A wheel relaxation line loosening detection method, device, equipment and medium

Abstract

The application provides a wheel relaxation line loosening detection method, device, equipment and medium. The method extracts a whole relaxation line area by performing a series of image processing on the acquired wheel image, and then preliminarily judges whether the relaxation line is loosened according to the area of the whole relaxation line area. If the preliminary judgment result is that the relaxation line is not loosened, the ratio of the area of the whole relaxation line area to the minimum circumscribed rectangle area of the whole relaxation line is further used to judge whether the relaxation line is loosened. In this way, the relaxation line loosening detection task is standardized, automatic detection is realized, and missed detection and errors caused by negligence in manual detection are avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wheel relaxation line detection, and particularly relates to a wheel relaxation line loosening detection method, device, equipment and medium. BACKGROUND

[0002] Locomotive automatic maintenance has been the goal of locomotive maintenance. Detection as a front-end process of repair has guiding significance. Wheel relaxation line detection is an important part of wheel detection, which can determine the loosening of the wheel rim by finding the misalignment of the relaxation line, so as to prevent possible major failures.

[0003] The existing wheel relaxation line loosening detection mainly relies on the naked eye judgment of maintenance workers, relies on a round of inspection around the locomotive, and checks the condition of each wheel to prevent the loosening of the wheel rim. However, through manual inspection, there may be missed detection and false detection caused by the negligence of maintenance workers. SUMMARY

[0004] To solve the technical problem of missed detection and false detection caused by the negligence of maintenance workers in the wheel relaxation line loosening detection relying on manual naked eye, the present application provides a wheel relaxation line loosening detection method, device, equipment and medium.

[0005] In a first aspect, the present application provides a wheel relaxation line loosening detection method, comprising:

[0006] Obtaining a wheel image, and the relaxation line in the wheel image is complete;

[0007] Converting the wheel image from an RGB color space to an HSV color space to obtain an HSV color space wheel image; and extracting a maximum region of a specified color in the HSV color space wheel image to obtain a wheel spoke region extraction image; the specified color is the color of the wheel spoke of the wheel to be detected;

[0008] Performing image morphological processing on the wheel spoke region extraction image to extract a partial relaxation line region;

[0009] Enlarging and selecting a region corresponding to the partial relaxation line region in the RGB color space wheel image based on the extracted partial relaxation line region to obtain a to-be-selected region image;

[0010] Extracting a pixel point with an RGB value satisfying a set condition from the to-be-selected region image to obtain an entire relaxation line region;

[0011] Preliminarily judging whether the relaxation line is loosened according to the area of the entire relaxation line region and a preset area threshold; if it is preliminarily judged that the relaxation line is not loosened, judging again whether the relaxation line is loosened according to the ratio of the area of the entire relaxation line region to the area of the minimum circumscribed rectangle of the entire relaxation line.

[0012] In some implementations, the region with the maximum specified color in the wheel image in the HSV color space is extracted to obtain a spoke region extraction image, including:

[0013] The wheel image in the HSV color space is binarized according to a preset specified color threshold region;

[0014] All contours in the binarized wheel image are found, and the areas of the regions enclosed by the contours in the binarized wheel image are counted;

[0015] The region with the maximum area of the regions enclosed by the contours is extracted from the binarized wheel image to obtain the spoke region extraction image.

[0016] In some implementations, the spoke region extraction image is subjected to image morphological processing to extract a partial relaxation line region, including:

[0017] The spoke region extraction image is subjected to morphological closing operation processing to obtain a closing operation result image;

[0018] The spoke region extraction image and the closing operation result image are subjected to difference operation processing to obtain a difference result image, and the difference result image represents a changed region image after the morphological closing operation processing;

[0019] All contours in the difference result image are found, and the areas of the regions enclosed by the contours in the difference result image are counted;

[0020] The region with the maximum area of the regions enclosed by the contours is extracted from the difference result image to obtain the partial relaxation line region.

[0021] In some implementations, after the difference result image is obtained, the difference result image is subjected to opening operation processing to eliminate part of the noise regions in the difference result image.

[0022] In some implementations, the selected region in the wheel image in the RGB color space is expanded based on the extracted partial relaxation line region to obtain a to-be-selected region image, including:

[0023] In the wheel image in the RGB color space, a whole circular region is extracted with the center of the minimum circumscribed rectangle of the partial relaxation line region as the center and a preset multiple of the length of the rectangle as the radius, and the whole circular region is the to-be-selected region image.

[0024] In some implementations, the condition is:

[0025] The R value of the pixel point is in [R mode -THD, R mode +THD];

[0026] the G value of the pixel point is in [G mode -THD, G mode +THD];

[0027] the B value of the pixel point is in [B mode -THD, B mode +THD];

[0028] wherein, R mode , G mode , B mode are the mode of the R value, the G value, and the B value of the partial relaxation line region respectively; and THD is a preset threshold value.

[0029] In some implementations, the area of the entire relaxation line region is compared with a preset area threshold value to preliminarily determine whether the relaxation line is loose; if it is preliminarily determined that the relaxation line is not loose, the ratio of the area of the entire relaxation line region to the minimum circumscribed rectangle area of the entire relaxation line region is used to determine again whether the relaxation line is loose, including:

[0030] comparing the area of the entire relaxation line region with a preset area threshold value to preliminarily determine whether the relaxation line is loose;

[0031] if the area of the entire relaxation line region is less than or equal to the preset area threshold value, it is determined that the relaxation line is loose;

[0032] if the area of the entire relaxation line region is greater than the preset area threshold value, the ratio of the area of the entire relaxation line region to the minimum circumscribed rectangle area of the entire relaxation line region is used to determine again whether the relaxation line is loose;

[0033] if the ratio of the area of the entire relaxation line region to the minimum circumscribed rectangle area of the entire relaxation line region is less than a preset duty cycle threshold value, it is determined that the relaxation line is loose;

[0034] if the ratio of the area of the entire relaxation line region to the minimum circumscribed rectangle area of the entire relaxation line region is greater than or equal to the preset duty cycle threshold value, it is determined that the relaxation line is not loose.

[0035] In a second aspect, an embodiment of the present application provides a wheel relaxation line looseness detection device, including:

[0036] an acquisition module configured to acquire a wheel image, wherein the relaxation line in the wheel image is complete;

[0037] a first region extraction module configured to convert the wheel image from an RGB color space to an HSV color space to obtain an HSV wheel image; and extract a region with the maximum specified color in the HSV wheel image to obtain a spoke region extraction image; the specified color is the color of the wheel spoke of a wheel to be detected;

[0038] The second region extraction module is configured to perform image morphological processing on the spoke region extraction image to extract a partial slack line region.

[0039] The third region extraction module is configured to expand the selected region corresponding to the partial slack line region in the wheel image in the RGB color space based on the extracted partial slack line region to obtain a to-be-selected region image.

[0040] The fourth region extraction module is configured to extract, from the to-be-selected region image, a pixel point whose RGB value satisfies a set condition to obtain an entire slack line region.

[0041] The judging module is configured to preliminarily judge whether the slack line is loose according to an area of the entire slack line region and a preset area threshold value, and if the slack line is preliminarily judged as not loose, judge whether the slack line is loose according to a ratio of the area of the entire slack line region to an area of a minimum circumscribed rectangle of the entire slack line.

[0042] In a third aspect, an embodiment of the present application provides a computer readable storage medium, and the computer readable storage medium stores a computer program.

[0043] In a fourth aspect, an embodiment of the present application provides an electronic device, which comprises a memory and at least one processor, and the memory stores a computer program.

[0044] One or more embodiments of the present application bring at least the following beneficial effects:

[0045] The present application extracts the entire slack line region by performing a series of image processing on the obtained wheel image, then preliminarily judges whether the slack line is loose according to the area of the entire slack line region, and if the slack line is preliminarily judged as not loose, further judges whether the slack line is loose according to the ratio of the area of the entire slack line region to the area of the minimum circumscribed rectangle of the entire slack line, so as to standardize the slack line loose detection task, realize automatic detection, and avoid missed detection and errors caused by manual detection. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope.

[0047] Figure 1 The wheel slack line loose detection method flowchart provided by the embodiments of the present application;

[0048] Figure 2 A wheel image provided by an embodiment of the present application with a standard relaxation line;

[0049] Figure 3 A wheel image provided by an embodiment of the present application after binarization processing;

[0050] Figure 4 A wheel spoke area extraction image schematic diagram provided by an embodiment of the present application;

[0051] Figure 5 A closed operation result image schematic diagram provided by an embodiment of the present application;

[0052] Figure 6 A difference result image schematic diagram provided by an embodiment of the present application;

[0053] Figure 7 A difference result image schematic diagram provided by an embodiment of the present application after open operation processing;

[0054] Figure 8 A partial relaxation line area extraction schematic diagram provided by an embodiment of the present application;

[0055] Figure 9 An enlarged selected to-be-selected area image schematic diagram provided by an embodiment of the present application;

[0056] Figure 10 An effect diagram of a whole relaxation line area provided by an embodiment of the present application;

[0057] Figure 11 A final effect diagram of a whole relaxation line area provided by an embodiment of the present application;

[0058] Figure 12 A whole relaxation line minimum circumscribed rectangle schematic diagram provided by an embodiment of the present application when the relaxation line is loose and when the relaxation line is not loose;

[0059] Figure 13 A structure diagram of a wheel relaxation line loosening detection device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0061] Embodiment one

[0062] The embodiment provides a wheel relaxation line loosening detection method, as shown in the figure, which comprises the following steps. Figure 1

[0063] In step S100, a wheel image is acquired, and a complete relaxation line is exposed in the wheel image.

[0064] The locomotive wheel can be roughly divided into two types of wheel rim wheels and integral wheels, and only the wheel rim wheel needs relaxation line detection. In the wheel image acquisition process, it is necessary to ensure that the current relaxation line is exposed completely. The collected wheel image with standard relaxation line is as shown in the figure. Figure 2

[0065] In step S200, the wheel image is converted from the RGB color space to the HSV color space to obtain the wheel image in the HSV color space, and the maximum region of a specified color in the wheel image in the HSV color space is extracted, which is a wheel spoke region extraction image. The specified color is the color of the wheel spoke of the wheel to be detected.

[0066] In the embodiment, the color of the wheel spoke of the wheel to be detected is unified as red, and the position and shape of the wheel spoke can be determined by extracting the red region in the figure.

[0067] The HSV color space is a more intuitive color model for an observer, and therefore, in the embodiment, the wheel image is first converted from the RGB color space to the HSV color space to obtain the wheel image in the HSV color space. The wheel spoke region extraction image is extracted by binarizing the wheel image in the HSV color space according to a preset red threshold. Specifically, the pixels in the wheel image in the HSV color space whose pixel values are in the preset red threshold region are marked as 1, and the pixels whose pixel values are not in the preset red threshold region are marked as 0. The wheel image after binarization is as shown in the figure. Figure 3 ​​As shown in FIG. 6, all contours in the binarized wheel image are searched, and the area of each contour in the binarized wheel image is counted. A region with the largest area enveloped by the contours is extracted from the binarized wheel image as a spoke region extraction image, as shown in FIG. 7. Figure 4

[0068] In step S300, image morphological processing is performed on the spoke region extraction image to extract a partial relaxation line region.

[0069] In some implementations, step S300 includes the following steps:

[0070] In step S310, morphological closing operation processing is performed on the spoke region extraction image to obtain a closing operation result image. The closing operation result image is as shown in FIG. 8. Figure 5

[0071] In step S320, difference operation processing is performed on the spoke region extraction image and the closing operation result image to obtain a difference result image, which represents a change region image after the morphological closing operation processing. The difference result image is as shown in FIG. 9. Figure 6 Figure 7

[0072] In step S330, all contours in the difference result image are searched, and the area of each contour in the difference result image is counted.

[0073] In step S340, a region with the largest area enveloped by the contours is extracted from the difference result image, which is a partial relaxation line region. The extraction of the partial relaxation line region is as shown in FIG. 10. Figure 8

[0074] In step S400, the extracted partial relaxation line region is used as a reference to expand and select a region corresponding to the partial relaxation line region in the wheel image in the RGB color space to obtain a to-be-selected region image.

[0075] In some implementations, a minimum circumscribed rectangle of the partial relaxation line region is taken as a center, and a rectangle length with a preset multiple is taken as a radius to extract an overall circular region in the wheel image in the RGB color space, which is the to-be-selected region image. The extracted to-be-selected region image is as shown in FIG. 11. Figure 9

[0076] In step S500, a pixel point with an RGB value satisfying a preset condition is extracted from the to-be-selected region image to obtain an entire relaxation line region. The preset condition is that: ​​​​​​

[0077] The R value of a pixel is in [R mode -THD, R mode +THD];

[0078] The G value of a pixel is in [G mode -THD, G mode +THD];

[0079] The B value of a pixel is in [B mode -THD, B mode +THD];

[0080] Among them, R mode G mode B mode These are the modes of the R, G, and B values ​​in the partial relaxation line region, respectively; THD is the preset threshold.

[0081] The specific implementation method includes: separately calculating the mode R of the R, G, and B values ​​in the partial relaxation line region. mode G mode B mode The process involves finding pixels in the image that meet preset conditions to obtain the entire relaxation line region. The resulting image of the entire relaxation line region is shown below. Figure 10 and Figure 11 As shown.

[0082] Step S600: Based on the area of ​​the entire relaxation line region and the preset area threshold, a preliminary judgment is made as to whether the relaxation line is loose; if the relaxation line is initially determined to be not loose, then based on the ratio of the area of ​​the entire relaxation line region to the area of ​​the smallest bounding rectangle of the entire relaxation line, a further judgment is made as to whether the relaxation line is loose.

[0083] If the relaxation line is loosened significantly, resulting in breaks on both sides, the area of ​​the entire relaxation line region will be much smaller. Therefore, to address this situation, an area threshold is set. If the area of ​​the entire relaxation line region is lower than this area threshold, it is determined that the relaxation line has become loose.

[0084] If the area of ​​the entire relaxation line region is greater than or equal to a preset area threshold, further assessment of the relaxation line's looseness is required. When the relaxation line is not loose, it forms a regular rectangle, so the ratio of the area of ​​the entire relaxation line region to the area of ​​the smallest bounding rectangle of the entire relaxation line will be close to 1. Figure 12 As shown in the left-hand diagram. If the slack line becomes loose, the slack lines between the wheel rim and spokes will misalign, resulting in many empty areas within the minimum bounding rectangle of the slack line, as shown below. Figure 12As shown in the right side of the figure, the ratio of the area of the entire relaxed line region to the area of the minimum circumscribed rectangle of the entire relaxed line will decrease. Therefore, by setting a duty cycle threshold, the above two cases can be distinguished. In this embodiment, the duty cycle threshold is set to 0.8, that is, if the ratio of the area of the entire relaxed line region to the area of the minimum circumscribed rectangle of the entire relaxed line is less than 0.8, it is determined that the relaxed line has loosened; if the ratio of the area of the entire relaxed line region to the area of the minimum circumscribed rectangle of the entire relaxed line is greater than or equal to 0.8, it is determined that the relaxed line has not loosened.

[0085] In this embodiment, the acquired wheel image is subjected to a series of image processing to extract the entire relaxed line region, and then the area of the entire relaxed line region is used to preliminarily determine whether the relaxed line has loosened. If the preliminary determination result is that the relaxed line has not loosened, the ratio of the area of the entire relaxed line region to the area of the minimum circumscribed rectangle of the entire relaxed line is used to further determine whether the relaxed line has loosened. In this way, the relaxed line loosening detection task is standardized, automatic detection is realized, and missed detection and errors caused by manual detection due to negligence are avoided.

[0086] Embodiment Two

[0087] This embodiment provides a relaxed line loosening detection device for a vehicle wheel, as shown in the figure, which comprises: Figure 13

[0088] The acquisition module 210 is configured to acquire a wheel image, and the relaxed line in the wheel image is complete.

[0089] Locomotive wheels can be roughly divided into two types: wheel rims and whole wheels, and only wheel rims need relaxed line detection. During wheel image acquisition, it is necessary to ensure that the current relaxed line is complete. The acquired wheel image with a standard relaxed line is as shown in the figure. Figure 2

[0090] The first region extraction module 220 is configured to convert the wheel image from an RGB color space to an HSV color space to obtain an HSV color space wheel image, and extract a region with the maximum specified color in the HSV color space wheel image to obtain a spoke region extraction image; the specified color is the color of the wheel spoke of the wheel to be detected.

[0091] In this embodiment, the color of the wheel spoke of the wheel to be detected is uniform, and the position and shape of the wheel spoke can be determined by extracting the red region in the figure.

[0092] ​​The HSV color space is a more intuitive color model for an observer, and thus, in this embodiment, the wheel image is first converted from the RGB color space to the HSV color space to obtain a wheel image in the HSV color space. The wheel spoke region extraction image is extracted by binarizing the wheel image in the HSV color space according to a preset red color threshold. Specifically, pixels in the wheel image in the HSV color space whose values are in the preset red color threshold region are marked as 1, and pixels whose values are not in the preset red color threshold region are marked as 0. The wheel image after binarization is shown in FIG. 8. Figure 3 All contours in the wheel image after binarization are found, and the areas of the regions enclosed by the contours in the wheel image after binarization are counted. The region with the largest area enclosed by the contours is extracted from the wheel image after binarization as the wheel spoke region extraction image, which is shown in FIG. 9. Figure 4

[0093] The second region extraction module 230 is configured to perform image morphological processing on the wheel spoke region extraction image to extract a partial relaxation line region.

[0094] Specifically, the process of extracting the partial relaxation line region by the second region extraction module 230 includes the following steps.

[0095] Step one: performing morphological closing operation processing on the wheel spoke region extraction image to obtain a closing operation result image. The closing operation result image is shown in FIG. 10. Figure 5

[0096] Step two: performing difference operation processing on the wheel spoke region extraction image and the closing operation result image to obtain a difference result image, which represents a change region image after the morphological closing operation processing. The difference result image is shown in FIG. 11. Figure 6 To eliminate part of the noise regions in the difference result image, the obtained difference result image is subjected to opening operation processing. The difference result image after the opening operation processing is shown in FIG. 12. Figure 7

[0097] Step three: finding all contours in the difference result image and counting the areas of the regions enclosed by the contours in the difference result image.

[0098] Step four: extracting the region with the largest area enclosed by the contours from the difference result image, which is the partial relaxation line region. The extraction of the partial relaxation line region is shown in FIG. 13. Figure 8

[0099] The third region extraction module 240 is configured to expand the region corresponding to the partial relaxation line region in the wheel image in the RGB color space based on the extracted partial relaxation line region to obtain a to-be-selected region image.​​​​

[0100] In some implementations, the third region extraction module 240 is configured to extract, in the wheel image in the RGB color space, a whole circular region as the to-be-selected region image, with the center of the minimum bounding rectangle of the partial cat-eye region as the center, and with the length of the rectangle as the radius. Figure 9

[0101] The fourth region extraction module 250 is configured to extract, from the to-be-selected region image, pixel points whose RGB values satisfy a set condition, to obtain the whole cat-eye region. The set condition is as follows:

[0102] The R value of the pixel point is in [R mode -THD, R mode +THD];

[0103] The G value of the pixel point is in [G mode -THD, G mode +THD];

[0104] The B value of the pixel point is in [B mode -THD, B mode +THD];

[0105] wherein R mode , G mode , and B mode are the mode of the R value, the G value, and the B value of the partial cat-eye region, respectively; and THD is a preset threshold.

[0106] Specifically, the fourth region extraction module 250 is configured to respectively calculate the mode R mode , G mode , and B mode of the R value, the G value, and the B value of the partial cat-eye region; and find, in the to-be-selected region image, pixel points that satisfy the preset condition, to obtain the whole cat-eye region. An effect diagram of the obtained whole cat-eye region is shown in FIGS. 5 and 6. Figure 10 Figure 11

[0107] The judging module 260 is configured to preliminarily judge whether the cat-eye is loose according to the area of the whole cat-eye region and a preset area threshold; and if the preliminary judgment is that the cat-eye is not loose, judge whether the cat-eye is loose according to the ratio of the area of the whole cat-eye region to the area of the minimum bounding rectangle of the whole cat-eye.

[0108] The judging rule of the judging module 260 for the cat-eye looseness is as follows:

[0109] ​​​If the area of the whole relaxed line region is less than the area threshold, it is determined that the relaxed line is loose; if the area of the whole relaxed line region is greater than or equal to the area threshold, it is further determined whether the relaxed line is loose according to a ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line, if the ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line is less than a preset duty cycle threshold, it is determined that the relaxed line is loose; if the ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line is greater than or equal to the preset duty cycle threshold, it is determined that the relaxed line is not loose.

[0110] The embodiment extracts the whole relaxed line region by performing a series of image processing on the obtained wheel image, and then preliminarily determines whether the relaxed line is loose according to the area of the whole relaxed line region, if the preliminary determination result is that the relaxed line does not loose, it is further determined whether the relaxed line is loose according to the ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line, so as to standardize the relaxed line loose detection task, realize automatic detection, and avoid missed detection and errors caused by human detection.

[0111] Embodiment three

[0112] The embodiment provides a computer readable storage medium, and the computer readable storage medium stores a computer program. When the computer program is executed by at least one processor, the machine vision-based wheel relaxed line loose detection method of the foregoing embodiment is implemented.

[0113] The computer readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, for example, static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0114] Embodiment four

[0115] The embodiment provides an electronic device, which includes a memory and at least one processor. The memory stores a computer program, and the computer program is executed by the at least one processor to implement the method of the foregoing embodiment.

[0116] The processor can be an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a controller, a microcontroller unit (MCU), a microprocessor, or other electronic elements, which are configured to execute the machine vision-based wheel relaxation line loosening detection method in the above embodiments.

[0117] Embodiment Five

[0118] The embodiment provides a computer program product, which executes the machine vision-based wheel relaxation line loosening detection method of the above embodiment when running on a processor.

[0119] In actual application, the computer program product can be implemented to run in an electronic device.

[0120] In the several embodiments provided by the present application, it should be understood that the disclosed device and method can also be implemented in other manners. The embodiments described above are only schematic.

[0121] It should be noted that, in this document, the terms "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. The terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or other elements inherent to such a process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0122] Although the embodiments of the present application have been disclosed with reference to the above embodiments, the above description is merely used to understand the present application and is not used to limit the present application. Any person skilled in the art, without departing from the spirit and scope of the present application, can make any modification and change in the form and details of the embodiments, but the patent protection scope of the present application should be subject to the scope defined by the appended claims.

Claims

1. A method of detecting relaxation of a wheel alignment, characterized by, The method comprises the following steps: acquiring a wheel image in which a slack line is completely exposed; converting the wheel image from an RGB color space to an HSV color space to obtain an HSV wheel image; and extracting a maximum region of a specified color in the HSV wheel image to obtain a wheel spoke region extraction image; the specified color is the color of a wheel spoke of a wheel to be detected; performing image morphological processing on the wheel spoke region extraction image to extract a partial slack line region; enlarging and selecting a region corresponding to the partial slack line region in the wheel image in the RGB color space based on the extracted partial slack line region to obtain a to-be-selected region image; extracting a pixel point with an RGB value satisfying a set condition from the to-be-selected region image to obtain an entire slack line region; preliminarily judging whether the slack line is loose according to a comparison between an area of the entire slack line region and a preset area threshold; if it is preliminarily judged that the slack line is not loose, judging again whether the slack line is loose according to a ratio of the area of the entire slack line region to an area of a minimum circumscribed rectangle of the entire slack line region; wherein the image morphological processing on the wheel spoke region extraction image to extract the partial slack line region comprises: performing a morphological closing operation on the wheel spoke region extraction image to obtain a closing operation result image; performing a difference operation on the wheel spoke region extraction image and the closing operation result image to obtain a difference result image, the difference result image representing a changed region image after the morphological closing operation; finding all contours in the difference result image and counting areas of regions enclosed by the contours in the difference result image; extracting a region with a maximum area from the regions enclosed by the contours in the difference result image to obtain the partial slack line region.

2. The wheel droop slack detection method according to claim 1, characterized by, The extraction of the maximum region of the specified color in the HSV wheel image to obtain the wheel spoke region extraction image comprises: performing a binarization operation on the HSV wheel image according to a preset specified color threshold region; finding all contours in the binarized wheel image and counting areas of regions enclosed by the contours in the binarized wheel image; extracting a region with a maximum area from the regions enclosed by the contours in the binarized wheel image to obtain the wheel spoke region extraction image.

3. The wheel looseness detection method according to claim 1, characterized by, After obtaining the difference result image, the method further comprises performing an opening operation on the obtained difference result image to eliminate part of noise regions in the difference result image.

4. The wheel droop looseness detection method according to claim 1, characterized by, The enlargement and selection of the region in the wheel image in the RGB color space based on the extracted partial slack line region to obtain the to-be-selected region image comprises: extracting an overall circular region in the wheel image in the RGB color space with a center of a minimum circumscribed rectangle of the partial slack line region as a center and a rectangular length of a preset multiple as a radius, the overall circular region being the to-be-selected region image.

5. The wheel looseness detection method according to claim 1, characterized by, The set condition is: The R value of the pixel point is in [R mode - THD, R mode + THD]. The G value of the pixel point is in [G mode -THD, G mode +THD] The B value of the pixel point is in [B mode - THD, B mode + THD] wherein R mode , G mode , and B mode are the mode of the R value, the G value, and the B value of the partial relaxation line region, respectively; and THD is a preset threshold value.

6. The wheel looseness detection method according to claim 1, characterized by, the preliminary judgment of whether the slack line is loose according to the comparison between the area of the entire slack line region and the preset area threshold; and the re-judgment of whether the slack line is loose according to the ratio of the area of the entire slack line region to the area of the minimum circumscribed rectangle of the entire slack line region if it is preliminarily judged that the slack line is not loose. According to the area of the whole relaxed line region and the preset area threshold, it is initially judged whether the relaxed line is loose or not; If the area of the whole relaxed line region is less than or equal to the preset area threshold, it is judged that the relaxed line is loose; If the area of the whole relaxed line region is greater than the preset area threshold, according to the ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line region, it is again judged whether the relaxed line is loose or not; If the ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line region is less than the preset duty cycle threshold, it is judged that the relaxed line is loose; If the ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line region is greater than or equal to the preset duty cycle threshold, it is judged that the relaxed line is not loose.

7. A wheel slack line loosening detection device, characterized in that, It comprises: An acquisition module is configured to acquire a wheel image, wherein a complete relaxed line is exposed in the wheel image; A first region extraction module is configured to convert the wheel image from an RGB color space to an HSV color space to obtain an HSV color space wheel image, and extract a maximum region of a specified color in the HSV color space wheel image to obtain a wheel spoke region extraction image, wherein the specified color is the color of a wheel spoke of a wheel to be detected; A second region extraction module is configured to perform image morphological processing on the wheel spoke region extraction image to extract a partial relaxed line region, wherein the image morphological processing comprises performing morphological closing operation processing on the wheel spoke region extraction image to obtain a closing operation result image, performing difference operation processing on the wheel spoke region extraction image and the closing operation result image to obtain a difference result image, finding all contours in the difference result image, and counting the areas of regions enclosed by the contours in the difference result image, and extracting a region with the largest area from the difference result image to obtain the partial relaxed line region; A third region extraction module is configured to expand and select a region corresponding to the partial relaxed line region in the RGB color space wheel image based on the extracted partial relaxed line region to obtain a region to be selected image; A fourth region extraction module is configured to extract pixel points with RGB values satisfying a set condition from the region to be selected image to obtain a whole relaxed line region; A judgment module is configured to initially judge whether the relaxed line is loose or not according to the area of the whole relaxed line region and a preset area threshold, and if it is initially judged that the relaxed line is not loose, to judge whether the relaxed line is loose or not according to the ratio of the area of the whole relaxed line region to the area of the minimum circumscribed rectangle of the whole relaxed line region.

8. An electronic device, comprising: The computer readable storage medium stores a computer program, and the computer program is executed by at least one processor to implement the method in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by at least one processor to implement the method in any one of claims 1 to 6.

Images