higher precision and efficiency.
Description
Technical Field
The invention belongs to the technical field of relay detection, and particularly relates to a non-contact measuring method for relay contact alignment based on machine vision.
Background
The relay plays roles of automatic control, relay protection, circuit switching and the like in a circuit, and generally has the advantages of high electrical reliability, good isolation performance, simple structure, good economy and the like, so the relay is widely applied to an automatic control system, a power relay protection system, a railway system, an aerospace system and the like, and is an important electrical equipment element in an industrial system in China.
In relay manufacturing enterprises, professional performance detection of relays is an important working step in order to ensure the qualification rate of products. Various parameters of the relay are criteria for evaluating whether a relay is acceptable or not. Currently, the performance parameters of relays are broadly divided into two main categories: mechanical parameters and electrical parameters. Common mechanical parameters mainly include contact alignment, contact sweep, absolute clearance, contact pressure, etc. The contact point alignment is an important mechanical parameter of the relay, when all the contact points of the same relay are simultaneously applied to a circuit, as the process errors in the production process make the simultaneous contact of multiple contact points difficult, the simultaneous contact errors of all the contact points are called the contact point alignment, and the smaller the value is, the better the action synchronism of the multiple contact points of the relay is. At present, a method for measuring the contact alignment of a relay usually adopts manual contact measurement, a measurer fixes the relay on a special base, manually fine-adjusts the relay and observes an on-off indicator lamp of the special base, and repeatedly performs contact measurement and records by using a dial indicator and other measuring tools. Because the measuring tool directly contacts with the mechanical structure of the relay to be measured in the contact measurement, the relay generates irreversible influence, and the manual contact type measuring precision is low, the operation is complex, the efficiency is low and the subjective influence of the measuring staff is large, so that the precision of the measuring result cannot be ensured.
The computer machine vision technology is an image processing technology and method for performing edge detection, feature recognition, template matching and parameter measurement on an image by adopting image acquisition equipment such as a high-speed camera and the like and a digital image processing technology to cooperatively work. The method has the advantages of high processing speed, high measurement precision, reasonable cost performance, no influence of subjective factors, realization of non-contact measurement and the like, and is increasingly applied to the technical field of relay detection.
Disclosure of Invention
The invention aims to solve the technical problem of providing a relay contact alignment non-contact measuring method based on machine vision by utilizing computer machine vision and image processing technology, which is used for solving the defects of the traditional contact measuring method.
The invention relates to a non-contact measuring method for the contact alignment degree of a relay, which comprises the following steps:
1. the method comprises the steps of using a time relay to regularly control the on-off of a power supply of the relay to be tested, reasonably selecting a relay action period, enabling each pair of contacts of the relay to be tested to reliably act, using a parallel light source to illuminate a plurality of pairs of contacts of the relay to be tested so as to improve the brightness of images, adjusting parameters of image acquisition equipment such as a high-speed camera and the like to enable the plurality of pairs of contacts of the relay to be tested to be imaged clearly, starting the image acquisition equipment such as the high-speed camera and the like to shoot in real time, and obtaining a sequence image of a complete suction process and a complete release process to be used as a sample for image processing.
2. Because the acquired sequential image samples are stored in the computer in the form of digital images, the storage form of single-frame images is a multidimensional matrix taking pixels as elements, the pixel value of a single point in the images corresponds to one element in the matrix, and the change range of the pixel value corresponding to the change from dark to light is 0 to 255. Therefore, normalization processing is needed between the actual length and the pixels, and a conversion relation between the pixels of the digital image and the actual length is obtained.
3. Before the image processing of the sequence image sample obtained by high-speed shooting by using the computer digital image processing technology, the image needs to be pre-processed for the convenience of processing and the reduction of the calculated amount, the elimination of some noise interference of the sample.
(1) Image noise filtering: in the process of capturing a sample of a sequence image by a high-speed camera, interference is inevitably generated on the sample image for various reasons, and therefore, noise filtering processing should be performed on the sample first. A designated odd-order matrix in the digital sample image
In (3) operating the pixel e value of the central point of the center, selecting a coefficient matrix ++>
The closer the center point e is in position, the larger the corresponding element value in the coefficient matrix is. Obtaining convolution sum of square matrix A and coefficient square matrix H>
The pixel value of the original center point e is replaced by sum. The noise filtering method is applied to carry out noise filtering processing on each pixel point in the image, so that the purposes of smoothing the image and filtering interference noise are achieved.
(2) Channel conversion of color images: in the process of shooting the relay by using the image acquisition equipment such as the high-speed camera, more complete real image data information is reserved for ensuring the sequence image samples, image distortion and feature loss are reduced, and the shot image is set to be a color image which contains data of three BGR color channels. In the image processing, in order to reduce the calculation amount and increase the operation speed, the original color image can be converted into a black-and-white image, and the color image with three color channels is converted into the black-and-white image with a single color channel on the premise of keeping all data of the original image.
(3) Background point pixel interference is eliminated: in order to conveniently detect the edge characteristic information of the target in the image, the color characteristic information is required to be used as a judgment basis. In a black-and-white image of a single color channel, in order to make the target color feature more prominent and eliminate the interference of the background pixel point, the original information can be retained by setting a pixel threshold value, when the pixel value of a certain point exceeds a given threshold value, and when the pixel value of the certain point is smaller than the given threshold value, the pixel value of the certain point is set to 0, so that the key pixel information is highlighted and the adverse effect of the background factor is ignored.
(4) Establishing a target template: on the basis of the processing of the three steps, in order to identify and position the contact point of the image relay, the edge characteristic information of the target needs to be extracted to form template data. By calculating the gradient of pixel points
Gradient direction θ=arctan (G y /G x ) To determine whether the pixel is a boundary. By transverse coefficient->
Longitudinal coefficient->
Multiplying the square matrix A to obtain transverse gradient +.>
And longitudinal gradient->
Meanwhile, in order to eliminate spurious response in the detection and calculation process, the non-maximum value of the spurious response is restrained; by setting an upper limit valueAnd a lower limit value to distinguish between true and potential edges; and finally, extracting edge characteristic information of the target by discarding the isolated weak edges. Drawing a minimum circumscribed rectangle of the edge of the target, intercepting an image in a rectangle frame, identifying characteristics in the frame, and storing to form a target template, so as to finish the establishment of the target template.
4. Based on the motion rule of the relay contacts, the motion range of the relay contacts is relatively fixed in an image, so that before the sample image is subjected to the preprocessing operation, the motion range of each pair of contacts can be defined, the motion range of a single pair of contacts is respectively defined by a rectangular frame in the image, and pixel values of areas except the rectangular frame are all set to be pure black, so that a key area is highlighted, adverse effects of background are ignored, and the processing speed is increased. The image preprocessing operation is applied to perform preprocessing on a sample image, the degree of difference between a single template and a place covered by the template in the single-frame sample image is calculated by setting a similarity value and applying a similarity matching algorithm, the area which is most similar to the established template in the sample image is found by iterative calculation, then the result is stored in a matrix, if the original image is of the AxB size and the template is of the MxN size, the matrix of the output result is of the (A-M+1) x (B-N+1) order, and in order to eliminate errors caused by angle change, the coordinate of the smallest circumscribed rectangle central point of the template is usually output as the processing result, and the coordinate value is in units of pixels. And after the similarity matching of the first frame of image is successful, repeating the similarity matching operation according to the frame reading sequence image in a circulating way, and obtaining the coordinate value of the corresponding central point after the matching of each frame of image. Repeating the above operation on multiple pairs of joints of the relay to be tested to obtain the coordinate value change condition of the multiple pairs of joints, obtaining the corresponding image frame number when the first pair of joints start to contact and the last pair of joints start to contact according to the analysis result, measuring the distance of the middle joint support of the two frames of images in the vertical direction, and obtaining the joint alignment value of the relay to be tested through normalization processing.
The invention has the advantages that: the measuring system has the advantages of high running speed, high measuring precision, good cost performance, no influence of subjective factors and the like, and has the greatest advantages of realizing non-contact measurement and real-time dynamic measurement with higher processing efficiency on the contact point alignment value of the measured relay, avoiding the irreversible influence of manual contact measurement on the relay, and greatly improving the measuring precision and efficiency of the contact point alignment of the relay.
Drawings
FIG. 1 is a flow chart of a non-contact relay contact alignment measurement method based on template matching
FIG. 2 is a schematic side view of a single pair of upper, middle and lower contacts of a relay
Fig. 3 is a schematic view of a relay front 8-butt joint
FIG. 4 is a schematic diagram of the positions of the 40 target templates of the relay
Detailed Description
The invention will now be described in detail with reference to the drawings and specific examples.
As shown in fig. 1, the embodiment of the invention provides a non-contact measurement method for relay contact alignment based on machine vision, which comprises the following steps:
step one: the relay to be tested is fixed on a special base, the on-off of a base power supply is controlled by a time relay to control whether the relay to be tested acts or not, and the primary action period is set to be 3s. On the basis of guaranteeing fixed action time, the shooting frequency of the high-speed camera is set to 1500Hz, the lens parameters are manually adjusted to enable 8 pairs of joints of the relay to be imaged clearly, and parallel light sources are used for assisting in illumination so as to increase the brightness of pictures. The time relay is turned on to enable the relay to be tested to act once, and meanwhile image acquisition equipment such as high-speed shooting is turned on to shoot once, so that acquisition of sequence sample images is completed, and 4500 frames of sample images are obtained.
Step two: and (3) carrying out normalization processing on the obtained sample image, and pasting a cross marker on the relay movable joint support 4 as shown in a front schematic diagram 5 of the relay joint in fig. 3, wherein the distance between the centers of two BMW marks is L=10mm. Selecting a first frame of picture to measure the circle centers of the two BMW marks to obtain circle center coordinates (526.21, 705.05), (530.89, 568.52), and obtaining a correction coefficient K=1 to obtain a conversion coefficient as follows:
step three: as shown in the following schematic side view of the single-docking point of fig. 2, the single-docking point of the relay includes an upper-docking point 1, a lower-docking point 2, a middle- docking point 3, and 3 total, and as can be seen from the schematic front view of the relay of fig. 3, there are 8 total-docking points of the relay to be tested, each of the upper-docking point and the lower-docking point has 2 contact points with the middle-docking point, so that 40 target templates should be determined, including 16 upper-docking point target templates, 8 middle-docking point target templates, and 16 lower-docking point target templates, wherein the single upper-docking point and lower-docking point include two contact point target templates.
The first frame image is read to carry out image pre-processing, and the image is sequentially subjected to image noise filtering processing, color image channel conversion processing, background point pixel interference elimination processing and target template establishment operation. In the background point pixel interference elimination processing, a threshold value is set to 127, a part with a pixel value larger than 127 is unchanged, and a pixel point pixel with a pixel value smaller than 127 is set to 0. In the operation of establishing a target template, the transverse gradient and the longitudinal gradient of each pixel point are calculated according to a gradient formula:
carry in->
θ＝arctan(G y /G x ) The gradient and gradient direction of the pixel point are obtained through formula calculation, and in order to simplify the calculation, the gradient direction of the pixel is dispersed into 8 directions corresponding to 8 pixels around the gradient direction of the pixel. Setting an upper limit value Q defining whether the pixel point is a boundary max And a lower limit value Q min When the point gradient G is greater than Q max Then identify the pointIs a boundary; when gradient G is smaller than Q min Then the point is identified as non-boundary and excluded; when gradient G is smaller than Q max And is greater than Q min Judging whether the boundary has a direct connection relation with the determined boundary, if so, judging the boundary, otherwise, judging the boundary as a non-boundary point. And extracting edge information of 40 target templates, respectively drawing the minimum circumscribed rectangles of the target templates, carrying out feature recognition on the target templates in the rectangles, and storing the target templates to finish the establishment of the target templates.
Step four: before the sample image is subjected to the preprocessing operation, a rectangular frame is used for defining the movement range of a plurality of pairs of joints, image information in the rectangular frame is reserved, and pixels outside the frame are set to 0. And (3) reading sample images according to frame sequences, processing each frame of image by using the image pre-processing method, setting a similarity index value to be 0.9, sequentially reading the established target templates in the third step, finding out the most similar area of each template meeting the similarity index value in the sample image through iterative calculation, drawing out by using a rectangular frame mark, outputting the coordinates of the central point of the rectangular frame as a result, sequentially reading 4500 sample images, and repeating the operation until the process is completed.
Step five: taking the sucking process as an example, as shown in the analysis result of the center point coordinate output in the fourth step, in the 1116 th frame image, the upper contact point and the middle contact point represented by P1 and P2 are in first contact with each other, the corresponding contact point group number is 1, and in the 1120 th frame image, the upper contact point and the middle contact point represented by P17 and P18 are in last contact with each other, the corresponding contact point group number is 5, and the coordinates are shown in the following table 1:
TABLE 1
Step six: the circle center coordinates of the upper BMW markers in the 1116 th and 1120 th frame images are (498.89, 731.11), (498.84, 734.16) respectively, the change of the circle center ordinate y represents the distance that the movable contact support 4 longitudinally walks, and the contact alignment value measured in a non-contact way can be obtained through calculation:
the above results indicate that: the invention can effectively carry out non-contact measurement on the contact point alignment degree of the relay.
Claims (7)
1. The non-contact relay contact alignment measuring method based on machine vision is characterized by comprising the following steps of:
s1, in the one-time action process of a plurality of pairs of contacts of a relay, acquiring sequence images of the contact movement process of the relay in real time by using an image acquisition device;
s2, establishing a conversion relation between the image pixels and the actual length, and obtaining coordinates of circle center points of two cross BMW marks adhered to the movable contact support
And->
The representative actual distance between two circle centers is L, and the correction coefficient K is selected to obtain the normalized conversion coefficient between the image pixel and the actual length +.>
；
S3, carrying out noise filtering processing on the acquired sequence images, and carrying out specified odd-order square matrix in the digital sample images
And coefficient square->
Performing convolution operation to obtain convolution sum
The e point pixel value is replaced by a sum value, and the effect of filtering noise interference can be achieved through the operation;
s4, converting the channel of the color image into a black-and-white image with a single color channel on the premise of keeping all data of the original image;
s5, eliminating the interference of the background point pixels, wherein the pixel threshold value is set, when the pixel value of a certain point exceeds a given threshold value, the pixel value is not processed, and when the pixel value is smaller than the given threshold value, the pixel value is set as a fixed value, and the interference of the background point pixels can be effectively eliminated through the operation;
s6, identifying and positioning the contact point of the image relay, extracting the edge characteristic information of the target, forming template data,
using pixel gradient magnitude and direction
、/>
To determine whether the pixel is boundary, and to determine the lateral coefficient +.>
Longitudinal coefficient->
Respectively multiplying the two square matrixes A to obtain transverse gradients
And longitudinal gradient->
The non-maximum value is restrained to eliminate the stray effect, the boundary point is determined by limiting the upper limit value and the lower limit value of the gradient to finish the edge detection of the target, and the minimum circumscribed rectangle of the target is drawn and intercepted to finish the establishment of the target template;
s7, in a limited motion area of the target joint, obtaining position coordinates of each target template in each frame of image by setting a similarity value and applying a similarity matching algorithm, analyzing a result and obtaining corresponding image frames when the first joint is contacted and the last joint is contacted;
and S8, measuring the distance travelled by the movable contact support between the two determined frames of images to obtain the relay contact alignment value.
2. The machine vision-based relay contact alignment non-contact measurement method according to claim 1, wherein: the relay is fixed on the base, and the contact action of the relay can be controlled by controlling the on-off of the base power supply.
3. The machine vision-based relay contact alignment non-contact measurement method according to claim 1, wherein: the time relay is connected in series in the base power supply loop, and the contact action of the relay can be periodically controlled by setting the timing time of the time relay so as to ensure the synchronism of the contact action and the shooting process.
4. The machine vision-based relay node alignment non-contact measurement method according to claim 1, wherein: and S2, when the normalization processing of the image pixels and the actual length is carried out, a cross BMW marker with the distance of 10mm between two circle centers is stuck on a moving contact support on the front of the relay in advance.
5. The machine vision-based relay node alignment non-contact measurement method according to claim 1, wherein: step S6, identifying and positioning the contact point of the image relay, and dispersing the gradient direction of the pixel into 8 directions corresponding to 8 pixels around the gradient direction when extracting the edge characteristic information of the target.
6. The machine vision-based relay node alignment non-contact measurement method according to claim 1, wherein: regarding the upper limit value, the lower limit value Qmax and Qmin of the gradient defined in the step S6, judging that when the gradient G of the point is larger than or equal to Qmax, the point is considered as a boundary; when the gradient G is less than or equal to Qmin, the point is determined to be a non-boundary and is excluded; when the gradient G is smaller than Qmax and larger than Qmin, judging whether the gradient G has a direct connection relation with the determined boundary, if so, judging the gradient G as the boundary, otherwise, judging the gradient G as a non-boundary point.
7. The machine vision-based relay node alignment non-contact measurement method according to claim 1, wherein: when the distance travelled by the middle joint support of the two obtained images is calculated, the circle center ordinate of the corresponding cross BMW mark in the two images can be measured respectively
And->
Then the contact alignment: />
。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010545565.3A CN111624203B (en) | 2020-06-16 | 2020-06-16 | Relay contact point alignment non-contact measurement method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010545565.3A CN111624203B (en) | 2020-06-16 | 2020-06-16 | Relay contact point alignment non-contact measurement method based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111624203A CN111624203A (en) | 2020-09-04 |
| CN111624203B true CN111624203B (en) | 2023-07-04 |
Family
ID=72259468
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010545565.3A Active CN111624203B (en) | 2020-06-16 | 2020-06-16 | Relay contact point alignment non-contact measurement method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111624203B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113327232B (en) * | 2021-05-28 | 2022-08-05 | 河北工业大学 | Relay contact set contact surface alignment method, apparatus and computer storage medium |
| CN113251939A (en) * | 2021-06-23 | 2021-08-13 | 沈阳铁路信号有限责任公司 | Method for measuring contact gap of non-contact railway relay |
| CN113344977B (en) * | 2021-06-29 | 2022-05-27 | 河北工业大学 | Contact pressure measurement model construction method based on image processing |
| CN115115584B (en) * | 2022-06-07 | 2024-03-29 | 河北工业大学 | Relay contact pressure measurement method based on image recognition tracking technology |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001074418A (en) * | 1999-09-09 | 2001-03-23 | Dainippon Printing Co Ltd | Method and apparatus for measuring moving object |
| CN1888929A (en) * | 2006-07-21 | 2007-01-03 | 李春鹤 | Apparatus and method for detecting railway signal relay connector uniform |
| US7860344B1 (en) * | 2005-05-06 | 2010-12-28 | Stochastech Corporation | Tracking apparatus and methods using image processing noise reduction |
| JP2011139123A (en) * | 2009-12-25 | 2011-07-14 | Casio Computer Co Ltd | Image processor and image processing program |
| CN106371009A (en) * | 2016-09-23 | 2017-02-01 | 上海地铁维护保障有限公司 | Test device and test method of safety type relay |
| CN107084664A (en) * | 2017-04-27 | 2017-08-22 | 沈阳铁路信号有限责任公司 | A kind of relay tip sweeps the measurement apparatus and its measuring method of journey |
| CN107894189A (en) * | 2017-10-31 | 2018-04-10 | 北京艾克利特光电科技有限公司 | A kind of EOTS and its method for automatic tracking of target point automatic tracing |
| CN109323729A (en) * | 2018-12-17 | 2019-02-12 | 桂林航天电子有限公司 | A kind of relay contact spring system automatic detection device and method |
| CN110008968A (en) * | 2019-04-09 | 2019-07-12 | 哈尔滨新中新电子股份有限公司 | A kind of robot clearing automatic trigger method based on image vision |
| CN110070557A (en) * | 2019-04-07 | 2019-07-30 | 西北工业大学 | A kind of target identification and localization method based on edge feature detection |
-
2020
- 2020-06-16 CN CN202010545565.3A patent/CN111624203B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001074418A (en) * | 1999-09-09 | 2001-03-23 | Dainippon Printing Co Ltd | Method and apparatus for measuring moving object |
| US7860344B1 (en) * | 2005-05-06 | 2010-12-28 | Stochastech Corporation | Tracking apparatus and methods using image processing noise reduction |
| CN1888929A (en) * | 2006-07-21 | 2007-01-03 | 李春鹤 | Apparatus and method for detecting railway signal relay connector uniform |
| JP2011139123A (en) * | 2009-12-25 | 2011-07-14 | Casio Computer Co Ltd | Image processor and image processing program |
| CN106371009A (en) * | 2016-09-23 | 2017-02-01 | 上海地铁维护保障有限公司 | Test device and test method of safety type relay |
| CN107084664A (en) * | 2017-04-27 | 2017-08-22 | 沈阳铁路信号有限责任公司 | A kind of relay tip sweeps the measurement apparatus and its measuring method of journey |
| CN107894189A (en) * | 2017-10-31 | 2018-04-10 | 北京艾克利特光电科技有限公司 | A kind of EOTS and its method for automatic tracking of target point automatic tracing |
| CN109323729A (en) * | 2018-12-17 | 2019-02-12 | 桂林航天电子有限公司 | A kind of relay contact spring system automatic detection device and method |
| CN110070557A (en) * | 2019-04-07 | 2019-07-30 | 西北工业大学 | A kind of target identification and localization method based on edge feature detection |
| CN110008968A (en) * | 2019-04-09 | 2019-07-12 | 哈尔滨新中新电子股份有限公司 | A kind of robot clearing automatic trigger method based on image vision |
Non-Patent Citations (4)
| Title |
|---|
| Observation of arc-emitted light between slowly opening electrical contacts using a high-speed camera;J. Sekikawa;《Proceedings of the 50th IEEE Holm Conference on Electrical Contacts and the 22nd International Conference on Electrical Contacts Electrical Contacts》;20041231;全文 * |
| Research on Arc Burning and Contact Bounce of DC Relays by High Speed Photograph Method;C. Zang;《2008 Proceedings of the 54th IEEE Holm Conference on Electrical Contacts》;20081231;全文 * |
| 航天继电器微观形貌参数与电气参数融合方法研究;李文华;《航天控制》;20191231;第37卷(第7期);全文 * |
| 铁路继电器的触头运动分析和性能退化研究;赵月山;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20220315(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111624203A (en) | 2020-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111624203B (en) | Relay contact point alignment non-contact measurement method based on machine vision | |
| CN108335331B (en) | Binocular vision positioning method and equipment for steel coil | |
| CN106780623A (en) | A kind of robotic vision system quick calibrating method | |
| CN109685078B (en) | Infrared image identification method based on automatic annotation | |
| CN108470356B (en) | Target object rapid ranging method based on binocular vision | |
| CN111260788B (en) | Power distribution cabinet switch state identification method based on binocular vision | |
| CN105740856A (en) | Method for reading readings of pointer instrument based on machine vision | |
| CN110334727B (en) | Intelligent matching detection method for tunnel cracks | |
| CN113870267B (en) | Defect detection method, defect detection device, computer equipment and readable storage medium | |
| CN104749801B (en) | High Precision Automatic optical detecting method and system | |
| CN111724354B (en) | Image processing-based method for measuring wheat ear length and wheat ear number of multiple wheat plants | |
| CN113538583A (en) | Method for accurately positioning position of workpiece on machine tool and vision system | |
| CN105701809A (en) | Flat-field correction method based on line-scan digital camera scanning | |
| CN110503623A (en) | A method of Bird's Nest defect on the identification transmission line of electricity based on convolutional neural networks | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN111652844B (en) | X-ray defect detection method and system based on digital image region growing | |
| CN113822810A (en) | Method for positioning workpiece in three-dimensional space based on machine vision | |
| CN112381751A (en) | Online intelligent detection system and method based on image processing algorithm | |
| CN112833821B (en) | Differential geometric three-dimensional micro-vision detection system and method for high-density IC welding spots | |
| CN113657339A (en) | Instrument pointer counting and reading method and medium based on machine vision | |
| CN110751690B (en) | Visual positioning method for milling machine tool bit | |
| CN116563391B (en) | Automatic laser structure calibration method based on machine vision | |
| CN114926417B (en) | Microscopic imaging detection method and system for weak scratches on surface of high-density flexible FPC | |
| CN113436214B (en) | Brinell hardness indentation circle measuring method and system and computer readable storage medium | |
| CN113536895A (en) | Disc pointer meter identification 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 | ||
| DD01 | Delivery of document by public notice |
Addressee: Xie Weidong Document name: Notification of Qualified Procedures |
|
| DD01 | Delivery of document by public notice | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |