An anomaly detection method and device for robotic arm collecting PCB boards

By using image processing technology, combined with grayscale image processing and Laplacian variance calculation, the problem of abnormal placement caused by board warping during the PCB board receiving process by the robotic arm was solved. This enabled efficient and low-cost anomaly detection and alarm, reducing the risk of board scrapping.

CN118429867BActive Publication Date: 2026-11-17KUSN HULI MICROELECTRONICS
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410706709.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-03
Publication Date
2026-11-17
Estimated Expiration
2044-06-03

AI Technical Summary

Technical Problem

Existing technologies cannot effectively detect abnormal placement caused by board warping during the PCB board receiving process by robotic arms, resulting in board scratches and scrapping. Furthermore, existing grating sensor methods are costly and not accurate enough.

Method used

A method combining image sharpness evaluation and object counting is adopted. By acquiring video of the area where the robotic arm picks up PCB boards, grayscale image processing, local threshold segmentation, morphological denoising, contour detection, and Laplacian variance calculation are performed to identify abnormal placement.

Benefits of technology

It achieves efficient and low-cost anomaly detection, reduces the risk of board scrap, lowers production costs, and can provide real-time alarms to handle abnormal situations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118429867B_ABST
    Figure CN118429867B_ABST
Patent Text Reader

Abstract

This invention discloses an anomaly detection method and apparatus for a robotic arm collecting PCB boards. The method includes acquiring video of the PCB board collecting area of ​​the robotic arm and extracting images at a set frequency; extracting detection regions from the images and converting them into grayscale images; performing local thresholding using Otsu's algorithm for thresholding and morphological denoising on the grayscale images to obtain segmented and denoised images; performing contour detection and drawing contours on the segmented and denoised images to obtain images with drawn contours; traversing all the grains in the images with drawn contours according to a set grain rule and calculating the Laplacian variance of the images with drawn contours; and determining whether the automatic PCB board collecting by the robotic arm is abnormal based on the number of grains in the traversed images and the calculated Laplacian variance of the images. This invention can efficiently filter and analyze key image data, significantly improving the accuracy and reliability of automated visual inspection and quality control systems.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to an anomaly detection method and device for a robotic arm collecting PCB boards, belonging to the field of robotic arm technology. Background Technology

[0002] Robotic arms are the most widely used automated mechanical devices in the field of robotics, found in industrial manufacturing, medical treatment, entertainment services, military, semiconductor manufacturing, and other fields. They can directly replace some jobs that previously required fixed manual handling, making them a crucial component in realizing intelligent manufacturing.

[0003] PCB manufacturing, a complex process integrating mechanics, chemistry, optics, and other fields, has seen the widespread adoption of robotic arms for handling PCBs in many manufacturing steps in recent years. In some PCB handling processes, to prevent friction between boards, direct contact is not allowed. Therefore, the boards must be placed in insertion carts, meaning each board needs to be inserted into the gaps in the cart. This places high demands on the precision of the robotic arms. With the integration of machine vision technology, current robotic arms can meet the requirements of such handling processes. However, inevitably, due to board bending or warping, occasionally severely bent boards cannot be properly inserted into the gaps and fall onto the insertion cart. These automated handling processes are generally performed without operator supervision. If personnel are not immediately notified and the issue is not addressed, the robotic arm's continued placement can easily cause scratches and render the boards unusable. Currently, the industry generally adds multiple optical sensors to the insertion cart's handling area to detect such board-falling anomalies. However, this method is costly and cannot detect such anomalies 100% accurately. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide an anomaly detection method and device for robotic arms to collect PCB boards. By combining image clarity evaluation and object counting functions when the number of warning images is particularly limited and the types of boards are particularly numerous, it can efficiently filter and analyze key image data, and significantly improve the accuracy and reliability of automated visual inspection and quality control systems.

[0005] To achieve the above objectives, the present invention is implemented using the following technical solution: In a first aspect, the present invention provides an anomaly detection method for a robotic arm receiving PCB boards, comprising: Acquire video of the area where the robotic arm collects PCB boards and extract images at a set frequency; Extract the detection region from the image and convert it to a grayscale image; The Otsu algorithm thresholding and morphological denoising are applied to the grayscale image to obtain the segmented and denoised image. Perform contour detection and draw contours on the segmented and denoised images to obtain images with drawn contours; According to the set grain rule, iterate through all the grains in the image after the outline is drawn, and calculate the Laplace variance of the image after the outline is drawn. Based on the number of rice grains in the image obtained after traversal and the calculated Laplace variance of the image, determine whether the automatic board retraction of the PCB arm is abnormal.

[0006] Furthermore, the Otsu algorithm thresholding for local thresholding of the grayscale image includes: The Otsu algorithm, which uses local thresholding, is used to process grayscale images. The grayscale image is divided into multiple sub-regions, and the Otsu algorithm is applied independently to each sub-region to calculate the local optimal threshold. The local optimum threshold is calculated based on the intra-class variance. and inter-class variance The calculation formula is: ; ; Where t is a set threshold for the region, categories with values ​​less than t are marked as 0, and categories with values ​​greater than t are marked as 1. and Indicates the threshold The ratio of the number of pixels in the two separated categories satisfies ; and It is the variance of pixel values ​​in these two categories. , It is the average of the pixel values ​​in these two categories; Find t such that S b 2 Maximize, as the local optimum threshold of the region; Thresholding segmentation is used to separate the target region from the background region in a grayscale image, resulting in a segmented image.

[0007] Furthermore, the morphological denoising of the grayscale image includes: Generate a 9*9 cross-shaped structuring element, apply the structuring element to the grayscale image, and perform erosion and then dilation operations on the grayscale image. In the erosion operation, if all pixels of the structuring element correspond to pixels in the grayscale image that are 1, then the center pixel remains 1; otherwise, it becomes zero. During the dilation operation, if a pixel in the grayscale image corresponding to the structuring element has a value of 1, the center pixel is set to 1. The number of iterations for the erosion and dilation operations is set to 1. After the iterations are completed, the denoised image is obtained.

[0008] Furthermore, the step of performing contour detection and contour drawing on the segmented and denoised image to obtain an image with drawn contours includes: Contour detection is performed on the segmented and denoised image. When detecting contours, only the outermost contour is detected, or only the inflection point information of the contour is saved. After the contour detection is completed, the contour is drawn to obtain the image with the contour drawn.

[0009] Furthermore, the step of traversing the image after outlining according to the set grain rule to count all grains and calculating the Laplacian variance of the image after outlining includes: Detect the bounding area of ​​each contour in the image after contour drawing, and define an area greater than or equal to 2 pixels as 1 grain of rice; Iterate through the image after the outline is drawn and count the number of all rice grains, with the count being X. Calculate the Laplacian variance Y of the image using the grayscale value of each pixel in the image after contouring. The formula for calculating the Laplacian variance is: ; in, Representing an image Laplace transform, This indicates variance calculation.

[0010] Furthermore, the step of determining whether the PCB arm's automatic board retraction is abnormal based on the total number of rice grains in the traversed image and the calculated Laplace variance of the image includes: The criteria for judging image anomalies are defined as simultaneously satisfying the number of rice grains X>a and the Laplace variance Y>b, where a is the threshold for the number of rice grains and b is the threshold for the Laplace variance. The settings of a and b are based on the values ​​of the number of rice grains and the Laplace variance when the empty insertion rack is in normal operation, when the arm is normally placing the PCB board, and when the arm is abnormally tilting the PCB board. When the condition is normal, the output disconnect signal is activated, the robotic arm operates normally, and the alarm does not sound. When an abnormality is detected, a conduction signal is output, the robotic arm stops placing the plate and activates the alarm.

[0011] Secondly, the present invention provides an anomaly detection device for a robotic arm collecting PCB boards, comprising: The extraction module is used to extract a frame image from the video of the PCB board receiving area of ​​the robotic arm at a set frequency. The detection and conversion module is used to extract the detection area from the image and convert it into a grayscale image; The segmentation and denoising module is used to perform local thresholding using the Otsu algorithm for thresholding and morphological denoising on grayscale images to obtain segmented and denoised images. The contour drawing module is used to perform contour detection and draw contours on the segmented and denoised images to obtain images with drawn contours. The traversal and calculation module is used to traverse the image after the outline is drawn according to the set rice grain rules, count the number of all rice grains, and calculate the Laplace variance of the image after the outline is drawn. The judgment module is used to determine whether the automatic board retraction of the PCB arm is abnormal based on the number of all rice grains in the image obtained after traversal and the calculated Laplace variance of the image.

[0012] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the preceding methods.

[0013] Fourthly, the present invention provides a computer device, comprising: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of any of the preceding methods.

[0014] Fifthly, the present invention provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0015] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: This invention provides a method and device for detecting anomalies in PCB board receiving by a robotic arm. When an anomaly is detected during board receiving, the arm stops receiving the board and an alarm is triggered to notify personnel for handling. This method can be applied to many automated PCB board receiving processes using robotic arms. It can detect anomalies in real time and quickly trigger alarms and handle them, reducing the risk of board scrap and further reducing production costs. Attached Figure Description

[0016] Figure 1 This is a flowchart of an anomaly detection method for a robotic arm receiving PCB boards provided in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the detection scenario provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the morphologically denoised image provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the detected contour and the image after contour drawing provided in an embodiment of the present invention; Figure 5 This is an illustrative diagram illustrating the situation when the rack vehicle is empty, as provided in an embodiment of the present invention. Figure 6 This is a schematic diagram of a morphologically denoised image when the insertion vehicle is empty, provided in an embodiment of the present invention. Figure 7 This is a schematic diagram illustrating the normal plate placement process provided in an embodiment of the present invention; Figure 8 This is a schematic diagram illustrating the morphological noise reduction during normal plate placement according to an embodiment of the present invention; Figure 9 This is a schematic diagram illustrating an abnormal plate receiving situation provided in an embodiment of the present invention; Figure 10 This is a schematic diagram illustrating the morphological denoising after handling abnormalities in an embodiment of the present invention. in, Figures 2-10 The horizontal and vertical coordinates are both in pixels. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention. Example 1

[0018] This embodiment describes an anomaly detection method for a robotic arm collecting PCB boards, including: Extract one frame from the video of the PCB board receiving area of ​​the robotic arm at a set frequency. Extract the detection region from the image and convert it to a grayscale image; The Otsu algorithm thresholding and morphological denoising are applied to the grayscale image to obtain the segmented and denoised image. Perform contour detection and draw contours on the segmented and denoised images to obtain images with drawn contours; According to the set grain rule, iterate through all the grains in the image after the outline is drawn, and calculate the Laplace variance of the image after the outline is drawn. Based on the number of rice grains in the image obtained after traversal and the calculated Laplace variance of the image, determine whether the automatic board retraction of the PCB arm is abnormal.

[0019] The present invention will be further described below with reference to the accompanying drawings and specific embodiments: The machine vision inspection system used in this invention includes a USB camera, a Raspberry Pi, and a relay. This embodiment uses a 2-megapixel 1080P high-definition USB camera, a Raspberry Pi 4B (4th generation), and an RS232 relay. The overall hardware combination meets the inspection accuracy requirements.

[0020] like Figure 1As shown in the figure, the abnormal detection method for robotic arm receiving PCB boards provided in this embodiment involves the following steps in its application process: Extract a frame from the video of the PCB board receiving area of ​​the robotic arm at a set frequency, for example... One frame is extracted every 10 frames. This frequency setting is related to the camera's frame rate (FPS) and the arm's retraction frequency, and must ensure that the arm can be detected every time it places a PCB board onto the insertion cart. Figure 2 The extracted images illustrate the detection scenario of the present invention, in which a robotic arm collects PCB boards one by one and inserts them vertically into a mounting cart.

[0021] Extract the detection region from the image and convert it to grayscale. Extract the location of the jump-start vehicle in the image. The detection area was extracted by cropping the image based on the diagonal pixel coordinates of the vehicle in the image. After extraction, the image was converted from a color RGB image to a grayscale image using the formula Gray = 0.299*R + 0.587*G + 0.114*B.

[0022] The Otsu algorithm is used for local thresholding and morphological denoising of grayscale images. The steps are as follows: 1) Otsu's Local Thresholding Algorithm for Threshold Segmentation. This step aims to segment a grayscale image into target and background regions using a thresholding algorithm. The Otsu algorithm for local thresholding is used to process the image, separating the target and background regions through thresholding. Specifically, the image is divided into multiple sub-regions, and the Otsu algorithm is independently applied to each sub-region to calculate the locally optimal threshold.

[0023] Based on the formulas for calculating within-class variance and between-class variance: Within-class variance ; Between-class variance ;

[0024] t is a set threshold for the region. Categories with values ​​less than t are marked as 0, and categories with values ​​greater than t are marked as 1. and Indicates the threshold The ratio of the number of pixels in the two separated categories (satisfying) ). and It is the variance of pixel values ​​in these two categories. , This is the average pixel value between the two categories. Find t such that S b 2 Maximizing this value is equivalent to finding the optimal threshold t for that region.

[0025] As shown in Figure 2, after step 2 (extracting the detection region and converting grayscale) and step 3 (Otsu algorithm segmentation and morphological denoising with local thresholding), the generated image is shown in Figure 3.

[0026] 2) Morphological Denoising. Morphological denoising uses structuring elements to perform dilation and erosion operations on the image to remove noise. First, a 9x9 intersecting structuring element is generated, and then an opening operation is performed, i.e., erosion followed by dilation. The erosion operation applies the structuring element to the image; if all pixels of the structuring element and all pixels in the original image are 1, the center pixel remains 1; otherwise, it becomes zero. The dilation operation sets the center pixel to 1 if any pixel in the original image corresponding to the structuring element is 1. The iteration count for both erosion and dilation operations is set to 1.

[0027] Perform contour detection on the image and draw the contours. When detecting contours, the contour retrieval method is to only detect contours. The outermost contour is measured, and an approximate method is to save only the inflection point information of the contour, and then draw the contour after the contour detection is completed. Image 4 is the image after contour detection and contour drawing of image 3.

[0028] Define the rice grain rule and iterate through all the rice grains in the image, while simultaneously calculating the Laplacian of the image. Laplacian variance. Based on the bounding area of ​​each detected contour, a grain of rice with an area greater than or equal to 2 pixels is defined as one grain of rice. The total number of grains of rice in the image is counted as X. Simultaneously, the Laplacian variance Y of the image is calculated using the grayscale value of each pixel in the image. The formula for calculating the Laplacian variance is: ; in, Representing an image Laplace transform, This indicates variance calculation.

[0029] Define the criteria for determining whether an image is abnormal. The criteria for determining an image as abnormal are: The condition is that the number of rice grains X>a and the Laplace variance Y>b are satisfied. The settings of a and b can be determined based on the number of rice grains and the Laplace variance values ​​when the PCB board is normally placed on the rack, when the PCB board is normally placed on the rack, and when the PCB board is abnormally tilted on the rack. That is, the abnormal image and the normal image can be distinguished by the number of rice grains a and the Laplace variance b. Figures 5-10 The images shown are from different scenarios in this case. After processing through the previous five steps, the criteria for determining the number of rice grains (a) and the Laplace variance (b) are set to 15 and 3500, respectively.

[0030] Upon detecting an anomaly, the Raspberry Pi controls the robotic arm's movements and alarm via the GPIO interface. Method When the system is functioning normally, it outputs a disconnect signal, allowing the robotic arm to operate normally without triggering the alarm. When an anomaly is detected, it outputs a conduction signal, stopping the robotic arm from placing the board and activating the alarm. Simultaneously, the Raspberry Pi saves an image of the anomaly and transmits information such as the time of the anomaly to the data backend. Example 2

[0031] This embodiment provides an anomaly detection device for a robotic arm collecting PCB boards, including: The extraction module is used to extract a frame image from the video of the PCB board receiving area of ​​the robotic arm at a set frequency. The detection and conversion module is used to extract the detection area from the image and convert it into a grayscale image; The segmentation and denoising module is used to perform local thresholding using the Otsu algorithm for thresholding and morphological denoising on grayscale images to obtain segmented and denoised images. The contour drawing module is used to perform contour detection and draw contours on the segmented and denoised images to obtain images with drawn contours. The traversal and calculation module is used to traverse the image after the outline is drawn according to the set rice grain rules, count the number of all rice grains, and calculate the Laplace variance of the image after the outline is drawn. The judgment module is used to determine whether the automatic board retraction of the PCB arm is abnormal based on the number of all rice grains in the image obtained after traversal and the calculated Laplace variance of the image. Example 3

[0032] This embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above. Example 4

[0033] This embodiment provides a computer device, including: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of any of the preceding methods. Example 5

[0034] This embodiment provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0035] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

[0036] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0037] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0038] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0039] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure and not to limit its protection scope. Although this disclosure has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading this disclosure, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the pending claims.

Claims

1. An anomaly detection method for a robotic arm receiving PCB boards, characterized in that, include: Acquire video of the area where the robotic arm collects PCB boards and extract images at a set frequency; Extract the detection region from the image and convert it to a grayscale image; The Otsu algorithm thresholding and morphological denoising are applied to the grayscale image to obtain the segmented and denoised image. Perform contour detection and draw contours on the segmented and denoised images to obtain images with drawn contours; Based on the defined grain rules, iterate through the image after outlining to count all grains, and calculate the Laplacian variance of the image after outlining, including: Detect the bounding area of ​​each contour in the image after contour drawing, and define an area greater than or equal to 2 pixels as 1 grain of rice; Iterate through the image after the outline is drawn and count the number of all rice grains, with the count being X. Calculate the Laplacian variance Y of the image using the grayscale value of each pixel in the image after contouring. The formula for calculating the Laplacian variance is: in, Representing an image Laplace transform, Indicates variance calculation; Based on the total number of rice grains in the images obtained after traversal and the calculated Laplace variance of the images, determine whether the PCB arm's automatic board retraction is abnormal, including: The criteria for judging image anomalies are defined as simultaneously satisfying the number of rice grains X>a and the Laplace variance Y>b, where a is the threshold for the number of rice grains and b is the threshold for the Laplace variance. The settings of a and b are based on the values ​​of the number of rice grains and the Laplace variance when the empty insertion rack is in normal operation, when the arm is normally placing the PCB board, and when the arm is abnormally tilting the PCB board. When the condition is normal, the output disconnect signal is activated, the robotic arm operates normally, and the alarm does not sound. When an abnormality is detected, a conduction signal is output, the robotic arm stops placing the plate and activates the alarm.

2. The anomaly detection method for a robotic arm receiving PCB boards according to claim 1, characterized in that, The Otsu algorithm thresholding method for local thresholding of grayscale images includes: The Otsu algorithm, which uses local thresholding, is used to process grayscale images. The grayscale image is divided into multiple sub-regions, and the Otsu algorithm is applied independently to each sub-region to calculate the local optimal threshold. The local optimum threshold is calculated based on the intra-class variance. and inter-class variance The calculation formula is: ; ; Where t is a set threshold for the region, categories with values ​​less than t are marked as 0, and categories with values ​​greater than t are marked as 1. and Indicates the threshold The ratio of the number of pixels in the two separated categories satisfies ; and It is the variance of pixel values ​​in these two categories. , It is the average of the pixel values ​​in these two categories; Find t such that S b 2 Maximize, as the local optimum threshold of the region; Thresholding segmentation is used to separate the target region from the background region in a grayscale image, resulting in a segmented image.

3. The anomaly detection method for a robotic arm receiving PCB boards according to claim 1, characterized in that, The morphological denoising of the grayscale image includes: Generate a 9*9 cross-shaped structuring element, apply the structuring element to the grayscale image, and perform erosion and then dilation operations on the grayscale image. In the erosion operation, if all pixels of the structuring element correspond to pixels in the grayscale image that are 1, then the center pixel remains 1; otherwise, it becomes zero. During the dilation operation, if a pixel in the grayscale image corresponding to the structuring element has a value of 1, the center pixel is set to 1. The number of iterations for the erosion and dilation operations is set to 1. After the iterations are completed, the denoised image is obtained.

4. The anomaly detection method for a robotic arm receiving PCB boards according to claim 1, characterized in that, The step of performing contour detection and contour drawing on the segmented and denoised image to obtain the image with drawn contours includes: Contour detection is performed on the segmented and denoised image. When detecting contours, only the outermost contour is detected, or only the inflection point information of the contour is saved. After the contour detection is completed, the contour is drawn to obtain the image with the contour drawn.

5. An anomaly detection device for a robotic arm collecting PCB boards, used to implement the anomaly detection method for a robotic arm collecting PCB boards as described in any one of claims 1-4, characterized in that, include: The extraction module is used to extract a frame image from the video of the PCB board receiving area of ​​the robotic arm at a set frequency. The detection and conversion module is used to extract the detection area from the image and convert it into a grayscale image; The segmentation and denoising module is used to perform local thresholding using the Otsu algorithm for thresholding and morphological denoising on grayscale images to obtain segmented and denoised images. The contour drawing module is used to perform contour detection and draw contours on the segmented and denoised images to obtain images with drawn contours. The traversal and calculation module is used to traverse the image after the outline is drawn according to the set rice grain rules, count the number of all rice grains, and calculate the Laplace variance of the image after the outline is drawn. The judgment module is used to determine whether the automatic board retraction of the PCB arm is abnormal based on the number of all rice grains in the image obtained after traversal and the calculated Laplace variance of the image.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When executed by a processor, the program implements the steps of the method described in any one of claims 1-4.

7. A computer device, characterized in that, include: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of the method according to any one of claims 1-4.

8. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-4.

Citation Information

Patent Citations

  • Extracting method for micro-image cellula target of acerous red tide algae

    CN101777121A

  • Bill recognition method and device, electronic equipment and storage medium

    CN111461100A