General pointer instrument visual identification method and system

Abstract

The invention relates to the technical field of visual identification, and discloses a universal pointer instrument visual identification method, which comprises the following steps of: obtaining a to-be-identified image containing a pointer instrument, a digital scale value list of the pointer instrument and clockwise and anticlockwise configuration instructions, the digital scale value list being a list arranged in sequence; adopting a target detection algorithm to obtain digital scale coordinates corresponding to the digital scale value list; according to the digital scale coordinates, obtaining a circular area in the center of an instrument panel and center coordinates of the pointer instrument; using a target threshold segmentation algorithm and according to the center coordinates, finding pointer tip coordinates of the pointer instrument; according to the digital scale coordinates, the center coordinates and the pointer tip coordinates, the reading value of the instrument pointer is obtained through calculation. According to the method, the pointer instrument with unclear scales can be identified.

Description

Technical Field

[0001] The present invention relates to the field of visual recognition technology, and in particular to a general pointer instrument visual recognition method and system. Background Art

[0002] By obtaining the readings of industrial instruments, we can monitor the detailed status of industrial production and intervene in industrial production in real time to achieve the goal of efficiency and safety. Pointer instruments account for a large proportion of various industrial instruments, and they are difficult to identify. With the development of image processing and artificial intelligence technology, it has become possible for cameras to automatically obtain real-time image data of instruments and automatically identify them visually, and engineers no longer need to conduct on-site inspections. There are two main categories of existing pointer instrument visual recognition algorithms: visual recognition algorithms based on traditional image processing; visual recognition algorithms based on deep learning; and visual recognition algorithms based on the combination of deep learning and traditional image processing.

[0003] However, the above-mentioned visual recognition algorithm has high training cost, cumbersome deployment and configuration, and is unable to recognize instrument images with blurred scale values. Summary of the invention

[0004] The main purpose of the present invention is to solve the technical problems of the existing visual recognition algorithm, such as high algorithm training cost, cumbersome deployment and configuration, and inability to recognize instrument images with blurred scale values. A general pointer instrument visual recognition method comprises the following steps: Acquire an image to be recognized that includes the pointer instrument, a list of digital scale values ​​of the pointer instrument, and clockwise and counterclockwise configuration instructions, wherein the list of digital scale values ​​is a list arranged in order; Using a target detection algorithm, obtain digital scale coordinates corresponding to the digital scale value list; according to the digital scale coordinates, obtain the circular area at the center of the instrument panel and the center coordinates of the pointer instrument; Using a target threshold segmentation algorithm and based on the center coordinates, find the coordinates of the pointer tip of the pointer instrument; The reading value of the instrument pointer is calculated according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates.

[0005] As a preferred technical solution, the target detection algorithm is used to obtain digital scale coordinates corresponding to the digital scale value list, including: Training a target detection model, wherein the target detection model is a pointer instrument and a target detection model of a digital scale in the pointer instrument; Using the target detection model to detect the target in the image to be identified, and obtaining a digital scale frame and digital scale coordinates; Each digital scale coordinate is matched one by one with the input digital scale value list.

[0006] As a preferred technical solution, the step of making each digital scale coordinate correspond to the input digital scale value list one by one includes: According to the digital scale coordinates, all digital scale coordinates are sorted, and the result of the sorting is that all digital scale coordinates form a circular single chain; According to the characteristic that the distance between the start and end points of the pointer instrument scale is longer, the connection between the start and end points of the circular single chain is disconnected to obtain the order of all digital scale coordinates; According to the clockwise and counterclockwise configuration instructions and the order of all digital scale coordinates, each scale coordinate is matched one by one with the input instrument digital scale value list.

[0007] As a preferred technical solution, the target threshold segmentation algorithm is used and the coordinates of the pointer tip of the pointer instrument are found according to the center coordinates, including: Generate a mask image, where the effective area of ​​the mask image is the circular area in the center of the instrument panel; Perform threshold segmentation in the effective area to find the approximate area of ​​the pointer, perform straight line fitting in the approximate area, filter out interfering straight lines, and select the longest straight line as the pointer; Calculate the distance between the two end points of the straight line and the center of the circle, and determine the endpoint with the larger distance as the tip of the pointer to obtain the coordinates of the pointer tip.

[0008] As a preferred technical solution, the specific steps of filtering the interference straight line are: The interference straight line is filtered out by the distance threshold between the straight line and the center of the circular area. If the distance between a straight line and the center of the circular area is less than the set threshold, the straight line is excluded. Interference lines are filtered out through the line length threshold. If the length of a line is less than the set threshold, the line is excluded.

[0009] As a preferred technical solution, the method of calculating the reading value of the instrument pointer according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates includes: Calculate the angle between the digital scale, the pointer tip and the center of the instrument panel according to the digital scale coordinates, the center coordinates and the pointer tip coordinates; The instrument is rotated with the center of the instrument as the rotation center, so that the angle between the maximum scale value of the instrument and the line connecting the center of the instrument is zero, and multiple interval scale angle ranges are obtained; Calculate the digital scale interval to which the pointer tip points; Calculate the value of the instrument pointer based on the interval scale angle range, scale value range, and pointer angle.

[0010] As a preferred technical solution, the reading value calculation formula is: When the scale is sorted clockwise: When the scale is sorted counterclockwise: Among them, x, y is the interval scale angle range [x, y], m, n is the scale value range, and the angle the needle points to is v.

[0011] A second aspect of the present invention provides a universal pointer instrument visual recognition system, comprising: An acquisition unit, the acquisition unit is used to acquire an image to be recognized containing the pointer instrument, a list of digital scale values ​​of the pointer instrument, and clockwise and counterclockwise configuration instructions, wherein the list of digital scale values ​​is a list arranged in order; A center coordinate calculation unit, the center coordinate calculation unit is used to use a target detection algorithm to obtain digital scale coordinates corresponding to the digital scale value list; according to the digital scale coordinates, obtain the circular area at the center of the instrument panel and the center coordinates of the pointer instrument; A pointer tip coordinate calculation unit, the pointer tip coordinate calculation unit is used to find the pointer tip coordinates of the pointer instrument by using a target threshold segmentation algorithm and according to the center coordinates; A reading calculation unit is used to calculate the reading value of the instrument pointer according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates.

[0012] The third aspect of the present invention provides an electronic device, comprising: a memory and at least one processor, wherein instructions are stored in the memory, and the memory and the at least one processor are interconnected via lines; the at least one processor calls the instructions in the memory so that the electronic device executes the above-mentioned general pointer instrument visual recognition method.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned general pointer instrument visual recognition method.

[0014] The present invention has the following beneficial effects: Adaptive recognition: The object detection algorithm of the present invention can adaptively recognize digital scale coordinates without presetting fixed positions or modes. This makes the algorithm highly adaptable to different pointer instruments, different installation positions and angles.

[0015] Accurately locate the pointer tip: Through threshold segmentation and straight line fitting technology, the present invention can accurately find the approximate area of ​​the pointer, filter out interfering straight lines, and select the longest straight line as the pointer. This processing method effectively improves the accuracy and robustness of pointer recognition.

[0016] Intelligent calculation of readings: Based on the digital scale coordinates, center coordinates and pointer tip coordinates, the solution can calculate the reading of the instrument pointer. The intelligent calculation method of the present invention not only improves the accuracy of the reading, but also reduces the need for manual intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a general pointer instrument visual recognition method provided by an embodiment of the present invention; Figure 2 1 is a diagram showing the recognition result of a pointer instrument and a digital scale in the pointer instrument according to an embodiment of the present invention.

[0018] Figure 3 This is a recognition result diagram showing a one-to-one correspondence between each scale point and an input instrument digital scale value list in an embodiment of the present invention.

[0019] Figure 4 This is a threshold segmentation result diagram of the circular area in the center of the instrument panel according to an embodiment of the present invention.

[0020] Figure 5 This is a diagram showing the identification result of the tip of a pointer and a reading according to an embodiment of the present invention.

[0021] Figure 6 A schematic diagram of the structure of a universal pointer instrument visual recognition device provided in an embodiment of the present invention.

[0022] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The embodiment of the present invention provides a method and system for visual recognition of a universal pointer instrument. The method comprises: obtaining an image to be recognized containing the pointer instrument, a list of digital scale values ​​of the pointer instrument, and clockwise and counterclockwise configuration instructions, wherein the list of digital scale values ​​is a list arranged in order; using a target detection algorithm to obtain digital scale coordinates corresponding to the list of digital scale values; obtaining the circular area at the center of the instrument panel and the center coordinates of the pointer instrument according to the digital scale coordinates; using a target threshold segmentation algorithm and according to the center coordinates, finding the coordinates of the pointer tip of the pointer instrument; and calculating the reading value of the instrument pointer according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates. The method of the present invention can identify pointer instruments with unclear scales.

[0024] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] There are two main categories of existing pointer instrument visual recognition algorithms: visual recognition algorithms based on traditional image processing; visual recognition algorithms based on deep learning; and visual recognition algorithms based on the combination of deep learning and traditional image processing.

[0026] A. Visual recognition algorithms based on traditional image processing can usually only recognize images of pointer instruments obtained by fixed cameras, and a large number of configuration items need to be manually added when deploying the algorithm - for example, the frame coordinates and pointers of the pointer instrument dial in the fixed camera field of view (this requires manual capture of fixed camera data and manual measurement during deployment); when identifying pointer instruments, first directly obtain the pointer instrument area from the configured frame coordinates, and then obtain the pointer angle in the local area through binarization, straight line fitting and other methods; finally, the final instrument reading value is calculated based on the configured instrument starting scale angle, scale value range and other information. The algorithm is not suitable for situations where instrument data is captured by non-completely fixed cameras, such as fixed-point shooting of industrial inspection robots (each time the inspection is carried out, the fixed-point shooting will have a certain offset, and the configuration items will partially fail, resulting in poor recognition results).

[0027] B. Based on deep learning, the visual recognition algorithm first uses a convolutional neural network to detect pointer instruments in camera images; then uses the convolutional neural network to segment the pointer instrument area, that is, to find the pixel-level precision outline of the regional dial scale and the pixel-level precision outline of the pointer, and then calculates the output instrument reading value based on the current instrument scale value range and other information. However, this type of recognition algorithm requires manual annotation of pixel-level precision segmentation model training data, which requires a lot of labor costs; in order to achieve algorithm universality, the comprehensiveness of training data is also a major test (including various types of pointer instruments and different lighting conditions of various pointer instruments); in actual application, the clarity of the obtained instrument image also has certain requirements.

[0028] C. Based on the visual recognition algorithm combining deep learning with traditional image processing, the convolutional neural network is used to detect the pointer instrument in the camera image; the convolutional neural network is then used to detect the digital scale and pointer in the pointer instrument area, perform OCR on the digital scale area, perform binarization in the pointer detection frame, or combine linear fitting to find the pointer; the pointer position, scale position and scale value of each scale are combined to calculate the reading value of the pointer instrument. However, this method will face a fatal problem in actual application. When the camera field of view is affected by light, the OCR cannot achieve the expected accuracy, or the pointer instrument photographed by the camera is rotated at an angle, the entire algorithm will fail.

[0029] In the face of various problems of existing pointer instrument visual recognition algorithms, this patent proposes a novel visual recognition algorithm based on the combination of deep learning and traditional image processing. It is a low-cost, highly robust and universal pointer instrument visual recognition method.

[0030] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 The first embodiment of the visual recognition method of a universal pointer instrument in the embodiment of the present invention includes: Acquire an image to be recognized that includes the pointer instrument, a list of digital scale values ​​of the pointer instrument, and clockwise and counterclockwise configuration instructions, wherein the list of digital scale values ​​is a list arranged in order; Specifically, the input is: an image of a pointer instrument; a list of digital scale values ​​of the pointer instrument (the content is: arranging the digital scales of the pointer instrument from small to large to form a list); the coordinates of the pointer scales arranged from small to large on the dial are described in counterclockwise or clockwise order (for example Figure 2 for clockwise direction).

[0031] Using a target detection algorithm, obtain digital scale coordinates corresponding to the digital scale value list; according to the digital scale coordinates, obtain the circular area at the center of the instrument panel and the center coordinates of the pointer instrument; Specifically, the target detection algorithm is used to obtain the digital scale coordinates corresponding to the digital scale value list, including: A target detection model is trained, wherein the target detection model is a pointer instrument, and a target detection model of a digital scale in the pointer instrument; the target detection model is used to detect the target in the image to be identified, and a digital scale border and digital scale coordinates are obtained; each digital scale coordinate is matched one by one with the input digital scale value list.

[0032] First, train the object detection model for pointer instruments and digital scales in pointer instruments (data enhancement of horizontal mirroring of images is required during training). Then, when identifying pointer instruments, use this model to detect objects in the target image. The recognition results are shown in Figure 2 .

[0033] Specifically, each digital scale coordinate is matched one by one with the input digital scale value list, including: According to the digital scale coordinates, all digital scale coordinates are sorted, and the result of the sorting is: all digital scale coordinates form a circular single chain; according to the characteristic that the starting point and the end point of the pointer instrument scale are longer, the connection between the starting point and the end point scale points in the circular single chain is disconnected to obtain the sorting of all digital scale coordinates; according to the clockwise and counterclockwise configuration instructions and the sorting of all digital scale coordinates, each scale coordinate is matched one by one with the input instrument digital scale value list.

[0034] For example, the digital scale boxes may be sorted using Dijkstra's algorithm, and each scale box may correspond one to one with the input digital scale value list of the instrument.

[0035] Get all the digital scale borders in the detected pointer instrument area, and find their center coordinates to represent each digital scale. Use Dijkstra algorithm to sort the digital scale center coordinates. The sorting result is: all scale coordinates form a circular single chain (such as connecting the six scale points in the 'pointer instrument recognition result' to form a closed area with the smallest perimeter); then, combined with the characteristics of the pointer instrument scale, disconnect the connection between the starting and ending scale points in the circular single chain. Combined with the input configuration instructions of whether the current instrument pointer scale is arranged counterclockwise or clockwise, each scale point is matched one by one with the input instrument digital scale value list. Recognition results are shown in Figure 3 .

[0036] Using a target threshold segmentation algorithm and based on the center coordinates, find the coordinates of the pointer tip of the pointer instrument; Specifically, the method of using a target threshold segmentation algorithm and finding the coordinates of the pointer tip of the pointer instrument according to the center coordinates includes: Generate a mask image, the effective area of ​​the mask image is the circular area in the center of the instrument panel; perform threshold segmentation in the effective area to find the approximate area of ​​the pointer, perform straight line fitting in the approximate area, filter out interfering straight lines, and select the longest straight line as the pointer; calculate the distance between the two end points of the straight line and the center of the circle, and judge the endpoint with a larger distance as the tip of the pointer to obtain the coordinates of the pointer tip.

[0037] Fit a circular area in the center of the instrument panel using the numeric scale positions: The circular area in the center of the instrument panel is fitted using the digital scale position, so as to obtain the radius of the circular area in the center of the instrument panel and the coordinates of the center point of the circle.

[0038] Perform target threshold segmentation in the circular area at the center of the instrument panel to find the pointer and the coordinates of the pointer tip: Generate a mask image, the effective area is the circular area in the center of the instrument panel, and perform threshold segmentation on the effective area of ​​the mask image corresponding to the original image (see Figure 4 ).

[0039] Specifically, the specific steps of filtering the interference straight line are: Interference lines are filtered out by the distance threshold between the line and the center of the circular area. If the distance between a line and the center of the circular area is less than the set threshold, the line is excluded. Interference lines are filtered out by the length threshold of the line. If the length of the line is less than the set threshold, the line is excluded.

[0040] Find the approximate area of ​​the pointer and fit a straight line in the approximate pointer area. Filter out interfering straight lines through the distance threshold between the straight line and the center of the circular area (set this threshold to 0.1 times the radius of the instrument center circle during implementation) and the length threshold of the straight line (set this threshold to 0.8 times the radius of the instrument center circle during implementation). Finally, select the longest straight line as the pointer. Finally, calculate the distance between the two end points of the straight line and the center of the circle, and determine the end point with the larger distance as the tip of the pointer (see Figure 5 ).

[0041] The reading value of the instrument pointer is calculated according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates.

[0042] Specifically, the calculating the reading value of the instrument pointer according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates includes: According to the digital scale coordinates, the center coordinates, and the pointer tip coordinates, the angles of the digital scale, the pointer tip, and the line connecting the center of the instrument panel are calculated; the instrument is rotated with the center of the instrument as the rotation center so that the angle between the maximum scale value of the instrument and the line connecting the center of the instrument is zero degrees, and multiple interval scale angle ranges are obtained; the digital scale interval to which the pointer tip points belongs is calculated; and the reading value of the instrument pointer is calculated in combination with the interval scale angle range, the scale value range, and the angle of the pointer.

[0043] The calculation formula for the reading value is: When the scale is sorted clockwise: When the scale is sorted counterclockwise: Among them, x, y is the interval scale angle range [x, y], m, n is the scale value range, and the angle the needle points to is v.

[0044] The details are as follows: Calculate the angle between each scale, the pointer tip and the center point of the circle (angle specification in polar coordinate system), the angle range is: [0°, 360°).

[0045] Calculate the digital scale interval of the direction pointed by the pointer tip Based on the result of the previous step, calculate the digital scale interval [m,n] to which the pointer tip points.

[0046] H. Calculate the value of the instrument pointer based on the interval scale angle range [x,y], scale value range [m,n], and pointer angle. Let the angle pointed by the pointer be v, Since the angle value is a polar coordinate system specification, the angle values ​​bound by the scale value cannot be directly added or subtracted. To facilitate the calculation between angle values, take the same example as above (the pointer scale values ​​are arranged from large to small, and the digital scale coordinates are arranged in a clockwise direction). Rotate the instrument with the center of the instrument as the rotation center so that the angle between the maximum scale value of the instrument and the line connecting the center of the instrument is zero degrees: Then the reading results: Assuming the pointer scale values ​​are arranged from large to small, the digital scale coordinates are sorted in a counterclockwise direction: Finally, the reading value of the instrument pointer is obtained. Figure 5 , which is 13.2.

[0047] In another embodiment, when the corresponding coordinates of the pointer scales are arranged from small to large in a clockwise direction on the dial, it is only necessary to horizontally mirror the image before step B "target detection to obtain the pointer instrument in the image, and the coordinates of the digital scale frame in the pointer instrument".

[0048] The above describes the visual recognition method of the universal pointer instrument in the embodiment of the present invention. The following describes the visual recognition device of the universal pointer instrument in the embodiment of the present invention. Figure 6 The first embodiment of the visual recognition device for universal pointer instruments in the embodiment of the present invention includes: An acquisition unit, the acquisition unit is used to acquire an image to be recognized containing the pointer instrument, a list of digital scale values ​​of the pointer instrument, and clockwise and counterclockwise configuration instructions, wherein the list of digital scale values ​​is a list arranged in order; A center coordinate calculation unit, the center coordinate calculation unit is used to use a target detection algorithm to obtain digital scale coordinates corresponding to the digital scale value list; according to the digital scale coordinates, obtain the circular area at the center of the instrument panel and the center coordinates of the pointer instrument; A pointer tip coordinate calculation unit, the pointer tip coordinate calculation unit is used to find the pointer tip coordinates of the pointer instrument by using a target threshold segmentation algorithm and according to the center coordinates; A reading calculation unit is used to calculate the reading value of the instrument pointer according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates.

[0049] Figure 7 7 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 700 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 710 (for example, one or more processors) and a memory 720, and one or more storage media 730 (for example, one or more mass storage devices) storing application programs 733 or data 732. Among them, the memory 720 and the storage medium 730 can be temporary storage or permanent storage. The program stored in the storage medium 730 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the electronic device 700. Furthermore, the processor 710 may be configured to communicate with the storage medium 730 to execute a series of instruction operations in the storage medium 730 on the electronic device 700.

[0050] The electronic device 700 may also include one or more power supplies 740, one or more wired or wireless network interfaces 750, one or more input and output interfaces 750, and / or one or more operating systems 731, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 7 The structure of the electronic device shown does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0051] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of a general pointer instrument visual recognition method.

[0052] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0053] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0054] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A general pointer instrument visual recognition method, characterized in that: The following steps are involved: Acquire an image to be recognized that includes the pointer instrument, a list of digital scale values ​​of the pointer instrument, and clockwise and counterclockwise configuration instructions, wherein the list of digital scale values ​​is a list arranged in order; Using a target detection algorithm, obtain digital scale coordinates corresponding to the digital scale value list; according to the digital scale coordinates, obtain the circular area at the center of the instrument panel and the center coordinates of the pointer instrument; Using a target threshold segmentation algorithm and based on the center coordinates, find the coordinates of the pointer tip of the pointer instrument; The reading value of the instrument pointer is calculated according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates.

2. A method for visually identifying a universal pointer instrument according to claim 1, characterized in that: The adopting of the target detection algorithm to obtain the digital scale coordinates corresponding to the digital scale value list includes: Training a target detection model, wherein the target detection model is a pointer instrument and a target detection model of a digital scale in the pointer instrument; Using the target detection model to detect the target in the image to be identified, and obtaining a digital scale frame and digital scale coordinates; Each digital scale coordinate is matched one by one with the input digital scale value list.

3. A method for visually identifying a universal pointer instrument according to claim 2, characterized in that: The step of making each digital scale coordinate correspond to the input digital scale value list one by one includes: According to the digital scale coordinates, all digital scale coordinates are sorted, and the result of the sorting is that all digital scale coordinates form a circular single chain; According to the characteristic that the distance between the start and end points of the pointer instrument scale is longer, the connection between the start and end points of the circular single chain is disconnected to obtain the order of all digital scale coordinates; According to the clockwise and counterclockwise configuration instructions and the order of all digital scale coordinates, each scale coordinate is matched one by one with the input instrument digital scale value list.

4. A method for visually identifying a universal pointer instrument according to claim 1, characterized in that: The method of using a target threshold segmentation algorithm and finding the coordinates of the pointer tip of the pointer instrument according to the center coordinates includes: Generate a mask image, where the effective area of ​​the mask image is the circular area in the center of the instrument panel; Perform threshold segmentation in the effective area to find the approximate area of ​​the pointer, perform straight line fitting in the approximate area, filter out interfering straight lines, and select the longest straight line as the pointer; Calculate the distance between the two end points of the straight line and the center of the circle, and determine the endpoint with the larger distance as the tip of the pointer to obtain the coordinates of the pointer tip.

5. A method for visually identifying a universal pointer instrument according to claim 4, characterized in that: The specific steps of filtering interference straight lines are: The interference straight line is filtered out by the distance threshold between the straight line and the center of the circular area. If the distance between a straight line and the center of the circular area is less than the set threshold, the straight line is excluded. Interference lines are filtered out through the line length threshold. If the length of a line is less than the set threshold, the line is excluded.

6. A method for visually identifying a universal pointer instrument according to claim 1, characterized in that: Calculating the reading value of the instrument pointer according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates includes: Calculate the angle between the digital scale, the pointer tip and the center of the instrument panel according to the digital scale coordinates, the center coordinates and the pointer tip coordinates; The instrument is rotated with the center of the instrument as the rotation center, so that the angle between the maximum scale value of the instrument and the line connecting the center of the instrument is zero, and multiple interval scale angle ranges are obtained; Calculate the digital scale interval to which the pointer tip points; Calculate the value of the instrument pointer based on the interval scale angle range, scale value range, and pointer angle.

7. A method for visually identifying a universal pointer instrument according to claim 6, characterized in that: The calculation formula of the reading value is: When the scale is sorted clockwise: When the scale is sorted counterclockwise: Among them, x, y is the interval scale angle range [x, y], m, n is the scale value range, and the angle the needle points to is v.

8. A universal pointer instrument visual recognition system, characterized in that: The computer bypass monitoring system comprises: An acquisition unit, the acquisition unit is used to acquire an image to be recognized containing the pointer instrument, a list of digital scale values ​​of the pointer instrument, and clockwise and counterclockwise configuration instructions, wherein the list of digital scale values ​​is a list arranged in order; A center coordinate calculation unit, the center coordinate calculation unit is used to use a target detection algorithm to obtain digital scale coordinates corresponding to the digital scale value list; according to the digital scale coordinates, obtain the circular area at the center of the instrument panel and the center coordinates of the pointer instrument; A pointer tip coordinate calculation unit, the pointer tip coordinate calculation unit is used to find the pointer tip coordinates of the pointer instrument by using a target threshold segmentation algorithm and according to the center coordinates; A reading calculation unit is used to calculate the reading value of the instrument pointer according to the digital scale coordinates, the center coordinates, and the pointer tip coordinates.

9. An electronic device, comprising a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory so that the electronic device executes each step of the universal pointer instrument visual recognition method as described in any one of claims 1-7.

Images