Intelligent detection method and system for breakage defect of middle frame bracket of electronic product

Abstract

The invention relates to an intelligent detection method and system for a fracture defect of a middle frame bracket of an electronic product, the intelligent detection method and system for the fracture defect of the middle frame bracket of the electronic product are ingenious in design and accurate in positioning, a bracket area is segmented and folded, iteration is continuously carried out based on the width and height of an extracted picture, and the detection accuracy is improved. The folding algorithm is ended until an iteration termination condition is met, so that the area where the fracture defect is located is continuously reduced and refined, and the accuracy and the detection precision of subsequent rapid detection of the fracture defect are further improved; then target detection is carried out on the detection area, each sub-image is independently calculated to obtain a detection result, finally, the detection results obtained by all the sub-images are combined to obtain a total detection result, and finally, back calculation is carried out on the total detection result, so that the accuracy of the detection results is improved; according to the detection method, the health state of the support material and the specific position of the fracture defect are obtained, the detection precision is higher, the detection speed is higher, and therefore the good practical value is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent detection of appearance defects, and more specifically, to an intelligent detection method and system for fracture defects of a frame bracket of an electronic product. Background Art

[0002] In the complex process from raw materials to final bracket molding, due to various factors such as transportation, process, and human factors, the edges of the frame brackets of electronic products often have various defects such as fractures, dents, and scratches. The existence of these defects will affect the overall quality of the product and reduce the user experience. Traditional bracket quality inspection uses manual inspection. It is generally difficult to detect too small fractures manually. Therefore, there are problems such as easy to miss inspections, low inspection efficiency, high labor costs, and inconsistent inspection standards. It does not meet the high standards of product quality required by modern intelligent production;

[0003] In recent years, as machine vision is applied to various aspects of automated production, machine vision technology continues to mature. Existing detection technologies generally use traditional machine vision detection algorithms, mainly including scale analysis, color extraction, contrast analysis, etc. Figure 1-Figure 3 Shown are enlarged images of the left side fracture of the electronic equipment bracket, enlarged images of fracture 1 on the right side, and enlarged images of fracture 2 on the right side; however, these algorithms have high requirements on the objects being inspected and are highly dependent on them. They also have problems such as low detection efficiency, high false detection rate, and the need for targeted processing of different targets. They can detect more obvious fractures, but it is difficult to achieve good results for materials with slight cracks or too small fractures. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for intelligently detecting fracture defects of a frame bracket in an electronic product in view of the above-mentioned defects of the prior art.

[0005] On the one hand, the technical solution adopted by the present invention to solve its technical problem is: an intelligent detection method for the fracture defect of the frame bracket of an electronic product, wherein the intelligent detection method comprises the following steps:

[0006] The camera acquires the bracket image in real time, selects the appropriate bracket image as the detection template image, delineates the image defect generation area and extracts all the contours in the area, selects the required feature contours from them, and then creates multiple image detection templates based on multiple feature contours at different positions of the image;

[0007] Find the feature contour in the image to be detected, and match the image to be detected with the corresponding image detection template. If the matching score is high and the contour match is correct, the affine matrix of the detection image is calculated through the affine transformation algorithm, otherwise the detection area is redefined;

[0008] The folding algorithm is used to cut and fold the original material image N times; the trained fracture detection model is used to perform fracture detection on multiple sub-images after cutting and folding, and it is determined whether there is a fracture defect in the sub-image after cutting and folding; if there is, the location of the fracture defect is marked; if not, the material OK signal is output;

[0009] The image detection results are summarized and back-calculated, and all sub-images are reassembled into the original image according to the above-mentioned cutting process. The health status of the stent material and the specific location of the fracture defect can be quickly detected through the marking information in the spliced ​​image;

[0010] The intelligent detection method for the fracture defect of the frame bracket of the electronic product of the present invention, wherein the method for screening the characteristic contour comprises the following steps: controlling the sensitivity of the outer edge detection of the image contour by adjusting the filter coefficient and the upper and lower limits of the filter, highlighting the potential contours and thus extracting all the contours in the fracture area; further screening the required characteristic contours by adjusting the upper and lower limits of the contour length in the contour screening;

[0011] The intelligent detection method for the fracture defect of the frame bracket of an electronic product described in the present invention comprises the following steps: before delineating the image defect generation area, single-point positioning is used to select the characteristic contour of a certain position of the detection template image, and the image detection template is created based on this contour;

[0012] In the intelligent detection method for the fracture defect of the frame support of an electronic product described in the present invention, the folding algorithm logic is as follows:

[0013] The width and height of the material image are w and h respectively. If the width value w of the material image is greater than the height value h, the height value h is used as the folding basis, and the material image is cut into two equal parts at w / 2. The cut sub-images are folded along the width value w. The width and height of the folded new image are w / 2 and 2h respectively. If 2h≥w, the folding algorithm is exited. Otherwise, it is continuously iterated until 2h≥w.

[0014] The intelligent detection method for the fracture defect of the frame bracket of the electronic product described in the present invention, wherein, if there is a fracture defect in the cut and folded picture, the specific position of the fracture defect is marked with a frame, and a material NG signal is output at the same time;

[0015] The present invention provides an intelligent method for detecting fracture defects of a frame bracket in an electronic product, wherein the fracture detection method comprises inputting the cut and folded sub-image and the image of the material to be detected after affine transformation into a fracture defect detection model of a detection device for comparative analysis;

[0016] The intelligent detection method for the fracture defect of the frame bracket of an electronic product according to the present invention further comprises:

[0017] According to the detection results of all sub-images, the location of the material fracture defect and the material status are presented on the original image of the detection device, and the area size of the fracture defect, the defect location coordinates, the number of fracture defects and the detection time are displayed on the data display interface;

[0018] On the other hand, the present invention provides an intelligent detection system for frame bracket fracture defects in electronic products, which includes:

[0019] The scanning and extraction module is used to obtain the stent image in real time through the camera, select the appropriate stent image as the detection template image, delineate the image defect generation area and extract all the contours in the area, and select the required feature contours;

[0020] A creation module is used to create multiple image detection templates according to multiple feature contours at different positions of the image;

[0021] The contour matching module is used to find the characteristic contour in the image to be detected and match the image to be detected with the corresponding image detection template. If the matching score is high and the contour matching is correct, the affine matrix of the detection image is calculated through the affine transformation algorithm, otherwise the detection area is redefined;

[0022] The fracture detection module is used to cut and fold the original material image N times; the fracture detection is performed on the multiple sub-images after cutting and folding through the trained fracture detection model to determine whether the sub-images after cutting and folding have fracture defects;

[0023] Marking module, used to mark the specific location of the fracture defect and output the material NG signal;

[0024] The summary back calculation module is used to summarize the image detection results and reassemble all sub-images into the original image according to the above-mentioned cutting process. The health status of the stent material and the specific location of the fracture defect can be quickly detected through the marking information in the spliced ​​image;

[0025] The display module is used to present the detection results of all sub-images, including the location of the material fracture defect and the material status, on the original image of the detection device, and at the same time display the area size of the fracture defect, the defect location coordinates, the number of fracture defects and the detection time on the data display interface.

[0026] The beneficial effects of the present invention are as follows: the intelligent detection method and system for the fracture defect of the frame bracket of the electronic product are cleverly designed, and multiple image detection templates are created by extracting characteristic contours at different positions. After the image to be detected is collected, it is necessary to find the characteristic contour in the corresponding area of ​​the material to be detected according to the existing template, so as to perform template matching, and then perform affine transformation on the material to be detected to ensure that the material to be detected is in the detection center and accurately positioned. By segmenting and folding the bracket area, it is continuously iterated based on the width and height of the extracted image until the folding algorithm is terminated when the iteration termination condition is met, so as to continuously reduce and refine the area where the fracture defect is located, thereby improving the accuracy and detection accuracy of subsequent rapid detection of fracture defects; then target detection is performed on the detection area, and the detection result is calculated separately for each sub-image, and finally the detection results obtained by all the sub-images are merged to obtain the total detection result, and finally the total detection result is back-calculated to improve the accuracy of the detection result, and the health status of the bracket material and the specific location of the fracture defect are obtained. The above detection method has higher detection accuracy and faster detection speed, and therefore has good practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. The drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work:

[0028] Figure 1 is an enlarged view of the left side of the electronic device bracket being broken in the background art;

[0029] Figure 2 is an enlarged view of the right side fracture 1 of the electronic device bracket in the background technology;

[0030] Figure 3 is an enlarged view of the right side fracture 2 of the electronic device bracket in the background technology;

[0031] Figure 4 It is a flow chart of an intelligent detection method for a frame bracket fracture defect in an electronic product according to a first embodiment of the present invention;

[0032] Figure 5 is a partial diagram of the detection result of the first embodiment of the present invention, taking the fracture of the left side of the electronic device bracket as an example;

[0033] Figure 6 It is a schematic diagram of an intelligent detection system for fracture defects of a frame bracket in an electronic product according to a second embodiment of the present invention. DETAILED DESCRIPTION

[0034] The terms "first", "second", "third" and "fourth" etc. in the specification and claims of the present invention and the drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0035] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0036] "Multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the related objects are in an "or" relationship.

[0037] Moreover, the terms "up, down, front, back, left, right, upper end, lower end, longitudinal" and the like indicating directions are all based on the posture and position of the device or equipment described in this solution during normal use.

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the following will be described clearly and completely in combination with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are partial embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0039] 1. Embodiment 1

[0040] A preferred embodiment of the present invention provides an intelligent detection method for a frame support fracture defect in an electronic product, such as Figure 4 As shown, the intelligent detection method includes the following steps:

[0041] S10: acquiring a bracket image in real time through a camera, selecting a suitable bracket image as a detection template image, delineating an image defect generating area and extracting all contours in the area, screening out required feature contours, and then creating multiple image detection templates according to multiple feature contours at different positions of the image;

[0042] Among them, a suitable bracket image is a picture with a correct position and a clear image; the method for screening characteristic contours in the present invention includes the following steps: by adjusting the filter coefficient and the upper and lower limits of the filter to control the sensitivity of the outer edge detection of the image contour, the purpose of smoothing the image and reducing noise interference can be achieved, thereby highlighting the potential contour and extracting all the contours in the fracture area; by adjusting the upper and lower limits of the contour length in the contour screening, the required characteristic contours are further screened out.

[0043] Alternatively, an edge detection algorithm in the prior art (such as Sobel operator, Canny operator, etc.) may be used to detect edges in the image, and then complete contour lines may be extracted from the edge detection results.

[0044] Furthermore, in order to quickly find the fracture defects, before delineating the image defect generation area, the present invention uses single-point positioning to select the characteristic contour of any position in the detection template image, and creates an image detection template based on this contour. By performing block detection on the defective area, small fracture defects can be well detected.

[0045] After collecting the image to be tested, it is necessary to find the feature contour in the corresponding area of ​​the material to be tested based on the existing template, so as to perform template matching and affine transformation to ensure that the material to be tested is in the center of the test.

[0046] S20: Find the characteristic contour in the image to be detected, and match the image to be detected with the corresponding image detection template. If the matching score is high and the contour matching is correct, the affine matrix of the detection image is calculated through the affine transformation algorithm to ensure that the detection material is in the detection center position and the positioning is accurate, thereby improving the accuracy of subsequent fracture detection; otherwise, the detection area is redefined;

[0047] When extracting the middle frame bracket ROI area (region of interest), the bracket area is first segmented and folded as follows:

[0048] S30: using a folding algorithm to cut and fold the original material image N times; using a trained fracture detection model to perform fracture detection on multiple sub-images after cutting and folding, to determine whether there is a fracture defect in the sub-images after cutting and folding; if there is a fracture defect, marking the location of the fracture defect; if not, outputting a material OK signal;

[0049] The folding algorithm logic of the present invention is as follows:

[0050] The width and height of the material image are w and h respectively. If the width value w of the material image is greater than the height value h, the height value h is used as the folding basis, and the material image is cut into two equal parts at w / 2. The cut sub-images are folded along the width value w. The width and height of the new image after folding are w / 2 and 2h respectively. If 2h≥w, the folding algorithm is exited, otherwise it is continuously iterated until 2h≥w.

[0051] Based on the width and height of the extracted image, the folding algorithm is continuously iterated until the iteration termination condition is met, so as to continuously reduce and refine the area where the fracture defect is located, thereby improving the accuracy and detection precision of the subsequent rapid detection of fracture defects, and providing a better user experience.

[0052] By performing the above-mentioned targeted target detection on the detection area in each sub-image, the detection result can be calculated separately for each sub-image, such as Figure 5 As shown, a partial diagram of the detection results is shown, taking the fracture on the left side of the electronic equipment bracket as an example, which improves the detection accuracy and avoids excessive external interference factors that reduce the detection rate.

[0053] In the detection results, if there is a break defect in the cut and folded image, the specific location of the break defect is marked with a frame, and the material NG signal is output at the same time; if no break defect is detected, then the material OK signal is output if it does not exist. There is no need to manually mark the status of the bracket material (OK or NG), saving time and effort.

[0054] S40: Summarize and reverse calculate the image detection results, and re-join all sub-images into the original image according to the above-mentioned cutting process. The health status of the stent material and the specific location of the fracture defect can be quickly detected through the marking information in the jointed image; by summarizing all the detection results and reversely joining all the sub-images into the original image, not only can re-inspection be performed to improve the accuracy of the detection results, but also the size of the image can be accurately controlled.

[0055] S50: The location of the material fracture defect and the material state are presented on the original image of the detection device according to the detection results of all sub-images. At the same time, the area size of the fracture defect, the defect location coordinates, the number of fracture defects and the detection time are displayed on the data display interface. The detection results are intuitive and clear.

[0056] The intelligent detection method and system design of the fracture defect of the frame bracket of the electronic product are ingenious. By extracting the characteristic contours at different positions, multiple image detection templates are created. After the image to be detected is collected, it is necessary to find the characteristic contour in the corresponding area of ​​the material to be detected according to the existing template, so as to perform template matching, and then perform affine transformation on the material to be detected to ensure that the material to be detected is in the detection center and accurately positioned. By segmenting and folding the bracket area, it is continuously iterated based on the width and height of the extracted image until the folding algorithm is terminated when the iteration termination condition is met, so as to continuously reduce and refine the area where the fracture defect is located, thereby improving the accuracy and detection accuracy of subsequent rapid detection of fracture defects; then target detection is performed on the detection area, and the detection result is calculated for each sub-image separately, and finally the detection results obtained by all the sub-images are merged to obtain the total detection result, and finally the total detection result is back-calculated to improve the accuracy of the detection result, and the health status of the bracket material and the specific location of the fracture defect are obtained. The above detection method has higher detection accuracy and faster detection speed, so it has good practical value.

[0057] 2. Embodiment 2:

[0058] The present invention also provides an intelligent detection system for fracture defects of a frame bracket in an electronic product, which is applicable to the intelligent detection method for fracture defects of a frame bracket in an electronic product as described in the first embodiment; Figure 6 As shown, it includes the following modules:

[0059] The scanning and extraction module is used to obtain the stent image in real time through the camera, select the appropriate stent image as the detection template image, delineate the image defect generation area and extract all the contours in the area, and select the required feature contours;

[0060] A creation module is used to create multiple image detection templates according to multiple feature contours at different positions of the image;

[0061] The contour matching module is used to find the characteristic contour in the image to be detected and match the image to be detected with the corresponding image detection template. If the matching score is high and the contour matching is correct, the affine matrix of the detection image is calculated through the affine transformation algorithm, otherwise the detection area is redefined;

[0062] The fracture detection module is used to cut and fold the original material image N times; the fracture detection is performed on the multiple sub-images after cutting and folding through the trained fracture detection model to determine whether the sub-images after cutting and folding have fracture defects;

[0063] Marking module, used to mark the specific location of the fracture defect and output the material NG signal;

[0064] The summary back calculation module is used to summarize the image detection results and reassemble all sub-images into the original image according to the above-mentioned cutting process. The health status of the stent material and the specific location of the fracture defect can be quickly detected through the marking information in the spliced ​​image;

[0065] The display module is used to present the detection results of all sub-images, including the location of the material fracture defect and the material status, on the original image of the detection device, and at the same time display the area size of the fracture defect, the defect location coordinates, the number of fracture defects and the detection time on the data display interface.

[0066] The intelligent detection method and system design of the fracture defect of the frame bracket of the electronic product are ingenious. By extracting the characteristic contours at different positions, multiple image detection templates are created. After the image to be detected is collected, it is necessary to find the characteristic contour in the corresponding area of ​​the material to be detected according to the existing template, so as to perform template matching, and then perform affine transformation on the material to be detected to ensure that the material to be detected is in the detection center and accurately positioned. By segmenting and folding the bracket area, it is continuously iterated based on the width and height of the extracted image until the folding algorithm is terminated when the iteration termination condition is met, so as to continuously reduce and refine the area where the fracture defect is located, thereby improving the accuracy and detection accuracy of subsequent rapid detection of fracture defects; then target detection is performed on the detection area, and the detection result is calculated for each sub-image separately, and finally the detection results obtained by all the sub-images are merged to obtain the total detection result, and finally the total detection result is back-calculated to improve the accuracy of the detection result, and the health status of the bracket material and the specific location of the fracture defect are obtained. The above detection method has higher detection accuracy and faster detection speed, so it has good practical value.

[0067] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. An intelligent detection method for fracture defects of frame brackets in electronic products, characterized in that: The intelligent detection method comprises the following steps: The camera acquires the bracket image in real time, selects the appropriate bracket image as the detection template image, delineates the image defect generation area and extracts all the contours in the area, selects the required feature contours from them, and then creates multiple image detection templates based on multiple feature contours at different positions of the image; Find the feature contour in the image to be detected, and match the image to be detected with the corresponding image detection template. If the matching score is high and the contour match is correct, the affine matrix of the detection image is calculated through the affine transformation algorithm, otherwise the detection area is redefined; The folding algorithm is used to cut and fold the original material image N times; the trained fracture detection model is used to perform fracture detection on multiple sub-images after cutting and folding, and it is determined whether there is a fracture defect in the sub-image after cutting and folding; if there is, the location of the fracture defect is marked; if not, the material OK signal is output; The image detection results are summarized and back-calculated, and all sub-images are reassembled into the original image according to the above-mentioned cutting process. The health status of the stent material and the specific location of the fracture defect can be quickly detected through the marking information in the spliced ​​image.

2. The intelligent detection method for fracture defects of frame brackets in electronic products according to claim 1 is characterized in that: The method for screening characteristic contours comprises the following steps: controlling the sensitivity of detecting the outer edge of the image contour by adjusting the filter coefficient and the upper and lower limits of the filter, highlighting the potential contours and extracting all the contours in the fracture area; and further screening the required characteristic contours by adjusting the upper and lower limits of the contour length in the contour screening.

3. The intelligent detection method for fracture defects of frame brackets in electronic products according to claim 2 is characterized in that: Before delineating the area where image defects occur, single-point positioning is used to select the characteristic contour of a certain position in the detection template image, and the image detection template is created based on this contour.

4. The intelligent detection method for fracture defects of frame brackets in electronic products according to claim 1, characterized in that: The folding algorithm logic is as follows: The width and height of the material image are w and h respectively. If the width value w of the material image is greater than the height value h, the height value h is used as the folding basis, and the material image is cut into two equal parts at w / 2. The cut sub-images are folded along the width value w. The width and height of the new image after folding are w / 2 and 2h respectively. If 2w≥w, the folding algorithm is exited, otherwise it is continuously iterated until 2h≥w.

5. The intelligent detection method for fracture defects of the frame support of an electronic product according to any one of claims 1 to 4, characterized in that: If there is a break defect in the image after cutting and folding, the specific location of the break defect is marked with a frame, and the material NG signal is output at the same time.

6. The intelligent detection method for fracture defects of frame brackets in electronic products according to claim 5, characterized in that: The fracture detection method includes inputting the cut and folded sub-image and the image of the material to be detected after affine transformation into a fracture defect detection model of a detection device for comparative analysis.

7. The intelligent detection method for fracture defects of a frame support in an electronic product according to any one of claims 1 to 4 and 6, characterized in that: The detection method further comprises: According to the detection results of all sub-images, the location of the material fracture defect and the material state are presented on the original image of the detection device. At the same time, the area size of the fracture defect, the defect location coordinates, the number of fracture defects and the detection time are displayed on the data display interface.

8. An intelligent detection system for fracture defects of frame brackets in electronic products, characterized in that: include: The scanning and extraction module is used to obtain the stent image in real time through the camera, select the appropriate stent image as the detection template image, delineate the image defect generation area and extract all the contours in the area, and select the required feature contours; A creation module is used to create multiple image detection templates according to multiple feature contours at different positions of the image; The contour matching module is used to find the characteristic contour in the image to be detected and match the image to be detected with the corresponding image detection template. If the matching score is high and the contour matching is correct, the affine matrix of the detection image is calculated through the affine transformation algorithm, otherwise the detection area is redefined; The fracture detection module is used to cut and fold the original material image N times; the fracture detection is performed on the multiple sub-images after cutting and folding through the trained fracture detection model to determine whether the sub-images after cutting and folding have fracture defects; Marking module, used to mark the specific location of the fracture defect and output the material NG signal; The summary back calculation module is used to summarize the image detection results and reassemble all sub-images into the original image according to the above-mentioned cutting process. The health status of the stent material and the specific location of the fracture defect can be quickly detected through the marking information in the spliced ​​image; The display module is used to present the detection results of all sub-images, including the location of the material fracture defect and the material status, on the original image of the detection device, and at the same time display the area size of the fracture defect, the defect location coordinates, the number of fracture defects and the detection time on the data display interface.

Images