A device detection method, device, equipment and medium based on infrared image

Abstract

The embodiment of the present application discloses a device detection method, apparatus, equipment and medium based on infrared images, and relates to the field of image processing technology. When performing circle detection, the method first performs circle recognition, and determines the center of the circle in two-dimensional space by using the characteristic that the perpendicular bisector of any two points on the edge contour passes through the center of the circle. After determining the center of the circle, the radius is determined by a number of distance values ​​from the center of the circle to the edge contour in various directions, thereby obtaining a corrected circular image with redundant occluded parts removed, and then performing circle recognition. On the one hand, it can reduce misidentification caused by direct recognition, and on the other hand, compared with the existing Hough transform that is performed in three-dimensional space, this method is completed in two-dimensional space, which significantly improves the detection efficiency for the Hough transform with large computational complexity.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a device detection method, apparatus, equipment and medium based on infrared images. Background Art

[0002] Thermal infrared imaging, a technical term usually referred to as infrared imaging or thermal imaging, is an imaging method that captures the infrared energy radiated from the surface of an object and converts it into a visible image. This technology has shown great application value in many fields, especially in equipment inspection scenarios, due to its unique non-contact measurement characteristics. Infrared images can not only provide clear visual information at night or in low-light environments, but also reveal the temperature distribution on the surface of an object, providing an important basis for fault diagnosis, energy efficiency evaluation, etc.

[0003] In the specific application of equipment detection, shape recognition in infrared images, especially the detection of circular targets, is particularly important. Circular targets such as bearings, circular heating elements, and circular electronic components play a key role in mechanical equipment, and their health directly affects the operating efficiency and safety of the entire system. Accurate detection of these circular targets through infrared images can timely detect potential faults or anomalies, so that corresponding measures can be taken to prevent or repair them, thereby improving the reliability and service life of the equipment.

[0004] However, in the actual detection process, it is often encountered that the circular target is partially obscured. This obstruction may be caused by reasons such as equipment layout, interference from other components, or limitations on the detection angle of view. The existence of the occlusion phenomenon makes the circular target in the infrared image appear irregular in shape, which brings challenges to the detection work. When faced with such complex scenes, traditional detection methods often have difficulty in accurately identifying the obscured circular target, resulting in reduced accuracy of the detection results. Summary of the invention

[0005] The embodiments of the present application provide a device detection method, apparatus, equipment and medium based on infrared images, which can solve the problem that it is difficult to accurately detect circular targets when there is occlusion.

[0006] On the one hand, an embodiment of the present application provides a device detection method based on infrared images, including:

[0007] Acquire a target infrared image of the target device, wherein the target infrared image at least includes a partially obscured circular target;

[0008] Preprocessing the target infrared image to obtain a preprocessed infrared image;

[0009] Performing edge detection on the preprocessed infrared image to obtain the edge contour of the circular target;

[0010] Taking the perpendicular bisectors of any two points on the edge contour, obtaining an intersection area formed by a plurality of perpendicular bisectors, and determining the assumed center of the circular target according to the intersection area;

[0011] Determine a plurality of distance values ​​from the assumed circle center to the edge contour in various directions, and determine the radius of the circular target according to the plurality of distance values;

[0012] According to the assumed center of the circle and the radius, a corrected circle corresponding to the circular target is obtained, and redundant occluded parts are eliminated according to the corrected circle to obtain a corrected circular image;

[0013] Based on the corrected circular image, the type of the component corresponding to the circular target is determined.

[0014] As an optional implementation manner, the intersection area includes a plurality of intersection points; and determining the assumed center of the circular target according to the intersection area includes:

[0015] Performing circular fitting on the plurality of intersection points to obtain a fitting circle;

[0016] The center of the fitting circle is determined as the assumed center.

[0017] As an optional implementation manner, determining the radius of the circular target according to the plurality of distance values ​​includes:

[0018] Traversing the plurality of distance values, and forming a histogram of the distance values ​​among the plurality of distance values;

[0019] The radius of the circular target is determined according to the histogram.

[0020] As an optional implementation manner, determining the radius of the circular target according to the histogram includes:

[0021] The radius of the circular target is determined by using the distance value that appears most frequently in the histogram or the distance value that appears more frequently than a threshold of the number of occurrences.

[0022] As an optional implementation manner, traversing the plurality of distance values ​​and forming a histogram of distance values ​​among the plurality of distance values ​​includes:

[0023] Determining a radius range based on a true radius of a circular component corresponding to the circular target;

[0024] The plurality of distance values ​​are traversed according to the radius range, and a histogram of distance values ​​among the plurality of distance values ​​is formed.

[0025] As an optional implementation manner, preprocessing the target infrared image to obtain a preprocessed infrared image includes:

[0026] Performing noise reduction processing on the target infrared image to obtain a noise-reduced infrared image;

[0027] Edge enhancement is performed on the de-noised infrared image to obtain a pre-processed infrared image.

[0028] As an optional implementation manner, performing edge detection on the preprocessed infrared image to obtain the edge contour of the circular target includes:

[0029] The Canny edge detection algorithm is used to perform edge detection on the preprocessed infrared image to obtain the edge contour of the circular target.

[0030] On the other hand, an embodiment of the present application provides a device detection apparatus based on infrared images, comprising:

[0031] An image acquisition module, used to acquire a target infrared image of a target device, wherein the target infrared image at least includes a partially obscured circular target;

[0032] An image preprocessing module, used for preprocessing the target infrared image to obtain a preprocessed infrared image;

[0033] An edge detection module, used for performing edge detection on the preprocessed infrared image to obtain the edge contour of the circular target;

[0034] A circle center determination module, used to take the perpendicular bisectors of any two points on the edge contour, obtain an intersection area formed by a plurality of perpendicular bisectors, and determine the assumed circle center of the circular target according to the intersection area;

[0035] A radius determination module, used to determine a number of distance values ​​from the assumed circle center to the edge contour in various directions, and determine the radius of the circular target according to the number of distance values;

[0036] An image correction module, used to obtain a corrected circle corresponding to the circular target according to the assumed center and the radius, and remove redundant occluded parts according to the corrected circle to obtain a corrected circular image;

[0037] The component recognition module is used to determine the type of the component corresponding to the circular target based on the corrected circular image.

[0038] On the other hand, an embodiment of the present application provides an electronic device, including:

[0039] A memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the aforementioned device detection method based on infrared images.

[0040] On the other hand, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and the processor executes the computer program to implement the aforementioned infrared image-based device detection method.

[0041] The beneficial effects of the present invention are:

[0042] The embodiments of the present application provide a device detection method, apparatus, equipment and medium based on infrared images. The method obtains a target infrared image of a target device, wherein the target infrared image includes at least a partially obscured circular target; preprocesses the target infrared image to obtain a preprocessed infrared image; performs edge detection on the preprocessed infrared image to obtain an edge contour of the circular target; takes the perpendicular bisectors of any two points on the edge contour to obtain an intersection area formed by several perpendicular bisectors, and determines an assumed center of the circular target based on the intersection area; determines several distance values ​​from the assumed center of the circle to the edge contour in various directions, and determines the radius of the circular target based on the several distance values; obtains a corrected circle corresponding to the circular target based on the assumed center of the circle and the radius, and eliminates redundant obscured parts based on the corrected circle to obtain a corrected circular image; and determines the type of component corresponding to the circular target based on the corrected circular image. That is, when performing circle detection, the method first performs circle recognition, and determines the center of the circle in two-dimensional space by using the characteristic that the perpendicular bisector of any two points on the edge contour passes through the center of the circle. After determining the center of the circle, the radius is determined by a number of distance values ​​from the center of the circle to the edge contour in various directions, thereby obtaining a corrected circular image with unnecessary occluded parts removed, and then performing circle recognition. On the one hand, it can reduce misrecognition caused by direct recognition, and on the other hand, compared with the existing Hough transform which is performed in three-dimensional space, this method is completed in two-dimensional space, which significantly improves the detection efficiency for the Hough transform with large computational complexity. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application;

[0045] Figure 2 is a flow chart of a device detection method based on infrared images provided in an embodiment of the present application;

[0046] Figure 3 is a flow chart of another device detection method based on infrared images provided in an embodiment of the present application;

[0047] Figure 4 is a flow chart of another device detection method based on infrared images provided in an embodiment of the present application;

[0048] Figure 5 It is a structural schematic diagram of an infrared image-based equipment detection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0050] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0051] In one embodiment, the device detection method based on infrared images provided by the present application can be applied to an electronic device. The electronic device can be a server or a client, and its internal structure diagram can be as follows: Figure 1As shown. The electronic device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor is used to provide computing and control capabilities; the memory includes a non-volatile storage medium and an internal memory, and the non-volatile storage medium stores an operating system, a computer program (the computer program implements a device detection method based on infrared images when executed by the processor) and a database, and the internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium; the database is used to store data involved in the execution of the device detection method based on infrared images, for example, the database can store data such as a collection of commodity information and recommended order time; the network interface is used to communicate with an external terminal through a network connection; the display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the housing of the electronic device, or an external keyboard, touchpad or mouse. Further, the client can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers and portable wearable devices, and the server can be implemented as an independent server or a server cluster composed of multiple servers.

[0052] Those skilled in the art will understand that Figure 1 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0053] Research has found that a common problem when using infrared equipment to monitor circular parts (such as bearings, circular heating elements, circular electronic components and other core components) is partial occlusion. This occlusion may be caused by the complexity of the internal structure of the equipment, the close arrangement of adjacent parts, or the limitation of the detection angle of view, causing the outline of the parts that should have been presented as clear circles in the infrared image to become blurred or irregular. For industrial processes that rely on circular recognition to confirm the identity of specific parts, the appearance of such irregular patterns undoubtedly increases the difficulty and uncertainty of recognition.

[0054] Traditionally, Hough Transform has been widely used in the field of circular target detection as a shape detection tool. Its advantage is that it can effectively process images containing a lot of noise and find peaks in the parameter space through a voting mechanism to determine the position and size of the circle in the image. However, when faced with complex infrared images with partial occlusion, the limitations of Hough Transform gradually become apparent. Occlusion not only changes the integrity of the circle, but also may introduce additional edge information, resulting in multiple local peaks in the parameter space, which in turn causes misrecognition, that is, irregular shapes are mistakenly judged as circles.

[0055] In addition, the computational complexity of the Hough transform is also an issue that cannot be ignored. Since it requires data transformation and cumulative voting in the three-dimensional parameter space, this process involves a large amount of computing resources. With the improvement of image resolution and the increase in detection requirements, the amount of calculation has increased dramatically, which not only affects the detection speed, but also may pose a challenge to industrial applications with high real-time requirements.

[0056] To solve the above problems, see Figure 2 , Figure 2 : is a schematic flow chart of a device detection method based on infrared images provided in an embodiment of the present application, the method comprising:

[0057] S10, acquiring a target infrared image of a target device, wherein the target infrared image at least includes a partially obscured circular target.

[0058] In the specific implementation process, the target infrared image refers to an infrared image taken of the target device, which is acquired by an infrared image acquisition device. It includes at least the partially obscured circular target mentioned above, and of course may also include other targets, such as unobstructed circular targets. The circular target is an image formed by a circular component, such as the end face of a bearing that generates heat during operation. A partially obscured circular target means that the circular target is obscured by components of other shapes, forming an irregular image area. The method of the present application is also applicable to unobstructed circular targets, but the effect is better for partially obscured circular targets.

[0059] It should be noted that the partially blocked circular target in this embodiment may not be an image area formed by a blocked circular component, but may be an irregular component.

[0060] S20, preprocessing the target infrared image to obtain a preprocessed infrared image.

[0061] In the specific implementation process, preprocessing may include noise reduction, edge enhancement, binarization, grayscale processing, etc. Noise reduction can reduce the noise in the image through methods such as Gaussian filtering or median filtering, edge enhancement can be achieved through deep learning, and OPENCV can be used to binarize the image and convert the image into a black and white image with only two pixel values ​​of 0 and 255. Binarization and grayscale processing can make the edge clearer and improve the efficiency and accuracy of subsequent edge detection.

[0062] As an optional implementation manner, preprocessing the target infrared image to obtain a preprocessed infrared image includes:

[0063] Performing noise reduction processing on the target infrared image to obtain a noise-reduced infrared image;

[0064] Edge enhancement is performed on the de-noised infrared image to obtain a pre-processed infrared image.

[0065] Noise reduction can be achieved by using methods such as Gaussian filtering or median filtering to reduce noise in images, and edge enhancement can be achieved through deep learning.

[0066] S30, performing edge detection on the preprocessed infrared image to obtain an edge contour of the circular target.

[0067] In a specific implementation process, the Canny edge detection algorithm may be used to perform edge detection on the pre-processed infrared image to obtain the edge contour of the circular target.

[0068] It is understandable that an edge is usually manifested as a significant change in pixel values, forming the outline of an object. Therefore, the edge outline of the circular object can be obtained by traversing pixels through the Canny edge detection algorithm.

[0069] It is understandable that if for a normal circular image without occlusion, a circular target image can be obtained by performing circle fitting after edge detection, but since the scene in this embodiment is partially occluded, the edge contour is not a regular circle, and it is impossible to directly determine whether it is a circular component. Therefore, it is necessary to process according to the subsequent steps to eliminate the influence of occlusion.

[0070] S40, taking the perpendicular bisectors of any two points on the edge contour, obtaining an intersection area formed by a plurality of perpendicular bisectors, and determining an assumed center point of the circular target according to the intersection area.

[0071] In the specific implementation process, the edge contour is composed of pixel points, so any points can be taken. For a regular circle, the perpendicular bisector of the straight line connecting any two points on the circumference must pass through the center of the circle. Therefore, in this embodiment, this point can be used to arbitrarily take points in different edge contour areas to obtain a number of perpendicular bisectors. Since the circular target is only partially blocked, these perpendicular bisectors of the retained circular part will intersect at the center of the circle under ideal conditions. However, since the edge contour cannot completely coincide with the real circle, these perpendicular bisectors will intersect in the vicinity of the true center of the circular target. Then, the hypothetical center of the circle can be determined through this intersection area. It is called the hypothetical center of the circle here because in actual operation, there is a deviation between the edge contour and the circular target.

[0072] In one possible case, the intersections of the perpendicular medians of the retained circular parts will form an intersection area with a relatively concentrated intersection. However, when the number of occurrences of the perpendicular medians is certain, if the number of retained circular parts is small, the number of intersections in this intersection area is not large. In this case, the number of intersections in the intersection area can be compared with a set threshold. If it exceeds the threshold, it can be defined as a true intersection area. Otherwise, this area can be determined as a non-circular area. Therefore, if there are too many obscured parts, it is almost impossible to determine whether the partially obscured circular target is originally a circle. On the other hand, the risk of misjudgment can be reduced by increasing the number of occurrences of the perpendicular medians and then making the above judgment.

[0073] It can be seen that in this embodiment, the number of intersection occurrences in the intersection area can also be used to preliminarily determine whether the partially blocked circular target is originally a circle.

[0074] In an alternative embodiment, see Figure 3 , the intersection area includes a plurality of intersection points; and determining the assumed center of the circular target according to the intersection area includes:

[0075] S402, performing circular fitting on the plurality of intersection points to obtain a fitting circle;

[0076] S404: Determine the center of the fitting circle as the assumed center.

[0077] In the specific implementation process, since several intersection points in the intersection area are all near the true center of the circle, several intersection points in the intersection area can be fitted into a circle by the least squares method, and then the center of the fitted circle is obtained to determine it as the assumed center of the circle.

[0078] It is understandable that the accuracy of the circle center determination is crucial to restore the circular target. The accuracy of the circle center determination can be improved by performing circular fitting on several intersection points in the intersection area.

[0079] S50, determining a plurality of distance values ​​from the assumed circle center to the edge contour in various directions, and determining the radius of the circular target according to the plurality of distance values.

[0080] In the specific implementation process, after determining the assumed center of the circle, in order to restore the circular target on the target infrared image, it is necessary to determine the radius of the circle corresponding to the circular target. Since the graphics corresponding to the partially obscured circular target are irregular in shape, the distance values ​​from the assumed center of the circle to the edge contour are also different. When the assumed center of the circle is relatively accurate, the distance values ​​from the edge of the retained circular part to the assumed center of the circle described above should be the same or similar, and these distance values ​​can be used to determine the radius.

[0081] For details, see Figure 4 , the step of determining the radius of the circular target according to the plurality of distance values ​​comprises:

[0082] S502, traversing the plurality of distance values, and forming a histogram of the distance values ​​among the plurality of distance values;

[0083] S504: Determine the radius of the circular target according to the histogram.

[0084] In a specific implementation process, a distance value may be obtained by traversing a number of distance values, thereby forming a histogram, so as to find the distance value (or distance value range) with the largest number of occurrences from the histogram and determine it as the radius.

[0085] In an optional implementation, determining the radius of the circular target according to the histogram includes:

[0086] The radius of the circular target is determined by using the distance value that appears most frequently in the histogram or the distance value that appears more frequently than a threshold of the number of occurrences.

[0087] It is understandable that the occurrence threshold can be set based on historical data, for example, the number of occurrences of distance values ​​that can be determined as the radius in the past in such circular target recognition. The number of occurrences in the histogram greater than the occurrence threshold can determine whether it can be determined as the radius on the one hand, and on the other hand, it can also be used to determine whether it was originally a circular target. This is because if the distance values ​​obtained in each direction are too few in the same or similar distance values, it is possible that the target itself is not circular, or the target is too blocked.

[0088] In some cases, the blocked portion is a relatively small portion, and therefore, when the distance values ​​obtained in various directions have a similar proportion, the distance value that appears most frequently in the histogram can be determined as the radius of the circular target.

[0089] In an optional implementation, the traversing of the plurality of distance values ​​and forming a histogram of the distance values ​​in the plurality of distance values ​​in S502 further includes the following steps:

[0090] S5022, determining a radius range based on a real radius of a circular component corresponding to the circular target;

[0091] S5024: Traverse the plurality of distance values ​​according to the radius range, and form a histogram of the distance values ​​among the plurality of distance values.

[0092] In the specific implementation process, the true radius refers to the radius of the circular component corresponding to the circular target on the target infrared image. The true radius belongs to the radius range and can be ±x based on the true radius, where x is a smaller natural number, such as 1, 2..., and the unit of x can be determined based on the true radius.

[0093] Traversing the distance values ​​according to the radius range, that is, traversing within the radius range, can significantly reduce the amount of traversed data and further improve the efficiency of device detection. And because the radius range is determined based on the real radius, the traversal result can ensure a certain degree of accuracy.

[0094] S60, obtaining a corrected circle corresponding to the circular target according to the assumed circle center and the radius, and removing redundant occluded parts according to the corrected circle to obtain a corrected circular image.

[0095] In the specific implementation process, the corrected circle is determined based on the assumed center and radius, and can be used as the imaging shape of the circular component. Therefore, by eliminating the redundant occluded part based on the corrected circle, the imaging shape of the circular component (i.e., the corrected circular image) can be obtained. It can be understood that eliminating the redundant occluded part means processing the pixel points outside the corrected circle to obtain the corrected circular image bounded by the corrected circle.

[0096] S70: Determine the type of the component corresponding to the circular target based on the corrected circular image.

[0097] In a specific implementation process, since the corrected circular image is a standard circular image, after the aforementioned recognition and processing, the component corresponding to the circular target that is partially blocked in the target infrared image can be determined as a circular component.

[0098] It can be seen that in this embodiment, by acquiring a target infrared image of a target device, the target infrared image includes at least a partially obscured circular target; preprocessing the target infrared image to obtain a preprocessed infrared image; edge detection is performed on the preprocessed infrared image to obtain an edge contour of the circular target; the perpendicular bisectors of any two points on the edge contour are taken to obtain an intersection area formed by several perpendicular bisectors, and an assumed center of the circular target is determined based on the intersection area; several distance values ​​from the assumed center of the circle to the edge contour in various directions are determined, and the radius of the circular target is determined based on the several distance values; based on the assumed center of the circle and the radius, a corrected circle corresponding to the circular target is obtained, and redundant obscured parts are eliminated based on the corrected circle to obtain a corrected circular image; based on the corrected circular image, the type of component corresponding to the circular target is determined. That is, when performing circle detection, circle recognition is performed first. The center of the circle is determined in two-dimensional space by using the characteristic that the perpendicular bisector of any two points on the edge contour passes through the center of the circle. After the center of the circle is determined, the radius is determined by several distance values ​​from the center of the circle to the edge contour in various directions, thereby obtaining a corrected circular image with unnecessary occluded parts removed, and then circle recognition is performed. On the one hand, it can reduce misidentification caused by direct recognition. On the other hand, compared with the existing Hough transform performed in three-dimensional space, this method is completed in two-dimensional space, which significantly improves the detection efficiency for the Hough transform with large computational complexity.

[0099] Based on the description of the above embodiments, the present application embodiment provides a device detection apparatus based on infrared images. Figure 5 , Figure 5 is a schematic diagram of the structure of an infrared image-based device detection apparatus provided in an embodiment of the present application, such as Figure 5 As shown, the device includes an image acquisition module 601, an image preprocessing module 602, an edge detection module 603, a circle center determination module 604, a radius determination module 605, an image correction module 606 and a component recognition module 607:

[0100] An image acquisition module 601 is used to acquire a target infrared image of a target device, wherein the target infrared image at least includes a partially obscured circular target;

[0101] An image preprocessing module 602 is used to preprocess the target infrared image to obtain a preprocessed infrared image;

[0102] An edge detection module 603 is used to perform edge detection on the pre-processed infrared image to obtain the edge contour of the circular target;

[0103] The center determination module 604 is used to take the perpendicular bisectors of any two points on the edge contour, obtain the intersection area formed by a plurality of perpendicular bisectors, and determine the assumed center of the circular target according to the intersection area;

[0104] A radius determination module 605 is used to determine a number of distance values ​​from the assumed circle center to the edge contour in various directions, and determine the radius of the circular target according to the number of distance values;

[0105] An image correction module 606 is used to obtain a correction circle corresponding to the circular target according to the assumed circle center and the radius, and remove redundant occluded parts according to the correction circle to obtain a correction circular image;

[0106] The component identification module 607 is used to determine the type of the component corresponding to the circular target based on the corrected circular image.

[0107] In one embodiment, based on the same inventive concept as the aforementioned embodiment, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and the processor executes the computer program to implement the method of the aforementioned embodiment.

[0108] It should be noted that the specific implementation methods and achievable technical effects of this embodiment can all refer to the contents of the aforementioned device detection method embodiment based on infrared images, which will not be repeated here.

[0109] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may 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.

[0110] What is disclosed above is only a partial embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. A person skilled in the art can understand that the entire or partial process of implementing the above embodiment and the equivalent changes made according to the claims of the present application are still within the scope of the invention.

Claims

1. A device detection method based on infrared images, characterized in that: include: Acquire a target infrared image of the target device, wherein the target infrared image at least includes a partially obscured circular target; Preprocessing the target infrared image to obtain a preprocessed infrared image; Performing edge detection on the preprocessed infrared image to obtain the edge contour of the circular target; Taking the perpendicular bisectors of any two points on the edge contour, obtaining an intersection area formed by a plurality of perpendicular bisectors, and determining the assumed center of the circular target according to the intersection area; Determine a plurality of distance values ​​from the assumed circle center to the edge contour in various directions, and determine a radius range based on a true radius of a circular component corresponding to the circular target; traverse the plurality of distance values ​​according to the radius range, and form a histogram of distance values ​​among the plurality of distance values; Determine the radius of the circular target by using the distance value that appears the most times in the histogram or the distance value that appears more than a threshold value of the number of times; According to the assumed center of the circle and the radius, a corrected circle corresponding to the circular target is obtained, and redundant occluded parts are eliminated according to the corrected circle to obtain a corrected circular image; Based on the corrected circular image, the type of the component corresponding to the circular target is determined.

2. The device detection method based on infrared images according to claim 1 is characterized in that: The intersection area includes a plurality of intersection points; and determining the assumed center of the circular target according to the intersection area includes: Performing circular fitting on the plurality of intersection points to obtain a fitting circle; The center of the fitting circle is determined as the assumed center.

3. The device detection method based on infrared images according to claim 1 is characterized in that: Preprocessing the target infrared image to obtain a preprocessed infrared image includes: Performing noise reduction processing on the target infrared image to obtain a noise-reduced infrared image; Edge enhancement is performed on the de-noised infrared image to obtain a pre-processed infrared image.

4. The device detection method based on infrared images according to claim 1 is characterized in that: Performing edge detection on the preprocessed infrared image to obtain the edge contour of the circular target includes: The Canny edge detection algorithm is used to perform edge detection on the preprocessed infrared image to obtain the edge contour of the circular target.

5. A device detection device based on infrared images, characterized in that: include: An image acquisition module, used to acquire a target infrared image of a target device, wherein the target infrared image at least includes a partially obscured circular target; An image preprocessing module, used for preprocessing the target infrared image to obtain a preprocessed infrared image; An edge detection module, used for performing edge detection on the preprocessed infrared image to obtain the edge contour of the circular target; A circle center determination module, used to take the perpendicular bisectors of any two points on the edge contour, obtain an intersection area formed by a plurality of perpendicular bisectors, and determine the assumed circle center of the circular target according to the intersection area; A radius determination module is used to determine a plurality of distance values ​​from the assumed circle center to the edge contour in various directions, and determine a radius range based on a true radius of a circular component corresponding to the circular target; traverse the plurality of distance values ​​according to the radius range, and form a histogram of the distance values ​​among the plurality of distance values; Determine the radius of the circular target by using the distance value that appears the most times in the histogram or the distance value that appears more than a threshold value of the number of times; An image correction module, used to obtain a corrected circle corresponding to the circular target according to the assumed center and the radius, and remove redundant occluded parts according to the corrected circle to obtain a corrected circular image; The component recognition module is used to determine the type of the component corresponding to the circular target based on the corrected circular image.

6. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement the method according to any one of claims 1 to 4.

Images