Method and device for detecting an evaporator

Abstract

The application provides a kind of evaporator detection method and device, the method includes by collecting image to be measured;Resolve the image to be measured, extract copper pipe area from the image to be measured;The pixel mean of the copper pipe area is calculated as the mean to be measured, the pixel variance of the copper pipe area is calculated as the variance to be measured;The variance to be measured and the preset copper pipe variance are compared, and the copper pipe knock detection result is obtained;The mean to be measured and the preset copper pipe mean are compared, and the copper pipe depth detection result is obtained.The design reduces the process of manual detection, reduces the input of human resources, saves detection time, avoids the low efficiency caused by manual fatigue, avoids the missed detection and misjudgment caused by manual visual error, and can also provide quantitative detection results, accurate evaluation of specific items and copper pipe assembly quality.

Description

Technical Field

[0001] This invention relates to the field of household appliance technology, and in particular to a method and apparatus for detecting evaporators. Background Technology

[0002] As one of the core components of an air conditioner, the evaporator plays a crucial role in the air conditioning cooling process. During the manufacturing process of the evaporator, inserting the U-shaped copper tubes into the fin assembly is a critical step, and its production quality and performance have a significant impact on the normal operation of the air conditioner.

[0003] U-shaped copper tubes and fins are not very hard, making them prone to impacts and deformation during evaporator assembly. Furthermore, the close proximity of the holes on the fins during assembly can lead to misinsertion or omissions of the copper tubes. The quality of evaporator assembly directly impacts the air conditioner's quality, performance, and brand reputation. The industry maintains strict requirements for evaporator assembly quality control.

[0004] Currently, evaporator assembly quality inspection relies heavily on manual visual inspection, which has several shortcomings. First, manual inspection requires significant manpower and time, and is prone to decreased efficiency due to fatigue or visual strain. Second, visual errors can occur during observation, leading to missed detections or misjudgments. Furthermore, traditional inspection methods cannot provide quantitative results when inspecting specific items, making it difficult to accurately assess the quality of specific items and copper tube assembly. Summary of the Invention

[0005] Therefore, it is necessary to provide a method and apparatus for detecting evaporators to address the aforementioned technical problems.

[0006] A method for detecting an evaporator, comprising: Acquire the image to be tested; Analyze the image to be tested and extract the copper pipe region from the image to be tested; The pixel mean of the copper tube region is calculated as the mean to be measured, and the pixel variance of the copper tube region is calculated as the variance to be measured. The test variance is compared with the preset copper tube variance to obtain the copper tube impact detection result; The measured mean value is compared with the preset copper tube mean value to obtain the copper tube depth detection result.

[0007] In one embodiment, the step of comparing the measured mean with the preset copper tube mean includes: Based on the copper pipe impact test results, the copper pipe areas whose impact test results are qualified are determined from several copper pipe areas as qualified impact pipe areas. The average value of the tested copper pipe in the qualified impacted pipe area is compared with the preset average value of the copper pipe to obtain the copper pipe depth detection result.

[0008] In one embodiment, after the step of acquiring the image to be tested, the following steps are included: The image to be tested is converted to grayscale to obtain a grayscale image; Perform Hough circle detection on the grayscale image to obtain several candidate circles; The diameter of each candidate circle is checked to see if it matches the preset screening diameter, and the result of copper tube insertion failure detection is obtained.

[0009] In one embodiment, the step of parsing the image to be tested and extracting the copper pipe region from the image to be tested includes: Based on a preset contour extraction algorithm, several contour rectangles are extracted from the image to be tested; Calculate the ratio of the width to the height of several outline rectangles as the scale to be measured; The test ratio is checked to see if it matches the preset sieve ratio. When the test ratio matches the preset sieve ratio, the area corresponding to the outline rectangle is determined to be the copper tube area.

[0010] In one embodiment, the method further includes: The center point of each copper tube region is detected, and the coordinates of the center point of each copper tube region are used as the coordinates to be measured. Based on a preset misinsertion detection algorithm, the coordinates to be tested are detected to obtain the copper tube misinsertion detection results.

[0011] In one embodiment, the preset misinsertion detection algorithm includes a first misinsertion detection algorithm, which includes: Based on the coordinates to be measured, the coordinates of the center point of the copper tube area in the first direction are obtained as the first coordinates to be measured; wherein, the first direction is parallel to the arrangement direction of each hole group, and each hole group is used to insert a copper tube. The difference between the first measured coordinate value and the reference coordinate value is obtained to obtain the distance difference between the center point of the copper tube area and the reference hole; wherein, the reference coordinate value is the coordinate value of the reference hole in the first direction, and the axis of the reference hole and the hole corresponding to the copper tube area is parallel to the first direction; Divide the distance difference by the preset interval difference to obtain the difference multiple; When the integer part of the difference multiple is even, the copper pipe misinsertion detection result in the first direction is determined to be no misinsertion; when the integer part of the difference multiple is odd, the copper pipe misinsertion detection result in the first direction is determined to be misinsertion.

[0012] In one embodiment, the preset misinsertion detection algorithm includes a second misinsertion detection algorithm, the second misinsertion detection algorithm including: Based on the coordinates to be measured, the coordinates of the center point of the copper tube area in the second direction are obtained as the second coordinates to be measured; wherein, the second direction is perpendicular to the arrangement direction of each hole group, and each hole group is used to insert a copper tube. The second coordinate value to be measured is compared with the preset second coordinate reference value to obtain the second direction detection result as the copper tube misinsertion detection result.

[0013] An evaporator detection device, comprising: The image acquisition module is used to acquire the image to be tested. The image analysis module is used to analyze the image to be tested and extract the copper pipe region from the image to be tested; The pixel calculation module is used to calculate the pixel mean of the copper tube region as the mean to be measured, and to calculate the pixel variance of the copper tube region as the variance to be measured. The impact detection module is used to compare the variance to be measured with the preset copper tube variance to obtain the copper tube impact detection result; The depth detection module is used to compare the measured mean value with the preset copper tube mean value to obtain the copper tube depth detection result.

[0014] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to perform the following steps: Acquire the image to be tested; Analyze the image to be tested and extract the copper pipe region from the image to be tested; The pixel mean of the copper tube region is calculated as the mean to be measured, and the pixel variance of the copper tube region is calculated as the variance to be measured. The test variance is compared with the preset copper tube variance to obtain the copper tube impact detection result; The measured mean value is compared with the preset copper tube mean value to obtain the copper tube depth detection result.

[0015] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor: Acquire the image to be tested; Analyze the image to be tested and extract the copper pipe region from the image to be tested; The pixel mean of the copper tube region is calculated as the mean to be measured, and the pixel variance of the copper tube region is calculated as the variance to be measured. The test variance is compared with the preset copper tube variance to obtain the copper tube impact detection result; The measured mean value is compared with the preset copper tube mean value to obtain the copper tube depth detection result.

[0016] The aforementioned evaporator detection method and apparatus involve: acquiring a test image; analyzing the test image to extract the copper tube region; calculating the pixel mean of the copper tube region as the test mean, and calculating the pixel variance of the copper tube region as the test variance; comparing the test variance with a preset copper tube variance to obtain the copper tube impact detection result; and comparing the test mean with a preset copper tube mean to obtain the copper tube depth detection result. This design reduces manual inspection steps, minimizes human resource investment, saves inspection time, avoids inefficiency due to human fatigue, avoids missed detections and misjudgments due to human visual errors, and provides quantitative inspection results to accurately assess specific items and copper tube assembly quality. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the detection method for the evaporator in one embodiment; Figure 2 This is a structural block diagram of the detection device for the evaporator in one embodiment; Figure 3 This is an internal structural diagram of a computer device in one embodiment; Figure 4 This is a schematic diagram of the evaporator in one embodiment; Figure 5 This is another flowchart illustrating the detection method for the evaporator in one embodiment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] Example 1 In this embodiment, as Figure 1 As shown, a method for detecting an evaporator is provided, which includes: Step 110: Acquire the image to be tested.

[0020] In this embodiment, the image to be tested is obtained by taking a picture of the evaporator to be tested, and the image to be tested contains copper tubes and fins.

[0021] Step 120: Analyze the image to be tested and extract the copper pipe region from the image to be tested.

[0022] In this embodiment, holes are formed on the fins for inserting copper tubes. Since the copper tubes are typically U-shaped, both ends of the U-shaped tube need to be inserted into the holes. Therefore, usually two holes form a hole group, and each hole group is used to insert one U-shaped copper tube. After the U-shaped copper tube is inserted into the hole, the image to be tested is taken from the side of the U-shaped copper tube facing away from the fins, and the bottom of the U-shaped copper tube is also photographed. The copper tube area in the image to be tested is detected using a contour extraction algorithm.

[0023] Step 130: Calculate the pixel mean of the copper tube region as the mean to be measured, and calculate the pixel variance of the copper tube region as the variance to be measured.

[0024] In this embodiment, the mean value to be measured is first calculated, and then the mean value to be measured is input into the formula for calculating the variance to be measured. The image values ​​of each pixel in the copper tube region are obtained, the total number of pixels contained in the copper tube region is calculated, and the pixel values ​​of all pixels in the copper tube region are summed to obtain the total image values ​​of the copper tube region. The ratio of the total image values ​​of the copper tube region to the total number of pixels is the mean value to be measured. The pixel value deviation between each pixel value in the copper tube region and the mean value to be measured is calculated, and the mean value of the summed pixel value deviations is then calculated to obtain the variance to be measured.

[0025] Step 140: Compare the variance to be measured with the preset copper pipe variance to obtain the copper pipe impact detection result.

[0026] In this embodiment, the preset copper tube variance is determined by photographing qualified evaporators, calculating the average and variance of the copper tube's pixel values ​​in the image, determining the preset copper tube variance based on the variance, and determining the preset copper tube mean based on the average. Since the color and brightness of the surface of the copper tube differ between the damaged and undamaged areas when it is bumped, and these differences are reflected in the pixel values ​​of the image being tested, the copper tube bump detection result can be determined based on the pixel values ​​of the copper tube area. The copper tube bump detection result includes "bump detection passed" and "bump detection failed." A passed bump detection indicates that the copper tube in the corresponding area has not been bumped, while a failed bump detection indicates that the copper tube in the corresponding area has been bumped.

[0027] Step 150: Compare the measured mean value with the preset copper tube mean value to obtain the copper tube depth detection result.

[0028] In this embodiment, the preset average value of the copper tube is obtained as in the previous step. Since the preset average value of the copper tube is calculated based on images of a normal or qualified evaporator, it can be used as a standard to measure the average value to be measured. If the average value to be measured is similar to the preset average value of the copper tube, it indicates that the insertion depth of the copper tube into the hole in the image to be measured is similar to the insertion depth under normal circumstances. The copper tube depth detection result includes depth detection qualified and depth detection unqualified.

[0029] The evaporator detection method in this embodiment involves: acquiring a test image; analyzing the test image to extract the copper tube region; calculating the pixel mean of the copper tube region as the test mean, and calculating the pixel variance of the copper tube region as the test variance; comparing the test variance with a preset copper tube variance to obtain the copper tube impact detection result; and comparing the test mean with a preset copper tube mean to obtain the copper tube depth detection result. This design reduces the number of manual inspection steps, reduces human resource investment, saves inspection time, avoids inefficiency due to human fatigue, avoids missed detections and misjudgments due to human visual errors, and can also provide quantitative inspection results to accurately assess specific items and copper tube assembly quality.

[0030] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0031] Example 2 This embodiment provides a method for detecting an evaporator, which includes: Step 1: Acquire the image to be tested.

[0032] In one embodiment, an image of the evaporator at the detection area is taken at a preset photographing position to obtain the image to be tested. The evaporators are arranged according to a preset placement rule. In this embodiment, the method of acquiring the image to be tested is standardized to reduce errors caused by parameter differences during acquisition. To facilitate the detection of each image to be tested, the evaporators to be tested are uniformly placed according to the preset placement rule, and images are taken at the same position on each evaporator. Further, the preset placement rule is that the side of the copper tube of the evaporator faces the photographing position, while the side of the copper tube faces away from the fins.

[0033] Step 2: Perform grayscale processing on the image to be tested to obtain a grayscale image.

[0034] In one embodiment, the images to be tested are converted to grayscale according to the acquisition order. In this embodiment, the images to be tested are processed sequentially according to the acquisition order. When grayscale images are confused, they can be sorted according to the acquisition order to ensure that the results of grayscale image detection correspond to the product.

[0035] Step 3: Perform Hough circle detection on the grayscale image to obtain several candidate circles; check whether the diameter of each candidate circle matches the preset screening diameter to obtain the copper tube missing insertion detection result.

[0036] In this embodiment, when a copper tube is inserted into a hole in the fin, the copper tube blocks the inserted hole, making it impossible to detect the corresponding hole in the grayscale image. For holes where no copper tube is inserted, the holes are recorded in the image to be tested during image acquisition. Even after grayscale processing, the holes are still recorded, and therefore can be detected using Hough circle detection. Furthermore, since defects may exist on the fin, these defects may be mistakenly detected as circles. Both the circles corresponding to defects and the circles corresponding to holes are candidate circles. Because the circles corresponding to defects and holes differ in diameter—that is, the diameters of candidate circles corresponding to different objects such as defects and holes are different—the detected circles can be filtered based on the diameter of the hole, selecting the circle corresponding to the hole from the candidate circles.

[0037] In one embodiment, the preset screening diameter is determined based on the diameter of the holes in a preset hole image. In this embodiment, in order to screen the detected circles, a preset hole image is obtained by taking a picture of the holes on the fins, and the diameter of the holes in the preset hole image is calculated to determine the preset screening diameter.

[0038] In one embodiment, the preset screening diameter is a numerical range, which includes the numerical value corresponding to the diameter of the hole in the preset hole image. In this embodiment, due to errors in the acquired image, the preset screening diameter is set as a numerical range to more accurately screen out the circles corresponding to the holes. When the diameter of the detected circle falls within this range, the hole is screened out, avoiding the inability to screen out the circles corresponding to the holes due to errors in photography or image processing, where the diameter of the circle corresponding to the hole differs from the diameter of the hole in the preset hole image. For example, if the numerical value corresponding to the diameter of the hole in the preset hole image is d, the numerical range of the preset screening diameter is (d-d1, d+d1), where d1 is a positive number. Within this range, the hole diameter d is the center of the range, and the range length is 2d1.

[0039] Step 4: Based on a preset contour extraction algorithm, extract several contour rectangles from the image to be tested; calculate the ratio of the width to the height of the contour rectangles as the test ratio; detect whether each test ratio matches the preset sieve ratio; when the test ratio matches the preset sieve ratio, determine that the area corresponding to the contour rectangle is the copper tube area.

[0040] In this embodiment, since the preset contour extraction algorithm extracts contours for both the hole and the bottom of the copper tube, it obtains the corresponding contour rectangles of the hole and the bottom of the copper tube in the image to be tested. However, the aspect ratios of the contour rectangles corresponding to the hole and the bottom of the copper tube are different. Therefore, the width and height of the extracted contour rectangles can be detected, and the ratio of height to width can be used as the ratio to be tested. Since the ratios to be tested for the contour rectangles corresponding to the hole and the bottom of the copper tube in the image are different, and the preset sieve ratio is determined based on the aspect ratio of the bottom of the copper tube, the contour rectangles can be screened using the preset sieve ratio. The contour rectangles that match the preset sieve ratio are the copper tube area, and the contour rectangles that do not match the preset sieve ratio are filtered out. The preset sieve ratio can be obtained by directly measuring the width and height of the bottom of the copper tube and comparing the measured width and height; or by taking a picture of a qualified evaporator to obtain an image of the qualified evaporator, detecting the width and height of the bottom of the copper tube from the image, and comparing the detected width and height.

[0041] In one embodiment, the step of extracting several contour rectangles from the image to be tested includes: extracting several contour rectangles from the grayscale image obtained after Hough circle detection, wherein the grayscale image obtained after Hough circle detection is the image after grayscale processing and Hough circle detection of the image to be tested. In this embodiment, contour rectangles are extracted from the image after detection in step three. The grayscale image of the image to be tested, i.e., the grayscale image, shows a large difference between the areas corresponding to the copper tube and the fins, which facilitates contour extraction.

[0042] In one embodiment, the width and height of the outline rectangle are obtained by detecting the outline rectangle using the `rectangle` function. In this embodiment, the outline rectangle is processed to obtain the width `w` and height `h` information of the circumscribed rectangle.

[0043] Step 5: Detect the center point of each copper tube region and obtain the coordinates of the center point of each copper tube region as the coordinates to be measured; based on the preset misinsertion detection algorithm, detect each of the coordinates to be measured and obtain the copper tube misinsertion detection result.

[0044] In this embodiment, the center point of the copper tube region can be obtained during the detection of the outline rectangle in step four. The center point of the copper tube region is the intersection of the diagonals of the corresponding outline rectangle. The coordinates of the center point of the copper tube region can be obtained by establishing a coordinate system for the image under test. For example, taking the corner point of the lower left corner of the fin as the origin of the coordinate system, the distance of the center point of the copper tube region from the origin of the coordinate system in different directions is detected to obtain the coordinates of the center point of the copper tube region, which are the coordinates to be measured. The preset misinsertion detection algorithm is determined based on the layout of the hole group on the fin. The position of the fin group is the qualified insertion method of the copper tube region. When the copper tube region is detected in the position of the hole group, it is determined that the copper tube has not been misinserted; when the copper tube is detected not in the position of the hole group, it is determined that the copper tube has been misinserted.

[0045] In one embodiment, the evaporator has a plurality of holes arranged in a predetermined number of rows and columns. The row spacing between adjacent holes is the same, and the column spacing between adjacent holes is the same. Two holes form a hole group, and each hole group is used to insert a copper tube. For example, the predetermined number of rows is two, and the predetermined number of columns is fourteen.

[0046] Furthermore, the hole groups are arranged along the axis of the row, with the direction of the row as the first direction, and the axes of the two holes in the hole group are parallel to the first direction. In this embodiment, the hole groups are arranged in rows, that is, as shown below. Figure 4 The horizontal arrangement in the middle.

[0047] In one embodiment, the preset misinsertion detection algorithm includes a first misinsertion detection algorithm, which includes: Based on the coordinates to be measured, the coordinates of the center point of the copper tube area in the first direction are obtained as the first coordinates to be measured; wherein, the first direction is parallel to the arrangement direction of each hole group, and each hole group is used to insert a copper tube. The difference between the first measured coordinate value and the reference coordinate value is obtained to obtain the distance difference between the center point of the copper tube area and the reference hole; wherein, the reference coordinate value is the coordinate value of the reference hole in the first direction, and the axis of the reference hole and the hole corresponding to the copper tube area is parallel to the first direction; Divide the distance difference by the preset interval difference to obtain the difference multiple; When the integer part of the difference multiple is even, the copper pipe misinsertion detection result in the first direction is determined to be no misinsertion; when the integer part of the difference multiple is odd, the copper pipe misinsertion detection result in the first direction is determined to be misinsertion.

[0048] In this embodiment, the preset interval difference is the row spacing between two adjacent holes, and also the row spacing between two adjacent groups of holes. See [link / reference]. Figure 4Since the two holes into which the copper tubes are inserted are set along the first direction, i.e., the row spacing between each hole is the same. For example, the row spacing between two adjacent holes, i.e., the preset interval difference, should be p. The center points of two adjacent copper tube areas should differ by 2p, which is an even multiple of the preset interval difference. Taking a hole in the first column as a reference hole, the distance from the center point of each copper tube area after the copper tube area in the first column to the reference hole in the first direction, i.e., the distance difference, is divided by the preset interval difference. The quotient obtained is the difference multiple. Given the arrangement of the holes and hole groups in the first direction, when the integer part of the difference multiple is even, it means that the copper tube corresponding to the difference multiple has not been misinserted and its insertion position in the first direction is correct. In this embodiment, the parity of the difference multiple needs to be checked for each copper tube region. When the integer part of the difference multiple corresponding to the copper tube region is even, the position of the copper tube corresponding to that copper tube region in the first direction of the fin is correct, and the copper tube has not been misinserted in the first direction. When the integer part of the difference multiple corresponding to the copper tube region is odd, the position of the copper tube corresponding to that copper tube region in the first direction of the fin is incorrect, possibly because it has been moved forward or backward by a preset interval difference in the first direction, and the copper tube has been misinserted in the first direction.

[0049] In one embodiment, the preset misinsertion detection algorithm includes a second misinsertion detection algorithm, the second misinsertion detection algorithm including: Based on the coordinates to be measured, the coordinates of the center point of the copper tube area in the second direction are obtained as the second coordinates to be measured; wherein, the second direction is perpendicular to the arrangement direction of each hole group, and each hole group is used to insert a copper tube. The second coordinate value to be measured is compared with the preset second coordinate reference value to obtain the second direction detection result as the copper tube misinsertion detection result.

[0050] In this embodiment, since the column spacing between two adjacent holes is equal, the layout of each hole in the column direction (i.e., the second direction) can be used to detect whether the copper tube is misinserted in the second direction. When the copper tube is inserted longitudinally, the center point of the copper tube area is located between two rows of holes, so it can be detected whether the center point of the copper tube area is located between two rows of holes to determine whether the copper tube is misinserted in the second direction. The preset second coordinate reference value is determined based on the area between two adjacent rows of holes, and the coordinate value of the area between two adjacent rows of holes in the second direction can be the preset second coordinate reference value. When the center point of the copper tube area falls within the range of the preset second coordinate reference value, it indicates that the copper tube corresponding to the copper tube area is misinserted in the second direction.

[0051] Further, the step of comparing the second measured coordinate value with a preset second coordinate reference value to obtain a second direction detection result as the copper pipe misinsertion detection result includes: comparing the second measured coordinate value with a preset second coordinate reference value; when the second measured coordinate value is within the range of the preset second coordinate reference value, determining that the second direction detection result indicates that the copper pipe has been misinserted; when the second measured coordinate value is outside the range of the preset second coordinate reference value, determining that the second direction detection result indicates that the copper pipe has not been misinserted.

[0052] Furthermore, when both the first and second direction test results indicate that the copper tube has not been misinserted, the copper tube misinsertion test result is qualified; when the first and / or second direction test results indicate that the copper tube has been misinserted, the copper tube misinsertion test result is unqualified.

[0053] Step 6: Calculate the pixel mean of the copper tube region as the mean to be measured, and calculate the pixel variance of the copper tube region as the variance to be measured.

[0054] In this embodiment, in order to perform the collision detection in step seven and the depth detection in step eight, it is necessary to first calculate the pixel mean (i.e., the measured mean) and pixel variance (i.e., the measured variance) of each copper tube region. Each copper tube region has a measured mean and a measured variance. Collision detection and depth detection need to be performed on each copper tube region to obtain the collision detection results and depth detection results of each copper tube region. Based on the collision detection results and depth detection results of each copper tube region, it can be determined that the corresponding copper tube has a collision problem or the insertion depth does not meet the requirements.

[0055] Step 7: Compare the measured variance with the preset copper pipe variance to obtain the copper pipe impact detection result.

[0056] In this embodiment, the preset copper tube variance is obtained by photographing the evaporator with undamaged copper tubes and calculating the preset copper tube variance of the copper tube area. When there are multiple evaporators with undamaged copper tubes, each evaporator is photographed separately, and the variance of the copper tube area is calculated separately. The mean of the variances of all copper tube areas is used as the preset copper tube variance. Alternatively, the preset copper tube variance can also be the maximum or minimum value of the variance of the copper tube area for each evaporator. It is important to understand that when a copper tube is damaged, the color and gloss of the damaged area differ from the normal area. This difference in color and gloss can be used to determine whether a damage has occurred. A larger measured variance for the copper tube area indicates a greater pixel difference in the copper tube area, which may be caused by a damage incident.

[0057] Step 8: Based on the copper pipe impact test results, determine the copper pipe areas whose impact test results are qualified from several copper pipe areas as qualified impact pipe areas; compare the test mean of the qualified impact pipe areas with the preset copper pipe mean to obtain the copper pipe depth test results.

[0058] In this embodiment, the copper tube depth detection is performed only after the copper tube impact detection result is qualified. This is because when an impact occurs, the pixels in the copper tube area change, thus affecting the measured mean value. The larger the error between the calculated measured mean value and the actual pixel mean value of the copper tube area, the greater the impact on the copper tube depth detection result. When the difference between the measured mean value and the preset copper tube mean value is large, it may be caused by the copper tube impact or by the copper tube insertion depth not meeting the requirements. Therefore, the depth detection is set after the impact detection. When the impact detection is qualified, the depth detection is performed to eliminate the possibility that the difference between the measured mean value and the preset copper tube mean value is caused by the copper tube impact. When the difference between the measured mean value and the preset copper tube mean value is found, it can be determined that the difference is caused by the copper tube insertion depth not meeting the requirements.

[0059] Further, the step of comparing the test mean of the qualified impact test area with the preset copper pipe mean to obtain the copper pipe depth detection result includes: comparing the test mean of the qualified impact test area with the preset copper pipe mean; when the test mean is different from the preset copper pipe mean, the copper pipe depth detection result is unqualified; when the test mean is the same as the preset copper pipe mean, the copper pipe depth detection result is qualified.

[0060] Furthermore, when the measured average value is greater than the preset average value of the copper tube, it is determined that the copper tube insertion depth has not reached the preset depth; when the measured average value is less than the preset average value of the copper tube, it is determined that the copper tube insertion depth has exceeded the preset depth, wherein the preset depth is the depth required for the copper tube to be inserted into the hole.

[0061] Example 3 In this embodiment, see Figure 5 This paper provides a method for detecting evaporators, which includes: an image acquisition step, a missing insertion detection step, a misinsertion detection step, a collision detection step, and a depth detection step. After these detection steps, the analysis results are returned, indicating whether the evaporator is qualified (OK) or unqualified (NG).

[0062] The specific steps are as follows: S1 Image Acquisition.

[0063] For the evaporator that needs to be inspected, take a side view of it, such as... Figure 4 The image shown is a schematic diagram of the evaporator side. This embodiment uses a 14*2 copper hole as an example, but it is not limited to this model. The captured image is then processed to obtain the desired image using grayscale conversion.

[0064] S2 copper pipe installation quality defect detection.

[0065] S21 Copper Pipe Misalignment Detection.

[0066] Missing copper tubes refer to instances where copper tubes are not inserted into the evaporator. In actual assembly, this omission creates a round hole, which can be detected by inspecting the image for these holes. A specific example is: based on... Figure 4 The side view of the evaporation device was photographed and then converted to grayscale. Next, Hough circle detection and edge detection were performed. Based on the reference value of the copper tube diameter d, a circular filter was applied to the image after the previous step. Figure 4 The evaporator fins have two rows of 14*2 holes. A U-shaped copper tube needs to be inserted into these holes. After inserting the U-shaped copper tube into the fins, the holes will no longer be visible in the side view. If circular holes are still detected after filtration, such as... Figure 4 As shown in D, there is a single round hole, which indicates that there is a problem with the copper tube not being inserted.

[0067] In the process of Hough circle detection, the first step is edge detection, which is a part of Hough circle detection. Hough circle detection obtains the center coordinates and radius of the circle. The purpose of Hough circle detection is to identify possible "circles". Edge detection is a part of Hough circle detection. At the production station, based on the images taken by the camera, we select some qualified circular holes. Then, we can write a program to process the images, calculate the diameter of these qualified circular holes, and take the average value, which is our reference value d. After determining the diameter d, we take a small value, such as 0.1, and add or subtract from the diameter, for example, (d-0.1, d+0.1), using this range to filter the diameter. Its function is to filter out "copper hole circles" because some circles obtained in the previous steps are not copper hole circles, so we are performing a filtering operation here. In circular filtration, after determining the diameter d, we take a small value, such as 0.1, and add or subtract from the diameter, for example, (d-0.1, d+0.1). We use this range to filter the diameter. Its purpose is to screen out the truly "copper hole circles," because some circles obtained in the previous steps are not copper hole circles. For example, defects on the fins might be misidentified as very small or very large circles. Therefore, we perform a filtering operation here. After circular hole filtration, what remains are the actual corresponding circular holes on the material, such as... Figure 4 These are the two rows of round holes that remain.

[0068] S22 Copper Pipe Misinsertion Detection.

[0069] Misinsertion of copper pipes still occurs during installation. Correct assembly requires inserting the copper pipe into two horizontal holes, ensuring all holes are fully inserted. Specifically, the center of the assembled conduit must align with the line connecting the horizontal holes, and the center point of the conduit must satisfy the parity requirement of the hole spacing. A specific example is as follows: by Figure 4 Establish an indicator system in the upper left corner, such as Figure 4 Regions C and E in the image represent two cases of misinterpretation. To detect misinterpretation in region C, a contour extraction operation is first performed on the grayscale image, and then the bounding rectangle of this contour is extracted, as shown below. Figure 4 As shown by the dashed line in region C, the width w and height h of the bounding rectangle are obtained. The presence of circular copper holes will also result in a bounding rectangle, as shown below. Figure 4 The dashed line represents region D. Region C is incorrectly inserted because the copper pipe is inserted vertically; it should be inserted into the two holes above columns 5 and 6. Region E is incorrectly inserted because the copper pipe is inserted one hole backward; it should be inserted into the two holes above columns 7 and 8. The grayscale image is the result of grayscale conversion and Hough circle detection. The `cv.rectangle()` algorithm from the OpenCV library can directly process the contours to obtain the `w` and `h` information of the bounding rectangle. Processing all contours in the above-processed image will yield the bounding rectangles of all contours. These bounding rectangles will be filtered later.

[0070] In the side view, the width w and height h of the copper tube always maintain a certain proportion. The true proportion threshold is measured and denoted as K. Using the proportion threshold K, the dashed lines in region C and region D can be distinguished, obtaining the side profiles of all copper tubes. Misalignment is divided into longitudinal misalignment and lateral misalignment. The true proportion threshold K can be directly measured with a ruler, as it is used as a threshold and does not need to be extremely precise in this scenario. The proportion threshold is both the lateral and longitudinal proportion threshold K of the copper tube. In our fin scenario, we insert a U-shaped copper tube into the fin, both laterally and longitudinally. The fact that the U-shaped copper tube is inserted longitudinally indicates that the lateral and longitudinal situations are the same; the U-shaped copper tube is the same, and the circumscribed rectangle of the U-shaped copper tube is the same. Therefore, the threshold K is also the same. Thus, this threshold is both lateral and longitudinal. Figure 4 To determine whether it's type C or type D, we use the center point of region C as the reference. Figure 4The center point can be obtained when finding the circumscribed rectangle. If the center point of region C is near the 1 / 2 mark in the vertical direction, we determine it to be case C. The center of region D is clearly not at the 1 / 2 mark in the vertical direction, so it can be distinguished from case C. Using a proportional threshold K, we can distinguish the dashed parts of region C from those of region D. First, we use the K value to differentiate between regions C, D, and E, identifying regions C and E. Second, for regions C and E, we further distinguish them based on whether their center points are at the 1 / 2 mark.

[0071] Vertical misalignment: For region C, the center point (x_c, y_c) of region C can be found using the circumscribed circle. If y_c = 0.5H, or y_c is approximately equal to 0.5H, then this region belongs to the situation shown in region C, which is a misalignment of the copper tube. In the previous step of calculating the circumscribed rectangle, we will obtain information such as w, h, and the center point.

[0072] Lateral misplacement: For region E, the center point (x_e, y_e) of region E can be obtained similarly. Let n = (x_e - x_a) / p. If the integer part of n is odd, it indicates that it belongs to region E and is a misplacement case. When the integer part of n is even, it indicates that the copper pipe is in the normal position.

[0073] S23 copper pipe impact test.

[0074] When assembled copper tubes are bumped or knocked, the metal surface will appear uneven in pixels when photographed. In the steps above, the mean and variance of pixels in the bottom side area of ​​the selected copper tubes are calculated. When the variance S of the pixel values ​​is greater than the standard value S_0, it indicates that the pixel values ​​in that area are unevenly distributed, indicating a bump or knocking incident. Figure 4 The copper tube in the F region is the normal area, while the opposite is the non-F region. Surface deformation caused by impact results in uneven pixel value distribution, with a significant difference between the pixel value variance and the standard value. The bottom side area of ​​the copper tube is... Figure 4 The five regions in the image represent the bottom of the copper tube, specifically the bottom of the U-shaped tube, captured by the camera. The side region of the copper tube bottom is the copper tube area in the image after grayscale processing, Hough circle detection, and edge detection. By aligning the actual material with the captured image, we identify those areas that are intact and without damage. We then find more of these areas and calculate the average value of these areas to obtain the standard value S_0.

[0075] The S23 copper pipe impact detection also requires a mean value, which is necessary in the variance formula. That is, when calculating the variance to determine if an impact has occurred, the mean value is used, and this mean value is the average value of the copper pipe in its normal state. Using this mean value is equivalent to comparing the impacted condition with the normal state of the copper pipe, avoiding the possibility that the impact itself is very uniform. This variance not only measures the degree of impact but also prevents the situation where the impact is uniform and therefore undetectable. Minor impacts are also treated as anomalies in air conditioning components. If the solution needs improvement, for example, if a minor impact is still detected, the detection algorithm can be corrected based on actual results, such as by statistically analyzing false detection data.

[0076] S24 copper pipe depth inspection.

[0077] Based on the above steps, the mean and variance of pixels in the bottom side area of ​​the selected copper tubes are calculated. A large deviation between the calculated mean pixel value M and the standard value M_0 indicates a large deviation between the insertion depth of the U-shaped copper tube and the standard deviation. Here, the standard is the insertion depth corresponding to normal insertion of the copper tube. Because when the insertion depth of the copper tube is inconsistent, the shooting distance between the copper tube and the camera differs from normal shooting, resulting in differences in the pixel values ​​displayed in the captured image. Figure 4 As shown in area B, the average pixel value deviates significantly from that of normal parts, and the copper tube insertion depth is inconsistent with the acceptable level, indicating an anomaly. The standard value M_0 is used for comparison with the mean value M. The standard mean value is the average pixel value corresponding to the normal copper tube insertion depth. The standard value M_0 is obtained by aligning the actual material with the photographed image, finding copper tube sections with normal depths, and then finding several more (multiple materials and multiple normal copper tube sections), aligning them, and calculating the average value. The standard value is derived from sampling and calculation of a certain number of normal parts. The issues of excessively deep or shallow insertion are essentially a matter of comparing the mean value with the standard value. Specifically, when the insertion is too deep, the mean value is smaller than the standard value; when it is too shallow, the mean value is larger than the standard value.

[0078] In this embodiment, steps S24 and S25 are sequential. After completing step S24, it indicates there are no more collision defects. Therefore, in step S25, the default variance is small, and no further explanation is given. Steps S24 and S25 are sequential. Depth measurement is only meaningful when there are no collision defects (small variance).

[0079] S3 Results Analysis.

[0080] Based on the detection results of missing insertion, incorrect insertion, insertion depth, and bumps, combined with preset standards and thresholds, a comprehensive analysis and judgment are performed. Based on the detection results, corresponding reports or marks are generated for recording and guiding subsequent processing.

[0081] In this embodiment, the image processing-based air conditioner evaporator assembly quality inspection method addresses several issues. For the problem of missing copper pipe insertion, it utilizes the Hough circle detection method, with the diameter of the copper pipe as the screening criterion, based on the fact that missing insertion will create a circular hole in actual assembly. For the problem of incorrect insertion, it establishes a Cartesian coordinate system for the assembled image, based on the requirement that the center of the assembled conduit should be aligned with the line connecting the horizontal circular holes. The center point of the copper pipe's side view is found using the circumscribed rectangle method, and incorrect insertion is determined based on the parity of the center point's position and the hole spacing. For depth and impact detection, the depth of copper pipe insertion causes pixel differences in the image, and uneven reflection from impacts leads to uneven pixel distribution. For the bottom side view of the U-shaped copper pipe, the average and variance of pixels in that area are calculated. If the pixel variance is small and the average differs significantly from the standard value, a depth problem exists; if the pixel variance is large compared to the standard value, an impact is present. Based on the detection of missing, incorrect, insertion depth, and impact issues, the air conditioner evaporator assembly quality inspection is achieved.

[0082] The order of collision detection and depth detection cannot be changed because when detecting inconsistent depths, the comparison between the pixel mean and the standard mean is considered. If it cannot be guaranteed that the copper tube is in a collision-free state, the pixel impact caused by the collision and the impact of inconsistent depth will overlap, making it impossible to determine whether the factor affecting the pixel mean is due to the collision or the inconsistent depth, thus making it impossible to distinguish the true influencing factor. Therefore, depth detection must be performed after collision detection. However, the order of incorrect insertion, missing insertion, and collision detection can be changed, as long as the order of collision detection and depth detection is maintained.

[0083] In this embodiment, an image processing-based method detects issues such as missing, incorrect, inadequately inserted, or damaged copper tubes in the evaporator by analyzing pixel features and grayscale detection of the target image, as well as utilizing the distribution of pixels and feature points. This avoids false positives and missed positives caused by prolonged human visual inspection, improving detection performance and production efficiency. Compared to traditional manual visual inspection, the evaporator inspection method in this embodiment achieves automated inspection, significantly improving efficiency and accuracy. The image processing algorithm can quickly and accurately analyze the copper tube assembly in the image, eliminating the tedious and time-consuming process of manual inspection.

[0084] The evaporator detection method in this embodiment addresses common problems during copper tube assembly on air conditioner evaporators, such as missing insertions, incorrect insertions, insufficient insertion depth, and collisions. Missing insertions are identified by the fact that they create circular holes during actual assembly. Incorrect insertions are identified by the requirement that the center of the assembled tube aligns with the horizontal line connecting the circular holes. For depth and collision detection, the depth of copper tube insertion causes pixel differences in the image, and uneven reflection from collisions leads to uneven image pixels. This embodiment detects depth and whether collisions have occurred. This embodiment is not limited to assembling objects with circular holes; other shapes such as square holes and polygons also follow the same concept. The circle detection module can be replaced according to specific detection requirements. Besides traditional Hough transform, edge detection, or template matching algorithms, more advanced circle detection algorithms, such as machine learning-based methods or deep learning models, can also be used.

[0085] Example 4 In this embodiment, as Figure 2 As shown, a detection device for an evaporator is provided, comprising: Image acquisition module 210 is used to acquire the image to be tested; Image parsing module 220 is used to parse the image to be tested and extract the copper pipe region from the image to be tested; The pixel calculation module 230 is used to calculate the pixel mean of the copper tube region as the mean to be measured, and to calculate the pixel variance of the copper tube region as the variance to be measured. The impact detection module 240 is used to compare the variance to be measured with the preset copper pipe variance to obtain the copper pipe impact detection result; The depth detection module 250 is used to compare the measured mean value with the preset copper tube mean value to obtain the copper tube depth detection result.

[0086] In one embodiment, the depth detection module 250 is used to determine, based on the copper pipe impact detection results, the copper pipe areas whose impact detection results are qualified from a plurality of copper pipe areas as qualified impact pipe areas; and compare the test mean of the qualified impact pipe areas with the preset copper pipe mean to obtain the copper pipe depth detection result.

[0087] In one embodiment, the device further includes an image grayscale module and a missing insertion detection module. The image grayscale module is used to perform grayscale processing on the image to be tested to obtain a grayscale image; The missing insertion detection module is used to perform Hough circle detection on the grayscale image to obtain a number of candidate circles; and to detect whether the diameter of each candidate circle matches the preset screening diameter to obtain the copper tube missing insertion detection result.

[0088] In one embodiment, the image parsing module 220 is used to extract a plurality of contour rectangles from the image to be tested based on a preset contour extraction algorithm; calculate the ratio of the width to the height of the plurality of contour rectangles as the ratio to be tested; detect whether each ratio to be tested matches a preset sieve ratio; and when the ratio to be tested matches the preset sieve ratio, determine that the region corresponding to the contour rectangle is the copper tube region.

[0089] In one embodiment, the device further includes a misinsertion detection module, which is used to detect the center point of each of the copper tube regions, obtain the coordinates of the center point of each of the copper tube regions as the coordinates to be measured, and detect each of the coordinates to be measured based on a preset misinsertion detection algorithm to obtain the copper tube misinsertion detection result.

[0090] In one embodiment, the misinsertion detection module is used to obtain the coordinate value of the center point of the copper tube region in a first direction as a first measured coordinate value based on the measured coordinates; wherein, the first direction is parallel to the arrangement direction of each hole group, and each hole group is used to insert a copper tube; the difference between the first measured coordinate value and the reference coordinate value is used to obtain the distance difference between the center point of the copper tube region and the reference hole; wherein, the reference coordinate value is the coordinate value of the reference hole in the first direction, and the axis of the reference hole and the hole corresponding to the copper tube region is parallel to the first direction; the distance difference is divided by a preset interval difference to obtain a difference multiple; when the integer part of the difference multiple is even, the copper tube misinsertion detection result in the first direction is determined to be no misinsertion; when the integer part of the difference multiple is odd, the copper tube misinsertion detection result in the first direction is determined to be misinsertion.

[0091] In one embodiment, the misinsertion detection module is used to obtain the coordinate value of the center point of the copper tube area in a second direction as a second coordinate value based on the coordinates to be measured; wherein, the second direction is perpendicular to the arrangement direction of each hole group, and each hole group is used to insert a copper tube; the second coordinate value to be measured is compared with a preset second coordinate reference value to obtain the second direction detection result as the copper tube misinsertion detection result.

[0092] Specific limitations regarding the evaporator detection device can be found in the above description of the evaporator detection method, and will not be repeated here. Each unit in the aforementioned evaporator detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These units can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each unit.

[0093] Example 5 In this embodiment, a computer device is provided. Its internal structure diagram can be shown as follows: Figure 3 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs, and also contains a database storing preset copper tube variances and preset copper tube mean values. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface communicates with other computer devices that have deployed application software. When executed by the processor, the computer program implements a method for detecting evaporators. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device casing, or an external keyboard, touchpad, or mouse.

[0094] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0095] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the evaporator detection method described in any of the above embodiments.

[0096] Example 6 In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the evaporator detection method described in any of the above embodiments.

[0097] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0098] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for detecting an evaporator, characterized in that, include: Acquire an image to be tested, which contains a copper tube and fins. The fins have holes for inserting the copper tube. Analyze the image to be tested and extract the copper pipe region from the image to be tested; The pixel mean of the copper tube region is calculated as the mean to be measured, and the pixel variance of the copper tube region is calculated as the variance to be measured. The test variance is compared with the preset copper tube variance to obtain the copper tube impact detection result; When the copper tube impact test result is deemed qualified, the measured mean value and the preset copper tube mean value are compared to obtain the copper tube depth test result. The copper tube depth test result is used to indicate whether the depth of the copper tube inserted into the hole on the fin is normal. When calculating the measured variance, the preset copper tube mean value is used as the benchmark value to eliminate the interference of uniform deformation on the impact test. After acquiring the image to be tested, the following steps are included: The image to be tested is converted to grayscale to obtain a grayscale image; Perform Hough circle detection on the grayscale image to obtain several candidate circles; The diameter of each candidate circle is checked to see if it matches the preset screening diameter, and the result of copper tube insertion failure detection is obtained.

2. The method according to claim 1, characterized in that, The step of comparing the measured mean with the preset copper tube mean includes: Based on the copper pipe impact test results, the copper pipe areas whose impact test results are qualified are determined from several copper pipe areas as qualified impact pipe areas. The average value of the tested copper pipe in the qualified impacted pipe area is compared with the preset average value of the copper pipe to obtain the copper pipe depth detection result.

3. The method according to claim 1, characterized in that, The step of parsing the image to be tested and extracting the copper pipe region from the image to be tested includes: Based on a preset contour extraction algorithm, several contour rectangles are extracted from the image to be tested; Calculate the ratio of the width to the height of several outline rectangles as the scale to be measured; The test ratio is checked to see if it matches the preset sieve ratio. When the test ratio matches the preset sieve ratio, the area corresponding to the outline rectangle is determined to be the copper tube area.

4. The method according to claim 1, characterized in that, The method further includes: The center point of each copper tube region is detected, and the coordinates of the center point of each copper tube region are used as the coordinates to be measured. Based on a preset misinsertion detection algorithm, the coordinates to be tested are detected to obtain the copper tube misinsertion detection results.

5. The method according to claim 4, characterized in that, The preset misinsertion detection algorithm includes a first misinsertion detection algorithm, which includes: Based on the coordinates to be measured, the coordinates of the center point of the copper tube area in the first direction are obtained as the first coordinates to be measured; wherein, the first direction is parallel to the arrangement direction of each hole group, and each hole group is used to insert a copper tube. The difference between the first measured coordinate value and the reference coordinate value is obtained to obtain the distance difference between the center point of the copper tube area and the reference hole; wherein, the reference coordinate value is the coordinate value of the reference hole in the first direction, and the axis of the reference hole and the hole corresponding to the copper tube area is parallel to the first direction; Divide the distance difference by the preset interval difference to obtain the difference multiple; The parity of the integer part of the difference multiple is checked to see if it matches the layout of the hole group in the first direction. When the integer part of the difference multiple is even, the copper tube misinsertion detection result in the first direction is determined to be that no misinsertion has occurred; when the integer part of the difference multiple is odd, the copper tube misinsertion detection result in the first direction is determined to be that misinsertion has occurred.

6. The method according to claim 4, characterized in that, The preset misinsertion detection algorithm includes a second misinsertion detection algorithm, which includes: Based on the coordinates to be measured, the coordinates of the center point of the copper tube area in the second direction are obtained as the second coordinates to be measured; wherein, the second direction is perpendicular to the arrangement direction of each hole group, and each hole group is used to insert a copper tube. The second coordinate value to be measured is compared with the preset second coordinate reference value to obtain the second direction detection result as the copper tube misinsertion detection result.

7. A detection device for an evaporator, characterized in that, include: An image acquisition module is used to acquire an image to be tested, which contains a copper tube and fins. Holes are provided on the fins for inserting the copper tube. The image analysis module is used to analyze the image to be tested and extract the copper pipe region from the image to be tested; The pixel calculation module is used to calculate the pixel mean of the copper tube region as the mean to be measured, and to calculate the pixel variance of the copper tube region as the variance to be measured. The impact detection module is used to compare the variance to be measured with the preset copper tube variance to obtain the copper tube impact detection result; The depth detection module is used to compare the measured mean value with the preset copper tube mean value when the copper tube impact test result is qualified, so as to obtain the copper tube depth detection result. The copper tube depth detection result is used to indicate whether the depth of the copper tube inserted into the hole on the fin is normal. When calculating the measured variance, the preset copper tube mean value is used as the benchmark value to eliminate the interference of uniform deformation on the impact test. The device also includes an image grayscale module and a missing insertion detection module. The image grayscale module is used to perform grayscale processing on the image to be tested to obtain a grayscale image; The missing insertion detection module is used to perform Hough circle detection on the grayscale image to obtain a number of candidate circles; and to detect whether the diameter of each candidate circle matches the preset screening diameter to obtain the copper tube missing insertion detection result.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images