Empty box air pressure indication value image acquisition system, recognition system and working method thereof

Abstract

The present invention discloses an empty box air pressure indication value image acquisition system, an identification system and a working method thereof. The empty box air pressure indication value image acquisition system of the present invention comprises a synchronous belt module, an industrial camera and a stepping motor, which can be effectively connected with the empty box air pressure indication value identification system and provide an image acquisition platform for the empty box air pressure indication value identification system. The empty box air pressure indication value identification system provided by the present invention comprises an image acquisition system and also comprises a high-speed switch, a host computer and a slave computer, and constructs a platform for indicating value identification of images collected by the empty box air pressure indication value image acquisition system. Through the platform and further integrating the working method of the empty box air pressure indication value identification system provided by the present invention, the indication information of the empty box air pressure gauge is obtained.

Description

Technical Field

[0001] The invention relates to an image acquisition system and a recognition system for indicating the value of air pressure in an empty box and a working method thereof, belonging to the technical field of machine vision and image processing equipment. Background Art

[0002] In the measurement industry and industrial production, it is often necessary to manually read the empty box pressure gauge. At present, the reading work is mainly done manually, which has the problems of large reading errors and low work efficiency. Using image processing and machine vision technology, we design an empty box pressure indication value recognition system, which can realize automatic empty box pressure gauge high-precision image acquisition and indication recognition. The whole process does not require manual participation, which is the problem we need to solve. Summary of the invention

[0003] The present invention provides an empty box air pressure indication value image acquisition system for constructing an image acquisition platform required by an empty box air pressure indication value recognition system; provides an empty box air pressure indication value recognition system for realizing connection with the empty box air pressure indication value image acquisition system and providing a platform for empty box air pressure indication value recognition; provides a working method of the empty box air pressure indication value recognition system for obtaining the pointer indication of an empty box air pressure gauge.

[0004] The technical solution of the present invention is: an empty box air pressure representation value image acquisition system, comprising a synchronous belt module 1, an industrial camera 2, and a stepper motor 5; the position of the synchronous belt module 1 in the X, Y two-dimensional plane is controlled by rotating two stepper motors 5; the position of the industrial camera 2 is changed by changing the position of the synchronous belt module 1 in the X, Y two-dimensional plane.

[0005] The first stepper motor 5 drives the slide of the X-direction synchronous belt module 1 to move, so that the Y-direction synchronous belt module 1 installed on the X-direction synchronous belt module 1 moves along the X direction; the second stepper motor 2 drives the slide of the Y-direction synchronous belt module 1 to move, so that the industrial camera 2 installed on the slide of the Y-direction synchronous belt module 1 moves along the Y direction.

[0006] The synchronous belt modules 1 are three, two of which are arranged in the X direction, and one is arranged in the Y direction. The installation method is as follows: the two X-direction synchronous belt modules 1 are installed in parallel above the supporting component 13, and the two X-direction synchronous belt modules 1 are connected as a whole through a coupling 6 and a connecting rod. The first stepper motor 5 is installed at the end of an X-direction synchronous belt module 1. The first stepper motor 5 rotates to drive the synchronous belts of the two X-direction synchronous belt modules 1 to rotate. The rotation of the synchronous belts of the two X-direction synchronous belt modules 1 drives the slide of the X-direction synchronous belt module 1 to move. The movement of the slide of the X-direction synchronous belt module 1 drives the Y-direction synchronous belt module 1 installed thereon to move along the X direction. The industrial camera 2 is installed on the Y-direction synchronous belt module 1 through the camera bracket 3 on the slide 4 of the Y-direction synchronous belt module 1. The bottoms of both ends of a Y-direction synchronous belt module 1 are respectively installed on two X-direction synchronous belt modules 1. The second stepper motor 5 is installed at the end of the Y-direction synchronous belt module 1. The second stepper motor 5 rotates to drive the synchronous belt of the Y-direction synchronous belt module 1 to rotate. The rotation of the synchronous belt of the Y-direction synchronous belt module 1 drives the slide 4 of the Y-direction synchronous belt module 1 to move. The industrial camera 2 is driven to move along the Y direction by the movement of the slide 4 on the Y-direction synchronous belt module 1.

[0007] An empty box air pressure indication value recognition system includes an image acquisition system, a high-speed switch 7, a host computer 8, and a slave computer 9; an industrial camera 2, the host computer 8, and the slave computer 9 are respectively connected to the high-speed switch 7 through a network cable, and the slave computer 9 is connected to a stepper motor 5 through an electric wire.

[0008] A working method of an empty box air pressure indication value recognition system, the method steps are as follows:

[0009] Step 1, place several empty box barometers 11 in the air pressure calibration box 12;

[0010] Step 2. Power on all devices and components in the system;

[0011] Step 3, perform communication and control tests on various devices and components, including industrial camera 2, stepper motor 5, high-speed switch 7, host computer 8, and slave computer 9;

[0012] Step 4: Perform zero-position calibration of industrial camera 2 to ensure that the center position of the industrial camera lens coincides with the physical center position of the field of view or is within the error range;

[0013] Step 5, obtain the center position information of each empty box barometer 11: the host computer 8 issues an image acquisition instruction, drives the industrial camera 2 to acquire an image and transmits it to the host computer 8, and the host computer 8 obtains the number and position information of the empty box barometers 11 in the field of view;

[0014] Step 6, the host computer 8 adjusts the position of the industrial camera 2 in turn according to the position information of each instrument in Step 5 to collect the positive view image of each empty box barometer 11 and transmit it to the host computer 8;

[0015] Step 7, after all the positive view images of the empty box barometers 11 are collected, the industrial camera 2 is reset to the center position of the industrial camera lens in Step 4;

[0016] Step 8, the host computer 8 performs indication recognition on the image data collected in Step 6 to recognize the pointer indication;

[0017] Step 9: The host computer 8 displays and stores the recognition results according to Step 8.

[0018] The error range is set to be: based on the physical center position of the field of view, the X and Y directions deviate from the position within ±30 pixels.

[0019] In the Step 4, the zero position calibration of the industrial camera 2 is specifically as follows:

[0020] S1. Place four circular marking points 10 at the bottom corners of the air pressure test box 11;

[0021] S2, the industrial camera 2 located at any position collects a frame of image and transmits it to the host computer 8;

[0022] S3, the upper computer 8 identifies the number and position of circular marking points in the image collected by S2, and sends the information to the lower computer 9, which drives the stepper motor 5 to work and control the position movement of the industrial camera 2;

[0023] S4. Repeat S2 and S3 until the center position of the lens of industrial camera 2 coincides with the physical center position of the field of view of the marking point, thus achieving zero calibration.

[0024] In the Step 5, the center position information of each empty box barometer 11 is obtained as follows:

[0025] After the position of P1 and industrial camera 2 is calibrated, the host computer 8 issues an image acquisition instruction. After receiving the instruction, the industrial camera 2 acquires the image and uploads it to the host computer 8. The host computer 8 performs mean filtering on the acquired image to eliminate noise in the image.

[0026] P2, the host computer 8 performs a power transformation operation on the result of P1 to expand the feature area and the background gray level;

[0027] P3, the host computer 8 performs threshold segmentation, edge detection and sharpening operations on the result of P2 to extract the basic contour information of the empty box barometer 11;

[0028] P4, the host computer 8 performs morphological operations on the results of P3, separates the contours of the barometers 11 of each empty box, and avoids adjacent barometers being connected to the same area;

[0029] P5 and the host computer 8 perform Hough circle detection on the result of P4 to obtain the coordinate information of the circle center.

[0030] In the Step 6, the process of collecting the front-view images of the empty box barometers 11 is as follows:

[0031] C1, the upper computer 8 sends the coordinate information of the center of each empty box barometer 11 in a field of view to the lower computer 9;

[0032] C2, the lower computer 9 converts the coordinate information into the rotation angle of the stepper motor 5 according to the received coordinate information, and sends the rotation angle signal to the two stepper motors 5 respectively;

[0033] C3. After receiving the rotation signal, the two stepper motors 5 respectively perform the rotation action, wherein one stepper motor 5 controls the rotation of the synchronous belt module 1 in the X direction, and the other stepper motor 5 controls the rotation of the synchronous belt module 1 in the Y direction, thereby realizing the position change of the industrial camera 2 in the X and Y two-dimensional planes;

[0034] C4, after the two stepper motors 5 rotate to the target position, they send a signal to the lower computer 9, and the lower computer 9 forwards the action completion signal to the upper computer 8;

[0035] C5. After receiving the signal, the host computer 8 sends an image acquisition instruction to the industrial camera 2;

[0036] C6. After receiving the image acquisition instruction, the industrial camera 2 performs image acquisition and sends the image to the host computer 8 for storage, so as to be used in the subsequent indication recognition process;

[0037] C7. Repeat C1 to C6 until the positive-view image acquisition of all the empty-box barometers 11 in the initial field of view is completed.

[0038] In the Step 8, the host computer 8 performs indication recognition on the front view image of an empty box barometer 11, specifically:

[0039] R1, the host computer 8 performs Hough circle detection on the collected image of a certain empty box barometer 11 in the positive view, and selects a radius to draw a white circle;

[0040] R2 and the host computer 8 perform threshold segmentation, sharpening, and contour detection operations on the R1 result;

[0041] R3, the host computer 8 performs a region of interest operation on the R2 result, segments the image of the single empty box barometer 11 facing the center of the industrial camera 2, and then rotates the image to the level of the thermometer according to the edge of the thermometer in the image;

[0042] R4 and the host computer 8 perform Gaussian filtering on the result of R3 to remove Gaussian noise, and then perform binarization processing;

[0043] R5 and the host computer 8 perform Laplace edge detection operation on the result of R3 to obtain the outline of the image content;

[0044] R6 and the host computer 8 perform Hough circle detection on the result of R5 to determine the center and radius of the dial;

[0045] R7 and host computer 8 perform Canny edge detection and Hough line detection operations on the result of R4, detect the pointer and draw the line;

[0046] R8, host computer 8 calculates the angle of the R7 result pointer straight line;

[0047] R9 and the upper computer 8 draw two concentric circles with different radii based on the result of R5 and the center and radius obtained by R6 and extract the scale line region of interest;

[0048] R10 and the host computer 8 perform binarization and contour extraction operations on the result of R9 to extract the longer scale lines;

[0049] R11, the host computer 8 performs corrosion and expansion operations on the result of R10, and connects the coordinates of the center point of the long scale line with the coordinates of the center of the circle, calculates and arranges the angles of the straight lines where each scale line is located and saves them;

[0050] R12, the upper computer 8 uses the result of R8 and the result of R11, and uses the pointer adjacent scale angle calculation method to calculate the pointer indication of the empty box barometer 11 image.

[0051] The beneficial effects of the present invention are as follows: the empty box air pressure indication value image acquisition system of the present invention can be effectively connected with the empty box air pressure indication value recognition system, and provide an image acquisition platform for the empty box air pressure indication value recognition system; the empty box air pressure indication value recognition system provided by the present invention constructs a platform for indicating value recognition of the image acquired by the empty box air pressure indication value image acquisition system, and the indication information of the empty box air pressure gauge is obtained through the platform and further integrated with the working method of the empty box air pressure indication value recognition system provided by the present invention. The working method realizes the positive angle image acquisition of each instrument through the target detection and positioning process, avoids the reading error introduced by the flexibility of the angle of view during the manual reading process, utilizes the digital image processing related algorithm to realize the pointer extraction and angle calculation, and simulates the manual reading method, and adopts the adjacent scale angle calculation method to obtain the pointer indication. Compared with the single angle calculation method, the algorithm avoids the error introduced by the uneven distribution of the dial scale, and further improves the indication recognition accuracy.

[0052] The entire invention utilizes digital image processing, embedded and machine vision technology to realize the automation of the existing manual calibration work of empty box barometers; the working process is highly automated, without the need for human participation, reducing the burden of manual calibration, the system modules have clear functions, clear division of labor, and a high degree of automation. Compared with manual readings, the error is small, the consistency is high, and the accuracy and efficiency of indication recognition are guaranteed; by setting up an upper computer and a lower computer, the entire invention has a good human-computer interaction interface, which improves the usability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the structure of the empty box air pressure indication value image acquisition system and indication value recognition system;

[0054] Figure 2 This is the zero calibration diagram for industrial cameras;

[0055] Figure 3 This is the P1 mean filtering result diagram;

[0056] Figure 4 This is the result of P2 power transformation;

[0057] Figure 5 It is the P3 basic contour extraction map;

[0058] Figure 6 This is the P4 separation contour result image;

[0059] Figure 7 This is the P5 color labeling result map;

[0060] Figure 8 is the orthographic image of an empty box barometer in the field of view;

[0061] Fig. 9 This is the white circle coverage of the instrument area at the positive viewing angle of R1;

[0062] Fig.10 This is the R2 contour detection result diagram;

[0063] Fig.11 This is the result of segmentation and image normalization of a single instrument of R3;

[0064] Fig.12 This is the Gaussian filtering and binarization result of a single instrument image of R4;

[0065] Fig.13 This is the Laplace edge detection result of a single instrument image of R5;

[0066] Fig.14 This is the Hough circle detection result of a single instrument image of R6;

[0067] Fig.15 This is the result diagram of the R7 pointer detection;

[0068] Fig.16 This is the result diagram of the R8 pointer angle calculation;

[0069] Fig.17 Extract result graph for R9 scale line;

[0070] Fig.18 Extract result graph for R10 long scale line;

[0071] Fig.19 It is the line connecting the long scale line of R11 and the center of the circle;

[0072] Fig. 20 This is the R12 indication identification result diagram;

[0073] Fig.21 A flowchart of the working method;

[0074] The numbers in the figure are: 1-synchronous belt module, 2-industrial camera, 3-camera bracket, 4-slide, 5-stepping motor, 6-coupling, 7-high-speed switch, 8-host computer, 9-slave computer, 10-circular marking point, 11-empty box barometer, 12-air pressure calibration box, 13-support component. DETAILED DESCRIPTION

[0075] The present invention will be further described below in conjunction with the accompanying drawings and embodiments, but the content of the present invention is not limited to the scope of the embodiments.

[0076] Example 1: Figure 1 As shown, an empty box air pressure representation value image acquisition system includes a synchronous belt module 1, an industrial camera 2, and a stepper motor 5; the position of the synchronous belt module 1 in the X, Y two-dimensional plane is controlled by rotating the two stepper motors 5; the position of the industrial camera 2 is changed by changing the position of the synchronous belt module 1 in the X, Y two-dimensional plane.

[0077] Furthermore, a first stepper motor 5 can be set to drive the slide movement of the X-direction synchronous belt module 1, so that the Y-direction synchronous belt module 1 installed on the X-direction synchronous belt module 1 moves along the X direction; the second stepper motor 2 drives the slide movement of the Y-direction synchronous belt module 1, so that the industrial camera 2 installed on the slide of the Y-direction synchronous belt module 1 moves along the Y direction.

[0078] Further, the number of the synchronous belt modules 1 can be three, two of which are arranged in the X direction and one in the Y direction, and the installation method is as follows: the two X-direction synchronous belt modules 1 are installed in parallel above the supporting component 13, and the two X-direction synchronous belt modules 1 are connected as a whole through a coupling 6 and a connecting rod, and the first stepper motor 5 is installed at the end of an X-direction synchronous belt module 1, and the first stepper motor 5 rotates to drive the synchronous belts of the two X-direction synchronous belt modules 1 to rotate, and the rotation of the synchronous belts of the two X-direction synchronous belt modules 1 drives the slide of the X-direction synchronous belt module 1 to move, and the movement of the slide of the X-direction synchronous belt module 1 drives the Y-direction synchronous belt module 1 installed thereon to move along the X direction, and the Y-direction synchronous belt module 1 is moved by the movement of the slide of the X-direction synchronous belt module 1. The synchronous belt module 1 moves along the X direction, driving the industrial camera 2 installed thereon to move along the X direction; the industrial camera 2 is installed on the slide 4 of the Y direction synchronous belt module 1 via the camera bracket 3, the bottoms of both ends of a Y direction synchronous belt module 1 are respectively installed on two X direction synchronous belt modules 1, and the two ends of the bottom are fixed to the slide of the X direction synchronous module by screws, and the second stepper motor 5 is installed at the end of the Y direction synchronous belt module 1. The rotation of the second stepper motor 5 drives the synchronous belt of the Y direction synchronous belt module 1 to rotate, and the rotation of the synchronous belt of the Y direction synchronous belt module 1 drives the slide 4 of the Y direction synchronous belt module 1 to move, and the movement of the slide 4 on the Y direction synchronous belt module 1 drives the industrial camera 2 to move along the Y direction. The entire image acquisition system is installed above the air pressure calibration box 12 through a supporting component 13. When the system is working, there are several empty box barometers 11 in the air pressure calibration box 12; the synchronous belt module includes a shaft, a track, a synchronous belt and a slide, the output shaft of the stepper motor and the shaft of the synchronous belt module are connected through a coupling, a synchronous belt is installed on the track, and a slide is installed on the synchronous belt.

[0079] An empty box air pressure indication value recognition system includes an image acquisition system, a high-speed switch 7, a host computer 8, and a slave computer 9; an industrial camera 2, the host computer 8, and the slave computer 9 are respectively connected to the high-speed switch 7 through a network cable, and the slave computer 9 is connected to a stepper motor 5 through an electric wire.

[0080] Furthermore, the industrial camera 2 may be a general industrial array camera, the port bandwidth of the high-speed switch 7 may be 1000 Mb / s or above, and the lower computer 9 may be an embedded development board with STM32 as the core.

[0081] The system of this application uses a specific industrial camera instead of others. This choice is a special design. The industrial camera has a unique high resolution and supports long-term continuous exposure, which allows the system to work continuously for a long time, thus better meeting the actual needs. At the same time, the industrial camera in this system is installed on the slide through the camera bracket, and the camera height can be flexibly changed to meet the needs of field of view size and focal length. The entire system can be used as an independent structure without being bound to a specific air pressure calibration box, making it applicable to various scenarios.

[0082] A working method of an empty box air pressure indication value recognition system, the method steps are as follows:

[0083] Step 1, place several empty box barometers 11 in the air pressure calibration box 12;

[0084] Step 2. Power on all devices and components in the system;

[0085] Step 3, conduct communication and control tests on various devices and components, including industrial camera 2, stepper motor 5, high-speed switch 7, host computer 8, and slave computer 9; (such as testing: whether the stepper motor is controlled and functions normally after the host computer sends control instructions; whether it can move normally up, down, left, and right, etc.);

[0086] Step 4: Perform zero-position calibration of industrial camera 2 to ensure that the center position of the industrial camera lens coincides with the physical center position of the field of view or is within the error range;

[0087] Step 5, obtain the center position information of each empty box barometer 11: the host computer 8 issues an image acquisition instruction, drives the industrial camera 2 to acquire an image and transmits it to the host computer 8, and the host computer 8 obtains the number and position information of the empty box barometers 11 in the field of view;

[0088] Step 6, the host computer 8 adjusts the position of the industrial camera 2 in turn according to the position information of each instrument in Step 5 to collect the positive view image of each empty box barometer 11 and transmit it to the host computer 8;

[0089] Step 7, after all the positive view images of the empty box barometers 11 are collected, the industrial camera 2 is reset to the center position of the industrial camera lens in Step 4;

[0090] Step 8, the host computer 8 performs indication recognition on the image data collected in Step 6 to recognize the pointer indication;

[0091] Step 9: The host computer 8 displays and stores the recognition results according to Step 8.

[0092] Furthermore, the error range can be set to: based on the physical center position of the field of view, the X and Y directions deviate from the position within ±30 pixels. (The selected industrial camera has 20 million pixels or more). This error range can make the comparison time more optimal while the image information collected by the camera has little effect on the accuracy of subsequent indication recognition.

[0093] Furthermore, in Step 4, the zero position calibration of the industrial camera 2 may be specifically set as follows:

[0094] S1. Place four circular marking points 10 at the bottom corners of the air pressure test box 11;

[0095] S2, the industrial camera 2 located at any position collects a frame of image and transmits it to the host computer 8;

[0096] S3, the upper computer 8 identifies the number and position of the circular marking points in the image collected by S2, and sends the information to the lower computer 9, which drives the stepper motor 5 to work and control the position movement of the industrial camera 2 (the movement here is a small movement, which can be set to move within a range of 20 pixels to make the movement effect better);

[0097] S4, repeat S2 and S3 until the industrial camera 2 captures an image such as Figure 2 As shown (the four corners in the figure are circular marking points, and the middle red dot is the physical center position of the field of view of the marking point, Figure 2 The red color is for clarity and can be replaced by other colors) until the center of the industrial camera 2 lens (center pixel) coincides with the physical center of the field of view of the marker point (the intersection of the lines connecting the diagonal circular marker points) to achieve zero calibration.

[0098] Furthermore, in Step 5, the center position information of each empty box barometer 11 can be obtained as follows:

[0099] After the position of P1 and industrial camera 2 is calibrated, the host computer 8 issues an image acquisition instruction. After receiving the instruction, the industrial camera 2 acquires the image and uploads it to the host computer 8. The host computer 8 performs mean filtering on the acquired image to eliminate the noise in the image. Figure 3 Results shown;

[0100] P2, the host computer 8 performs a power transformation operation on the result of P1 to expand the feature area and the background gray level; Figure 4 Results shown;

[0101] P3, the host computer 8 performs threshold segmentation, edge detection and sharpening operations on the result of P2 to extract the basic contour information of the empty box barometer 11; the result is as follows Figure 5 As shown;

[0102] P4, the host computer 8 performs morphological operations on the results of P3, separates the contours of the barometers 11 in each empty box, and avoids adjacent barometers being connected to the same area; the results are as follows Figure 6 As shown;

[0103] P5 and the host computer 8 perform Hough circle detection on the result of P4 to obtain the coordinate information of the circle center. For the subsequent positive view image acquisition, color labeling is completed. The result is as follows Figure 7 As shown in the figure, color marking can be used to improve the efficiency of human-computer interaction.

[0104] Furthermore, in Step 6, the front-view image acquisition process of each empty box barometer 11 can be set as follows:

[0105] C1, the upper computer 8 sends the coordinate information of the center of each empty box barometer 11 in a field of view to the lower computer 9;

[0106] C2, the lower computer 9 converts the coordinate information into the rotation angle of the stepper motor 5 according to the received coordinate information, and sends the rotation angle signal to the two stepper motors 5 respectively;

[0107] C3. After receiving the rotation signal, the two stepper motors 5 respectively perform the rotation action, wherein one stepper motor 5 controls the rotation of the synchronous belt module 1 in the X direction, and the other stepper motor 5 controls the rotation of the synchronous belt module 1 in the Y direction, thereby realizing the position change of the industrial camera 2 in the X and Y two-dimensional planes;

[0108] C4, after the two stepper motors 5 rotate to the target position, they send a signal to the lower computer 9, and the lower computer 9 forwards the action completion signal to the upper computer 8;

[0109] C5. After receiving the signal, the host computer 8 sends an image acquisition instruction to the industrial camera 2;

[0110] C6. After receiving the image acquisition instruction, the industrial camera 2 performs image acquisition and sends the image to the host computer 8 for storage, so as to be used in the subsequent indication recognition process;

[0111] C7. Repeat C1 to C6 until the positive-view image acquisition of all the empty-box barometers 11 in the initial field of view is completed.

[0112] Furthermore, in the Step 8, the host computer 8 can be set to perform indication recognition on the positive view image of an empty box barometer 11, specifically (taking the empty box barometer 11 in the lower right corner of the picture as the target instrument as an example):

[0113] R1, host computer 8 collects Figure 8 The image of an empty box barometer 11 (the empty box barometer 11 in the lower right corner of the picture is the target instrument) in the positive view shown in the figure performs Hough circle detection and selects a white circle with an appropriate radius; the result is as follows Fig. 9 As shown;

[0114] R2, host computer 8 performs threshold segmentation, sharpening, and contour detection operations on the result of R1; the result is as follows Fig.10 As shown;

[0115] R3, the host computer 8 performs the region of interest operation on the result of R2, and segments the image of the single empty box barometer 11 facing the center of the industrial camera 2, and then rotates the image to the level of the thermometer according to the edge of the thermometer in the image; the result is as follows Fig.11 As shown;

[0116] R4 and the host computer 8 perform Gaussian filtering on the result of R3 to remove Gaussian noise, and then perform binarization processing; the result is as follows Fig.12 As shown;

[0117] R5 and the host computer 8 perform Laplace edge detection on the result of R3 to obtain the outline of the image content; the result is as follows Fig.13 As shown;

[0118] R6 and host computer 8 perform Hough circle detection on the result of R5 to determine the center and radius of the dial. The result is as follows: Fig.14 As shown;

[0119] R7 and host computer 8 perform Canny edge detection and Hough line detection on the result of R4, detect the pointer and draw the line; the result is as follows Fig.15 As shown; in order to better display the effect, the pointer is marked in green;

[0120] R8, host computer 8 calculates the angle of the R7 result pointer straight line; taking the starting scale (indication 49) as the zero point, the result is as follows Fig.16 As shown;

[0121] R9, host computer 8, draws two concentric circles with different radii based on the result of R5 and the center and radius obtained by R6, and extracts the region of interest of the scale line; the result is as follows Fig.17 As shown;

[0122] R10 and host computer 8 perform binarization and contour extraction operations on the result of R9 to extract the longer scale lines; the result is as follows Fig.18 As shown;

[0123] R11, host computer 8 performs corrosion and expansion operation on the result of R10, connects the coordinates of the center point of the long scale line with the coordinates of the center of the circle, calculates and arranges the angles of the straight lines where each scale line is located and saves them; the result is as follows Fig.19 As shown;

[0124] R12, the host computer 8 uses the result of R8 and the result of R11 to calculate the pointer value of the empty box barometer 11 image using the pointer adjacent scale angle calculation method. The result is as follows Fig. 20 shown.

[0125] The specific implementation modes of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above implementation modes, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.

Claims

1. A working method of an empty box air pressure indication value recognition system, Features: The method steps are as follows: Step 1. Place a number of empty box barometers (11) in a barometric pressure test box (12); Step 2. Power on all devices and components in the system; Step 3, conduct communication and control tests on various devices and components, including industrial cameras (2), stepper motors (5), high-speed switches (7), host computers (8), and slave computers (9); Step 4: Perform zero-position calibration of the industrial camera (2) to ensure that the center position of the industrial camera lens coincides with the physical center position of the field of view or is within the error range; Step 5, obtaining the center position information of each empty box barometer (11): the host computer (8) issues an image acquisition command to drive the industrial camera (2) to acquire an image and transmit it to the host computer (8), and the host computer (8) obtains the number and position information of the empty box barometers (11) in the field of view; Step 6, the host computer (8) adjusts the position of the industrial camera (2) in turn according to the position information of each instrument in Step 5 to collect the positive-view image of each empty box pressure gauge (11) and transmits it to the host computer (8); Step 7, after the positive view images of all the empty box barometers (11) are collected, the industrial camera (2) is reset to the center position of the industrial camera lens in Step 4; Step 8, the host computer (8) performs indication value recognition on the image data collected in Step 6 to identify the pointer indication value; Step 9, the host computer (8) displays and stores the recognition results according to Step 8; In Step 6, the process of collecting the positive-view images of each empty box barometer (11) is as follows: C1, the upper computer (8) sends the coordinate information of the center of each empty box barometer (11) in the field of view to the lower computer (9); C2, the lower computer (9) converts the coordinate information into the rotation angle of the stepper motor (5) according to the received coordinate information, and sends the rotation angle signal to the two stepper motors (5) respectively; C3. After receiving the rotation signal, the two stepper motors (5) respectively perform rotation actions, wherein one stepper motor (5) controls the rotation of the synchronous belt module (1) in the X direction, and the other stepper motor (5) controls the rotation of the synchronous belt module (1) in the Y direction, thereby realizing the position change of the industrial camera (2) in the X and Y two-dimensional planes; C4. After the two stepper motors (5) rotate to the target position, they send a signal to the lower computer (9), and the lower computer (9) forwards the action completion signal to the upper computer (8); C5. After receiving the signal, the host computer (8) sends an image acquisition instruction to the industrial camera (2); C6, after receiving the image acquisition instruction, the industrial camera (2) performs image acquisition and sends the image to the host computer (8) for storage, for use in the subsequent indication recognition process; C7, repeat C1 to C6 until the positive-view image acquisition of all empty-box barometers (11) in the initial field of view is completed; The empty box air pressure indication value recognition system comprises an image acquisition system, and further comprises a high-speed switch (7), a host computer (8), and a slave computer (9); the industrial camera (2), the host computer (8), and the slave computer (9) are respectively connected to the high-speed switch (7) via a network cable, and the slave computer (9) is connected to the stepping motor (5) via an electric wire; The image acquisition system comprises a synchronous belt module (1), an industrial camera (2), and a stepper motor (5); the position of the synchronous belt module (1) is controlled within an X, Y two-dimensional plane by rotating two stepper motors (5); and the position of the industrial camera (2) is changed by changing the position of the synchronous belt module (1) within the X, Y two-dimensional plane.

2. The working method according to claim 1, Features: The first stepper motor (5) drives the slide of the X-direction synchronous belt module (1) to move, thereby enabling the Y-direction synchronous belt module (1) installed on the X-direction synchronous belt module (1) to move along the X-direction; the second stepper motor (2) drives the slide of the Y-direction synchronous belt module (1) to move, thereby enabling the industrial camera (2) installed on the slide of the Y-direction synchronous belt module (1) to move along the Y-direction.

3. The working method according to claim 1, Features: The synchronous belt modules (1) are three in number, two of which are arranged in the X direction and one in the Y direction. The installation method is as follows: the two X-direction synchronous belt modules (1) are installed in parallel above the support component (13); the two X-direction synchronous belt modules (1) are connected as a whole via a coupling (6) and a connecting rod; a first stepper motor (5) is installed at the end of an X-direction synchronous belt module (1); the first stepper motor (5) rotates to drive the synchronous belts of the two X-direction synchronous belt modules (1) to rotate; the rotation of the synchronous belts of the two X-direction synchronous belt modules (1) drives the slide of the X-direction synchronous belt module (1) to move; the movement of the slide of the X-direction synchronous belt module (1) drives the Y-direction synchronous belt module (1) installed thereon to move along the X direction; The Y-direction synchronous belt module (1) moves along the X-direction, driving the industrial camera (2) mounted thereon to move along the X-direction; the industrial camera (2) is mounted on the slide table (4) of the Y-direction synchronous belt module (1) via a camera bracket (3); the bottoms of both ends of a Y-direction synchronous belt module (1) are respectively mounted on two X-direction synchronous belt modules (1); a second stepper motor (5) is mounted at the end of the Y-direction synchronous belt module (1); the second stepper motor (5) rotates to drive the synchronous belt of the Y-direction synchronous belt module (1) to rotate; the rotation of the synchronous belt of the Y-direction synchronous belt module (1) drives the slide table (4) of the Y-direction synchronous belt module (1) to move; and the industrial camera (2) moves along the Y-direction through the movement of the slide table (4) on the Y-direction synchronous belt module (1).

4. The working method according to claim 1, Features: The error range is set to: based on the physical center position of the field of view, the X and Y directions deviate from the position within ±30 pixels.

5. The working method according to claim 1, Features: In the Step 4, the zero position calibration of the industrial camera (2) is specifically as follows: S1. Place four circular marking points (10) at the corners of the bottom surface of the air pressure test box (11); S2, an industrial camera (2) located at any position captures a frame of image and transmits it to a host computer (8); S3, the upper computer (8) identifies the number and position of circular marking points in the image captured by S2, and sends the information to the lower computer (9), and the lower computer (9) drives the stepper motor (5) to work and thus controls the position movement of the industrial camera (2); S4. Repeat S2 and S3 until the center position of the lens of the industrial camera (2) coincides with the physical center position of the field of view of the marking point, thus achieving zero calibration.

6. The working method according to claim 1, Features: In Step 5, the center position information of each empty box air pressure gauge (11) is obtained as follows: After the position of P1 and the industrial camera (2) is calibrated, the host computer (8) issues an image acquisition instruction. After receiving the instruction, the industrial camera (2) acquires the image and uploads it to the host computer (8). The host computer (8) performs mean filtering on the acquired image to eliminate noise in the image. P2, the host computer (8) performs a power transformation operation on the result of P1 to expand the feature area and the background grayscale level; P3 and the host computer (8) perform threshold segmentation, edge detection and sharpening operations on the results of P2 to extract the basic contour information of the empty box barometer (11); P4, the host computer (8) performs morphological operations on the results of P3 to separate the contours of the barometers (11) of each empty box to prevent adjacent barometers from being connected to the same area; P5 and the host computer (8) perform Hough circle detection on the result of P4 to obtain the coordinate information of the circle center.

7. The working method according to claim 1, Features: In the Step 8, the host computer (8) performs indication recognition on the front view image of an empty box pressure gauge (11), specifically: R1, the host computer (8) performs Hough circle detection on the collected image of a certain empty box barometer (11) in the positive view, and selects a radius to draw a white circle; R2 and the host computer (8) perform threshold segmentation, sharpening, and contour detection operations on the results of R1; R3, the host computer (8) performs a region of interest operation on the result of R2, segments the image of the single empty box barometer (11) facing the center of the industrial camera (2), and then rotates the image to the level of the thermometer according to the edge of the thermometer in the image; R4 and the host computer (8) perform Gaussian filtering on the result of R3 to remove Gaussian noise, and then perform binarization processing; R5 and the host computer (8) perform Laplace edge detection on the result of R3 to obtain the outline of the image content; R6 and the host computer (8) perform Hough circle detection on the result of R5 to determine the center and radius of the dial; R7 and the host computer (8) perform Canny edge detection and Hough line detection on the result of R4, detect the pointer and draw the line; R8, the host computer (8) calculates the angle of the R7 result pointer straight line; R9, the host computer (8) draws two concentric circles with different radii based on the result of R5 and the center and radius obtained by R6 and extracts the scale line area of ​​interest; R10 and the host computer (8) perform binarization and contour extraction operations on the result of R9 to extract the longer scale lines; R11, the host computer (8) performs corrosion and expansion operations on the result of R10, connects the coordinates of the center point of the long scale line with the coordinates of the center of the circle, calculates and arranges the angles of the straight lines where each scale line is located and saves them; R12, the host computer (8) uses the result of R8 and the result of R11 to calculate the pointer indication of the empty box barometer (11) image using the pointer adjacent scale angle calculation method.

Images