CN110610516B - Railway fastener nut center positioning method - Google Patents
Railway fastener nut center positioning method Download PDFInfo
- Publication number
- CN110610516B CN110610516B CN201910789963.7A CN201910789963A CN110610516B CN 110610516 B CN110610516 B CN 110610516B CN 201910789963 A CN201910789963 A CN 201910789963A CN 110610516 B CN110610516 B CN 110610516B
- Authority
- CN
- China
- Prior art keywords
- fastener
- image
- spike
- area
- center
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 229910000831 Steel Inorganic materials 0.000 claims description 15
- 239000010959 steel Substances 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 9
- 241001669679 Eleotris Species 0.000 claims description 7
- 230000001174 ascending effect Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 4
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 3
- 230000001629 suppression Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004575 stone 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
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a railway fastener nut center positioning method, which mainly comprises the following implementation processes: and detecting the center of the spike by using the contour and the regional texture information of the spike, and determining the center of the nut by using the center of the spike. The invention can accurately realize the accurate positioning of the center of the fastener nut, and has better positioning accuracy and positioning effect for the fastener nuts under different environments.
Description
Technical Field
The invention relates to the field of rail traffic detection equipment, in particular to a method for positioning the center of a railway fastener nut.
Background
Fasteners are important components of railway systems that require periodic disassembly for maintenance to ensure safe operation of the railway. At present, the fastener is dismantled mainly and relies on artifical the dismantlement, and is not only inefficient like this, has certain danger moreover. Therefore, the design of the automatic fastener dismounting device has very important significance. For the automatic dismounting device, the dismounting of the fastener nut is mainly completed, so that the completion of the accurate positioning of the fastener nut is the key for the successful design of the automatic dismounting device of the fastener.
The accurate positioning of the nut mainly comprises two steps, namely, the pixel coordinate of the center of the nut is accurately positioned from the obtained fastener image; secondly, the obtained pixel coordinates are converted into world coordinates, so that the distance between the nut and the dismounting device is obtained, and the accurate physical positioning of the nut is realized. The method provided by the patent mainly realizes the accurate positioning of the coordinates of the central pixel of the nut.
At present, related researches on precise positioning of coordinates of a central pixel of a fastener nut are few, and a method for realizing the positioning of the center of the nut by utilizing texture inhibition and combination characteristic grading exists. The method comprises the steps of firstly, carrying out texture suppression on a fastener outline image; then positioning the center of a circular gasket below the nut and the center of the nut by using the processed contour image; and finally, the final nut center is obtained through the combination of the two positioned centers to realize the positioning. The process is schematically illustrated in figure 1. In addition, most studies have focused on the location of the fastening region, i.e., extracting the fastening region from the original image, and have no reference to the nut center location.
Although the texture suppression and feature classification combined method can be used for positioning the coordinates of the central pixel of the nut, the positioning accuracy is low. The characteristic that the centers of the circular gasket and the nut are the same and the respective shape characteristics are utilized, the center of the gasket and the center of the nut are obtained through positioning, and the final center of the nut is obtained through the common action of the two centers. However, in an actual railway fastener system, the centers of the nut and the washer are fundamentally different, and a certain deviation exists between the centers of the nut and the washer. Meanwhile, the shape characteristic of the nut is unstable, and the hexagonal characteristic of the nut cannot be basically recognized due to factors such as corrosion, oil stain coverage and the like. At this time, the center positioning of the nut by using the characteristic of "common center" and the shape characteristic will generate larger deviation, thereby causing the positioning failure and lowering the positioning accuracy. In addition, the fastener region positioning is only for the whole fastener, so that the fastener region can be accurately obtained from the original image as a target, and the accurate positioning of the nut center cannot be completed.
It can be seen that the texture suppression and combination feature classification method has a low accuracy in positioning the center of the nut of the fastener, and particularly, in the case of substantial loss of the shape feature of the nut in fig. 2, the center of the nut cannot be directly positioned. The center of the positioning washer is indirectly used as the center of the nut, so that large deviation exists inevitably, and the positioning fails.
Disclosure of Invention
The invention aims to solve the technical problem that the center of a railway fastener nut is accurately positioned by providing a method for positioning the center of the railway fastener nut aiming at the defects of the prior art.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: a railway fastener nut center positioning method mainly comprises the following implementation processes: and detecting the center of the spike by using the contour and the regional texture information of the spike, and determining the center of the nut by using the center of the spike.
The method comprises the following concrete implementation steps:
1) Acquiring a steel rail image, and binarizing the steel rail image, wherein the maximum connected region of the obtained binarized image is a steel rail region;
2) Dividing the steel rail image into two parts by using the steel rail area, respectively intercepting sub-images from the two parts of images to carry out binarization processing, and carrying out AND operation on the two obtained binarization images to obtain a total binarization image, wherein the maximum communication area in the total binarization image is the sleeper area;
3) Obtaining a rough positioning image of the fastener by utilizing the steel rail area and the sleeper area;
4) Selecting a standard fastener image as a fastener template image, calculating the correlation between the rough positioning image of the fastener and the fastener template image, and obtaining the position with the maximum correlation value, namely the accurate position of the fastener, thus finishing the accurate positioning of the fastener and simultaneously extracting and obtaining a fastener sub-image;
5) Extracting the initial contour of the sub-image of the fastener, and removing the irrelevant contour in the image to obtain a fastener contour image;
6) Performing circular fitting on the fastener outline image to obtain an initial spike region, and screening the initial spike region to obtain a candidate spike region;
7) Calculating a gray level mean value m and an information entropy e of the candidate spike region, normalizing the gray level mean value m and the information entropy e to obtain a normalized gray level mean value and an information entropy texture feature, and selecting a final spike region by using the gray level mean value and the information entropy texture feature;
8) And determining the center coordinates of the spike by using the final spike area, wherein the center coordinates are the center coordinates of the nut.
In step 4), the calculation formula of the correlation R is as follows:
wherein, theta m,n And θ' m,n The pixel points (m, n) of the fastener coarse positioning image and the pixel points (m, n) in the fastener template image are respectively the orientation field characteristic values.
Characteristic value of the direction field
I x And I y Pixel horizontal and vertical gradients.
In step 5), defining that the profile satisfying any one of the following conditions is an irrelevant profile which is obviously larger than the spike profile, namely the profile to be removed:
wherein D is L And D W Pixel values for each contour connected region along the length and width of the image are shown, respectively, and L and W represent the length and width of the fastener sub-image.
The initial spike zone set diameter is in the interval
And the spike region with the circle center pixel abscissa located in the range of 1/3 width in the middle of the fastener outline image is a candidate spike region set.
The calculation formula of the gray level mean value m and the information entropy e is as follows:
z i the gray value of the pixel point in the candidate trace nail region, p (z) i ) The number of occurrences of each gray value.
The normalized gray level mean value and information entropy texture feature calculation formula is as follows:
wherein c is a candidate laneThe staple zone radius.
In step 7), the step of selecting a final spike region by using the gray average value and the information entropy texture features comprises:
1) Sorting the gray level mean value and the information entropy texture features of each candidate spike region, wherein the gray level mean value is sorted in a descending order, and the information entropy texture features are sorted in an ascending order;
2) Giving the most votes to the areas with the maximum gray mean value and the minimum information entropy texture feature, and giving the sequentially decreased votes to other areas according to the descending order of the gray mean value and the ascending order of the information entropy texture feature;
3) And adding the votes obtained from each candidate spike area to obtain the area with the most votes, namely the final spike area.
Compared with the prior art, the invention has the beneficial effects that: the invention can accurately realize the accurate positioning of the center of the fastener nut, and has better positioning accuracy and positioning effect for the fastener nuts under different environments.
Drawings
FIG. 1 is a schematic view of a texture suppression binding feature classification method for centering a nut;
FIG. 2 is a diagram of a nut and a circular washer in an actual railway; FIG. 2 (a) the nut shape feature disappears; FIG. 2 (b) circular shims are biased;
FIG. 3 is a block diagram of a positioning algorithm flow;
FIG. 4 shows the fastener in rough orientation;
FIG. 5 shows the fastener in a fine position;
FIG. 6 (a) an initial profile image; FIG. 6 (b) the processed profile image;
FIG. 7 illustrates a range of positions of the spike in the image;
FIG. 8 (a) initial spike zone coarse positioning; FIG. 8 (b) coarse location of the candidate spike zone;
FIG. 9 spike fine positioning;
FIG. 10 is a view of the precise positioning of the nut center;
FIG. 11 shows the centering result for different types of fasteners; the fastener comprises (a) a normal fastener, (b) a surface oil stain fastener and (c) a background sundry fastener.
Detailed Description
The center of the spike is the same as that of the nut, and the circular characteristic of the spike is more stable, so that the center of the nut is positioned by detecting the center of the spike. The novel method can accurately detect the center of the spike by utilizing the profile and the regional texture information of the spike so as to realize the accurate positioning of the center of the nut.
The basic flow of the positioning algorithm of the invention is as follows.
The whole positioning algorithm mainly comprises two parts, namely fastener region positioning and accurate positioning of the nut center from the obtained fastener sub-image. A block diagram of the system is shown in fig. 3.
The fastener region positioning is used for extracting a fastener region from the acquired original image, and the nut center positioning is mainly completed by three steps of fastener sub-image contour acquisition, spike rough positioning and spike fine positioning. The images used were acquired in an actual rail environment.
The positioning algorithm of the present invention is implemented as follows.
The first step is as follows: fastener zone location
(1) Fastener coarse positioning
The rough positioning mainly utilizes the position relation among the fasteners, the steel rail and the sleeper, and the fastener area is preliminarily determined by detecting the steel rail and the sleeper area. And (3) carrying out binarization on the original image by using the characteristics that the rail area is over-exposed and the brightness is obviously higher than that of other areas, and setting a threshold value to be 0.75, wherein the largest connected area in the obtained binary image is the rail area.
The original image is divided into two parts by the detected rail region, sub-images of 100 pixels in width are respectively cut out from the two parts of images for binarization, and the threshold value is set to be 0.38. And performing AND operation on the two obtained binary images, wherein the maximum communication area in the binary images is the sleeper area. And the rough positioning of the fastener can be completed after the steel rail and the area are detected.
(2) Fastener fine positioning
The fine positioning is achieved by template matching. Selecting a standard fastener image as a fastener template image, respectively calculating a rough positioning fastener image and a fastener template image direction field, then sliding the fastener template image on the rough positioning image, simultaneously calculating the correlation of the characteristics of the direction field of the overlapped area of the rough positioning fastener image and the fastener template image, and obtaining the position with the maximum correlation value as the accurate position of the fastener, thus finishing the accurate positioning of the fastener and simultaneously extracting to obtain a fastener subimage.
Formula for calculating direction field
Correlation calculation formula
The second step is that: obtaining sub-image outline of fastener
And extracting the initial contour of the sub-image of the fastener by using a canny operator. Then according to the standard of the railway industry, the diameter of the upper end of the spike is 24mm, the length of the whole fastener area is 142mm, and the width of the whole fastener area is 106mm, so that the proportion of the spike in the length direction and the width direction in the whole fastener area is about 1/6 and 1/4. In the fastener sub-image, this ratio is also satisfied. According to this proportional relationship, the profile defined to satisfy any of the conditions described below is an extraneous profile that is significantly larger than the spike profile and needs to be removed.
Of the two conditions mentioned above, D L And D W Pixel values for each contour connected region along the length and width of the image are shown, respectively, and L and W represent the length and width of the fastener sub-image. The threshold is chosen to expand the strict scaling relation by a factor of 2. Meanwhile, 10 pixels are used as a threshold value, and the contour with the number of the pixel points smaller than the threshold value is defined as being obviously smaller than the contour of the spikeExtraneous contours are removed. The obtained outline image is shown in fig. 6 (b).
The third step: spike coarse positioning
After the fastener image profiles are obtained, a circle fitting is performed on each profile by using a least square method, and the obtained circles form an initial spike region set, as shown in fig. 8 (a).
Least square method circle fitting formula
(3)
According to the camera field of view calculation formula, once the camera-to-object distance is known, the camera coverage field of view can be obtained. According to imaging proportional relation
(D is the physical size of the spike diameter, W is the camera field of view width, W is the resulting image width, and D is the pixel width of the spike diameter), can be found { (R) } in the mean dimension of the spike diameter>
Accordingly, in the fastener image, the ratio of the pixel width of the spike diameter to the image width is known, and the ratio is 1/4 (in the figure, the diameter of the spike is 50 pixels, and the image width is 200 pixels). Then the pixel width of the spike diameter must be £ greater>
(δ is an allowable offset value, and is set to 5 pixels). Also in the fastener system, the spike must be located at its central position, so that the spike position can be limited to the central 1/3 width region of the image in the accurately extracted sub-image of the fastenerWithin the compartment, as shown in fig. 7.
Screening the obtained initial spike region set by the two limiting conditions, wherein the diameter is in the interval
Meanwhile, the spike region with the center pixel abscissa located in the width range of 1/3 of the middle of the image is a candidate spike region set. Thus, most of the non-spike areas can be removed, and the spike coarse positioning is completed.
The fourth step: spike fine positioning and nut center positioning
(1) Spike fine positioning
Because the spikes have the greatest height in the fastener system, the spike regions in the fastener image are generally brighter regions, appearing as more gray-scale values. Meanwhile, the gray level consistency of the spike region in the fastener image is good, a large amount of gray level mutation generally does not exist, and the information entropy is small. Combining the above analysis, the gray level mean and the information entropy are used as texture features to complete the fine positioning of the spike. And respectively calculating the gray average m and the information entropy e of the candidate spike area.
After the gray level mean value and the information entropy are obtained through calculation, the spike region cannot be screened out by directly comparing the gray level mean value with the information entropy due to the fact that the size of each candidate spike region is different. Therefore, the gray level mean and the information entropy of each candidate region are normalized.
Normalized calculation formula
And after the normalized gray level mean value and the normalized information entropy texture feature of each candidate region are obtained, selecting a final spike region by using the texture feature and adopting a voting method. The voting rules are as follows:
(1) sorting the gray level mean value and the information entropy of each candidate region, wherein the gray level mean value is sorted in a descending order, and the information entropy is sorted in an ascending order;
(2) giving the most votes (the most votes are the number of candidate regions) to the region with the maximum gray mean and the minimum information entropy, and giving the sequentially decreased votes to other regions according to the descending order of the gray mean and the ascending order of the information entropy;
(3) and adding the votes obtained from each candidate area to obtain the area with the most votes, namely the final spike area.
(2) Nut centering
After the final spike area is obtained through fine positioning, the pixel coordinate of the circle center of the spike area is also obtained, and the coordinate is also the central coordinate of the nut, so that the accurate positioning of the center of the nut of the fastener is completed.
Examples
The center positioning experiment was performed using 1500 fastener pictures taken at the line of the common stone (hedder-glottis). The acquisition of pictures comprises three types of fasteners: normal, surface greasy dirt and background sundries, 500 pictures of each type. For the nut center (namely the spike center) in the actual environment, the precise position of the nut center is difficult to know, so the invention takes the manually marked center as the standard. Meanwhile, in order to reduce errors caused by manual labeling, multiple times of labeling are adopted, and the magnitude of the Mean Absolute Error (MAE) value is used as a criterion for judging whether the center positioning is accurate or not. Here, MAE < =6pixels is considered to be accurate for centering.
The accuracy of the centering of the three types of fastener nuts (the accuracy is the ratio of the correct centering picture to the total number of pictures) is shown in table 1, and the results of partial centering are shown in fig. 11.
TABLE 1 nut centering accuracy for different types of fasteners
As can be seen from the data in table 1 and the positioning results in fig. 8 (a) and 8 (b), the algorithm can accurately realize the accurate positioning of the center of the fastener nut, and meanwhile, the algorithm still has good positioning accuracy and positioning effect for the fastener nuts in different environments.
Claims (8)
1. A railway fastener nut center positioning method is characterized by comprising the following implementation processes: detecting the center of the spike by using the contour and the regional texture information of the spike, and determining the center of the nut by using the center of the spike;
the concrete implementation steps comprise:
1) Acquiring a steel rail image, and binarizing the steel rail image, wherein the maximum communication area of the obtained binarized image is a steel rail area;
2) Dividing the steel rail image into two parts by using the steel rail area, respectively intercepting sub-images from the two parts of images to carry out binarization processing, and carrying out AND operation on the two obtained binarization images to obtain a total binarization image, wherein the maximum communication area in the total binarization image is the sleeper area;
3) Obtaining a rough positioning image of the fastener by utilizing the steel rail area and the sleeper area;
4) Selecting a standard fastener image as a fastener template image, calculating the correlation between the rough positioning image of the fastener and the fastener template image, and obtaining the position with the maximum correlation value, namely the accurate position of the fastener, thus finishing the accurate positioning of the fastener and simultaneously extracting and obtaining a fastener sub-image;
5) Extracting the initial contour of the sub-image of the fastener, and removing the irrelevant contour in the image to obtain a fastener contour image;
6) Performing circular fitting on the fastener outline image to obtain an initial spike region, and screening the initial spike region to obtain a candidate spike region;
7) Calculating a gray level mean value m and an information entropy e of a candidate spike region, normalizing the gray level mean value m and the information entropy e to obtain normalized gray level mean values and information entropy texture characteristics, and selecting a final spike region by using the gray level mean values and the information entropy;
8) And determining the center coordinates of the spike by using the final spike area, wherein the center coordinates are the center coordinates of the nut.
2. The method for centering a railway fastener nut as claimed in claim 1, wherein in step 4), the correlation R is calculated by the formula:
wherein, theta m,n And θ' m,n The pixel points (m, n) of the fastener coarse positioning image and the pixel points (m, n) in the fastener template image are respectively assigned with orientation field characteristic values.
3. The method of claim 2 wherein the direction field characteristic value is a value obtained by centering a railway fastener nut
I x And I y Pixel horizontal and vertical gradients.
4. The method of claim 1, wherein in step 5), the profile defined to satisfy any one of the following conditions is an extraneous profile substantially larger than the spike profile, i.e., the profile to be removed:
wherein D is L And D W Pixel values for each contour connected region along the length and width of the image are shown, respectively, and L and W represent the length and width of the fastener sub-image.
5. Railway fastener nut centering according to claim 1A bit method, wherein said initial set of spike zone diameters are in intervals
And the spike area with the circle center pixel abscissa located in the width range of 1/3 of the middle of the fastener outline image is a candidate spike area set.
6. The method for centering a railway fastener nut as claimed in claim 1, wherein the calculation formula of the mean value m of the gray scale and the entropy e of the information is as follows:
z i the gray value of the pixel point of the candidate spike region, p (z) i ) The number of occurrences of each gray value.
7. The method of claim 6, wherein the normalized mean gray scale value and entropy calculation formula is:
where c is the radius of the candidate spike zone.
8. The method for centering railway fastener nuts as claimed in any one of claims 1 to 7, wherein the step of selecting the final spike region using the mean grayscale value and entropy in step 7) comprises:
1) Sorting the texture features of the gray level mean value and the information entropy of each candidate spike area, wherein the gray level mean value is sorted in a descending order, and the information entropy is sorted in an ascending order;
2) Giving the most votes to the areas with the maximum gray mean value and the minimum information entropy, and giving the sequentially decreased votes to other areas according to the descending order of the gray mean value and the ascending order of the information entropy;
3) And adding the votes obtained from each candidate spike area to obtain the area with the most votes, namely the final spike area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910789963.7A CN110610516B (en) | 2019-08-26 | 2019-08-26 | Railway fastener nut center positioning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910789963.7A CN110610516B (en) | 2019-08-26 | 2019-08-26 | Railway fastener nut center positioning method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110610516A CN110610516A (en) | 2019-12-24 |
| CN110610516B true CN110610516B (en) | 2023-03-24 |
Family
ID=68890772
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910789963.7A Active CN110610516B (en) | 2019-08-26 | 2019-08-26 | Railway fastener nut center positioning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110610516B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112001908B (en) * | 2020-08-25 | 2021-03-09 | 哈尔滨市科佳通用机电股份有限公司 | Railway freight car sleeper beam hole carried foreign matter detection method |
| CN112102272A (en) * | 2020-09-03 | 2020-12-18 | 常州工学院 | Nut aperture detection method and device, storage medium and terminal equipment |
| CN112862889A (en) * | 2021-01-29 | 2021-05-28 | 珠海迪沃航空工程有限公司 | Recognition correction system, method and device based on image recognition |
| CN112962373A (en) * | 2021-02-20 | 2021-06-15 | 中国铁建重工集团股份有限公司 | Spike identification method based on line structured light and spike identification operation vehicle |
| TWI818837B (en) * | 2022-03-09 | 2023-10-11 | 四零四科技股份有限公司 | Object recognition method and object recognition device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719305A (en) * | 2016-01-25 | 2016-06-29 | 成都国铁电气设备有限公司 | Assembly falloff defect identification method and system of overhead contact system |
| CN108154506A (en) * | 2017-12-27 | 2018-06-12 | 合肥市雅视智能科技有限公司 | A kind of hex nut automatic testing method |
-
2019
- 2019-08-26 CN CN201910789963.7A patent/CN110610516B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719305A (en) * | 2016-01-25 | 2016-06-29 | 成都国铁电气设备有限公司 | Assembly falloff defect identification method and system of overhead contact system |
| CN108154506A (en) * | 2017-12-27 | 2018-06-12 | 合肥市雅视智能科技有限公司 | A kind of hex nut automatic testing method |
Non-Patent Citations (1)
| Title |
|---|
| 低压断路器智能测控系统中螺母的图像识别;耿璐等;《计算机工程与设计》;20111016(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110610516A (en) | 2019-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110610516B (en) | Railway fastener nut center positioning method | |
| CN108389179B (en) | Machine vision-based can cover surface defect detection method | |
| Wang et al. | Research on crack detection algorithm of the concrete bridge based on image processing | |
| CN107967681B (en) | Elevator compensation chain impact defect detection method based on machine vision | |
| CN101571419B (en) | Method adopting image segmentation for automatically testing LED indicator light of automobile instruments | |
| CN104966049B (en) | Lorry detection method based on image | |
| CN106934795A (en) | The automatic testing method and Forecasting Methodology of a kind of glue into concrete beam cracks | |
| CN104794491A (en) | Fuzzy clustering steel plate surface defect detection method based on pre classification | |
| CN104792792A (en) | Stepwise-refinement pavement crack detection method | |
| CN103839268A (en) | Method for detecting fissure on surface of subway tunnel | |
| CN111060442A (en) | Oil particle detection method based on image processing | |
| CN111539927B (en) | Detection method of automobile plastic assembly fastening buckle missing detection device | |
| CN102737252A (en) | Method for detecting faults caused by foreign body pollution between electrified railway insulator plates based on affine invariant moment | |
| CN106504231A (en) | Component defect detection method and system | |
| Abbas et al. | Automated pavement distress detection using image processing techniques | |
| CN111539436A (en) | Rail fastener positioning method based on straight template matching | |
| CN116012348A (en) | Maximum crack width measuring method suitable for pavement cracks | |
| CN110490163B (en) | Intelligent processing method and device for railway video data | |
| CN103927512B (en) | vehicle identification method | |
| CN116485719A (en) | Self-adaptive canny method for crack detection | |
| CN112085699A (en) | Pavement crack extraction method based on two-dimensional image | |
| CN108492306A (en) | A kind of X-type Angular Point Extracting Method based on image outline | |
| CN118501177A (en) | Appearance defect detection method and system for formed foil | |
| CN114926415A (en) | Steel rail surface detection method based on PCNN and deep learning | |
| CN114155226A (en) | Micro defect edge calculation 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 |