Relay actuation time analysis method based on target matching pursuit

Abstract

The application relates to the technical field of relays, and discloses a relay contact action time analysis method based on target matching tracking, which is used for improving the accuracy when analyzing the relay contact action time. The method comprises the following steps: performing power-on testing on a target relay, collecting contact images of the target relay in the power-on testing process, and obtaining a contact image set; pre-processing the contact image set to obtain a processed image set; performing feature recognition on each processed image in the processed image set to obtain contact position information in each processed image in the processed image set; and performing contact action time analysis on the target relay based on the contact position information in each processed image in the processed image set to determine a target contact action time.

Description

Technical Field

[0001] This invention relates to the field of relay technology, and in particular to a relay actuation time analysis method based on target matching and tracking. Background Technology

[0002] The timing parameters of electrical appliances are important indicators reflecting their operating characteristics and reliability. Accurate measurement and calculation of these parameters not only help analyze the failure mechanisms of relays but also provide crucial data support for performance analysis and lifespan prediction. Actuation time, a critical parameter reflecting reliability, refers to the time from energizing the excitation circuit coil to the start of action of the relay's moving contact.

[0003] The speed at which a relay actuates affects the safe operation of the entire system. For example, when normally closed and normally open relays using the same moving contact to disconnect one circuit and connect another, it often happens that one circuit hasn't had time to completely disconnect before the other circuit closes. If both circuits close simultaneously, a serious short circuit will occur. Therefore, the accuracy of the relay's actuation time measurement is crucial. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a relay triggering time analysis method based on target matching and tracking, which solves the technical problem of low accuracy in relay triggering time analysis.

[0005] This invention provides a relay actuation time analysis method based on target matching and tracking, comprising: performing an energization test on a target relay, and simultaneously acquiring contact images of the target relay during the energization test to obtain a set of contact images; preprocessing the set of contact images to obtain a set of processed images; performing feature recognition on each processed image in the set of processed images to obtain contact position information in each processed image in the set of processed images; and performing actuation time analysis on the target relay based on the contact position information in each processed image in the set of processed images to determine the target actuation time.

[0006] In this invention, the step of preprocessing the set of touch images to obtain a processed image set includes: performing grayscale processing on the set of touch images to obtain a grayscale image set; performing image filtering processing on the grayscale image set to obtain a filtered image set; and performing binarization processing on the filtered image set to obtain a processed image set.

[0007] In this invention, the step of performing feature recognition on each processed image in the processed image set to obtain touch point position information in each processed image in the processed image set includes: performing feature recognition on each processed image in the processed image set to determine the rectangular frame position information of each processed image;

[0008] By analyzing the position information of the rectangular frame of each processed image, the touch point position information in each processed image in the processed image set is determined.

[0009] In this invention, the step of performing feature recognition on each processed image in the processed image set to determine the bounding box position information of each processed image includes: performing touch point region analysis on each processed image in the processed image set to determine candidate touch point regions for each processed image; performing line detection on the candidate touch point regions of each processed image to determine multiple lines in the candidate touch point regions of each processed image; and performing bounding box position analysis on the processed image set using the multiple lines in the candidate touch point regions of each processed image to determine the bounding box position information of each processed image.

[0010] In this invention, the step of performing touch point position analysis on the set of processed images using the bounding box position information of each processed image to determine the touch point position information in each processed image of the set of processed images includes: performing center point position parameter analysis on the bounding box position information of each processed image to determine a set of position parameters; and performing touch point position analysis on the set of position parameters to determine the touch point position information in each processed image of the set of processed images.

[0011] In this invention, the step of analyzing the touch time of the target relay based on the touch point position information in each processed image of the processed image set to determine the target touch time includes: performing position offset analysis based on the touch point position information in each processed image of the processed image set to determine the position offset analysis result; performing target image matching based on the position offset analysis result to determine the target image set; and performing touch time analysis on the target relay based on the target image set to determine the target touch time.

[0012] In this invention, the step of analyzing the triggering time of the target relay using the target image set to determine the target triggering time includes: performing time-series feature analysis on the target image set to determine a time-series feature set; and performing triggering time analysis on the target relay using the time-series feature set to determine the target triggering time.

[0013] In this invention, a target relay is subjected to an energization test, and simultaneously, contact images of the target relay during the energization test are acquired to obtain a set of contact images. The set of contact images is preprocessed to obtain a set of processed images. Feature recognition is performed on each processed image in the set of processed images to obtain contact position information in each processed image. Based on the contact position information in each processed image, the trigger time of the target relay is analyzed to determine the target trigger time. This invention indirectly achieves the measurement of trigger time using image processing technology based on target point feature capture and tracking. Its application enables accurate tracking of moving contacts throughout the entire movement process. By processing the parameters obtained from the tracking, the final trigger time value is obtained. The process is more intuitive and clear, eliminating the influence of human subjectivity and further improving the accuracy of relay trigger time analysis. Attached Figure Description

[0014] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0015] Figure 1 This is a flowchart of the relay triggering time analysis method based on target matching and tracking in an embodiment of the present invention.

[0016] Figure 2 This is a flowchart illustrating feature recognition for each processed image in the processed image set in an embodiment of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0019] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0020] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 , Figure 1 This is a flowchart of the relay triggering time analysis method based on target matching and tracking according to an embodiment of the present invention, as shown below. Figure 1 As shown, the flowchart includes the following steps:

[0021] S101. Perform an energization test on the target relay, and simultaneously acquire contact images of the target relay during the energization test to obtain a set of contact images;

[0022] Specifically, adjust the high-speed camera lens so that the lens axis is on the same horizontal line as the front of the target relay, and ensure that the target relay is displayed in the center of the entire page to avoid pixel value errors caused by angle. Install and fix the target relay on the relay base. The base not only provides power but also provides stability to prevent the target relay from being unstable during shooting, which could lead to inaccurate target recognition and affect the final result. Appropriately adjust the light intensity and exposure to achieve the ideal image of the front of the relay. The ring light source ensures uniform light distribution, making it easier to highlight the edges and height changes of the moving contacts, highlighting parts that are originally difficult to see. Then, conduct a power-on test on the target relay and simultaneously collect contact images of the target relay during the power-on test to obtain a set of contact images.

[0023] Meanwhile, in this embodiment of the invention, a self-designed microcontroller-based synchronous generator is used to output excitation signal 1 and excitation signal 2. Excitation signal 1 is a high-low level electrical signal that forms a circuit with the 24V power supply and the solid-state relay, acting as a switch and generating the power supply signal for the relay coil to operate. The high level of excitation signal 1 is set to 1.5s, the low level to 1.5s, and the entire process is 3s. It should be noted that excitation signal 2 is a high-frequency clock signal used as an external trigger for image acquisition by the industrial high-speed camera, thereby realizing the synchronous triggering of image acquisition and relay coil power supply, so that the starting time of shooting is the same, and a complete motion sequence image is obtained, avoiding the time error caused by manual shooting. Therefore, the output time of excitation signal 2 (clock pulse) is set to 3 seconds and the frequency is 2200HZ / S, so that the shooting frequency of the shooting device is 2200 frames / second, and the image of the entire motion process of the relay front is captured and saved.

[0024] S102. Preprocess the set of contact images to obtain a processed set of images;

[0025] Specifically, the preprocessing includes image segmentation, image grayscale conversion, image filtering, image binarization, dilation, and erosion operations. In this embodiment of the invention, the server performs preprocessing on the touch point images to obtain a processed image set.

[0026] S103. Perform feature recognition on each processed image in the processed image set to obtain the touch point location information in each processed image in the processed image set;

[0027] Specifically, a rectangular feature recognition method is used. The built-in straight line detection function is used to identify multiple straight lines, determine the intersection of the obtained straight lines and their extensions, take out four intersection points to determine whether they are rectangles, set a threshold to remove rectangles that do not meet the requirements, and when the moving contact moves, the position parameter of the center point of the moving contact will change. Then, the starting time T2 of the moving contact is obtained by judging the position change.

[0028] S104. Based on the contact position information in each processed image in the processed image set, perform contact time analysis on the target relay to determine the target contact time.

[0029] In this embodiment of the invention, the initial moment T1 of the pulse output and the moment T2 when the moving contact begins to move are obtained by programming. The time difference between T2 and T1 is the triggering time of the relay.

[0030] By performing the above steps, the target relay is energized for testing. Simultaneously, contact images of the target relay during the energization test are acquired, resulting in a set of contact images. This set of contact images is then preprocessed to obtain a set of processed images. Feature recognition is performed on each processed image in the set to obtain contact position information. Based on the contact position information in each processed image, the target relay's activation time is analyzed to determine the target activation time. In this invention, image processing technology based on target point feature capture and tracking is used to indirectly measure the activation time. Its application enables accurate tracking of the moving contact throughout the entire motion process. By processing the parameters obtained from the tracking, the activation time value is finally obtained. The process is more intuitive and clear, eliminating the influence of human subjectivity and further improving the accuracy of relay activation time analysis.

[0031] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0032] (1) Perform grayscale processing on the set of touch point images to obtain a set of grayscale images;

[0033] (2) Perform image filtering on the grayscale image set to obtain a filtered image set;

[0034] (3) Binarize the filtered image set to obtain the processed image set.

[0035] Specifically, converting the set of touch point images into a set of grayscale images can be achieved using common algorithms such as the averaging method, weighted averaging method, and maximum value method. These methods convert the values ​​of the RGB three channels into single-channel grayscale values. Filtering is then applied to the grayscale image set to remove noise and enhance image features. Common filtering algorithms include Gaussian filtering, median filtering, and mean filtering. The resulting grayscale image set is then binarized, with pixels having grayscale values ​​exceeding a set threshold set as white (or foreground) and pixels below the threshold set as black (or background). This process ensures that each image contains only two different colors of pixels, facilitating subsequent image analysis and processing.

[0036] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0037] (1) Perform position offset analysis based on the touch point position information in each processed image in the processed image set, and determine the position offset analysis result;

[0038] (2) Match the target images using the position offset analysis results to determine the target image set;

[0039] (3) Analyze the triggering time of the target relay by using the target image set to determine the target triggering time.

[0040] Specifically, for the contact point position information of each processed image in the processed image set, position offset analysis is performed, that is, comparing the position changes of the same contact point in different images to determine the position offset analysis result; based on the position offset analysis result, target images are matched, that is, the image most similar to the target image is selected from the original image library to determine the target image set; in the target image set, for each target relay contact point, the contact point activation timestamp is extracted from multiple images, and the target activation time is determined by analyzing these timestamp data.

[0041] In one specific embodiment, the process of determining the target trigger time by analyzing the trigger time of the target relay using a set of target images can specifically include the following steps:

[0042] (1) Perform temporal feature analysis on the target image set to determine the temporal feature set;

[0043] (2) The triggering time of the target relay is determined by analyzing the triggering time of the target relay through the timing feature set.

[0044] Specifically, the timing feature analysis of the target image set is performed, that is, the timing data of each target image is extracted and analyzed to obtain the timing feature set of the target contact, such as start time, duration, frequency, etc. Using the timing feature set, the contact time of each target relay is determined by analyzing its timing feature set.

[0045] In one specific embodiment, such as Figure 2 As shown, the process of executing step S103 can specifically include the following steps:

[0046] S201. Perform feature recognition on each processed image in the processed image set to determine the bounding box position information of each processed image;

[0047] S202. By analyzing the position information of the rectangular frame of each processed image, the touch point position information in each processed image in the processed image set is determined.

[0048] Specifically, feature recognition is performed on each processed image in the image set. This involves using computer vision techniques (such as deep learning and image segmentation) to detect and locate the target region in each image and determine the bounding box position information of each target region. Based on the bounding box position information of each processed image, touch point position analysis is performed within the target region. This involves using image processing techniques to extract and analyze the touch point position information inside each bounding box to determine the touch point position information in each processed image in the image set.

[0049] In one specific embodiment, the process of performing step S201 may specifically include the following steps:

[0050] (1) Perform touch point region analysis on each processed image in the processed image set to determine the candidate touch point region for each processed image;

[0051] (2) Perform line detection on the candidate touch area of ​​each processed image to determine multiple lines in the candidate touch area of ​​each processed image;

[0052] (3) Perform rectangular frame position analysis on the set of processed images by multiple straight lines in the candidate touch point area of ​​each processed image to determine the rectangular frame position information of each processed image.

[0053] Specifically, touch point region analysis is performed on each processed image in the image set. Image processing techniques (such as edge detection, binarization, morphological operations, etc.) are used to extract and segment candidate touch point regions in the images to determine the candidate touch point regions for each processed image. For the candidate touch point regions of each processed image, line detection is performed, that is, multiple lines in the candidate touch point regions are extracted using algorithms such as Hough transform. Based on the information of multiple lines in the candidate touch point regions, the position of the bounding box is analyzed for each processed image, that is, the position information of the bounding box containing the touch point is determined based on the relationship and position of the lines.

[0054] In one specific embodiment, the process of performing step S202 may specifically include the following steps:

[0055] (1) Analyze the center point position parameters of the rectangular box position information of each processed image to determine the set of position parameters;

[0056] (2) Perform touch point position analysis on the processed image set through the position parameter set to determine the touch point position information in each processed image in the processed image set.

[0057] Specifically, for the bounding box position information of each processed image in the image set, center point position parameter analysis is performed, that is, the center point coordinates of each bounding box are calculated and these coordinates are used as a set of position parameters. Using the set of position parameters, for each processed image, touch point position analysis is performed within its corresponding bounding box. That is, based on the center point coordinates in the set of position parameters, the touch point position information in the bounding box is accurately located and extracted to determine the touch point position information in each processed image in the image set.

[0058] The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A method for analyzing relay actuation time based on target matching and tracking, characterized in that, The methods include: The target relay is subjected to an energization test, and at the same time, contact images of the target relay during the energization test are acquired to obtain a set of contact images; The set of touch images is preprocessed to obtain a processed set of images; Feature recognition is performed on each processed image in the processed image set to obtain the touch point location information in each processed image in the processed image set; For each processed image in the set of processed images, feature recognition is performed to determine the bounding box position information of each processed image; For each processed image in the processed image set, perform touch point region analysis to determine the candidate touch point region for each processed image; Line detection is performed on the candidate touch point regions of each processed image to determine multiple straight lines in the candidate touch point regions of each processed image; The bounding box position information of each processed image is determined by performing rectangular box position analysis on the set of processed images using multiple straight lines in the candidate touch point region of each processed image. By analyzing the position information of the rectangle in each processed image, the touch point position information in each processed image in the processed image set is determined. The target relay is analyzed for touch time based on the touch point position information in each processed image in the processed image set to determine the target touch time; Based on the touch point position information in each processed image in the processed image set, position offset analysis is performed to determine the position offset analysis result; The target image set is determined by matching the position offset analysis results. The target relay is subjected to trigger time analysis using the target image set to determine the target trigger time; Perform temporal feature analysis on the target image set to determine the temporal feature set; The target relay is activated by analyzing the activation time using the set of timing features to determine the target activation time.

2. The relay actuation time analysis method based on target matching and tracking according to claim 1, characterized in that, The step of preprocessing the set of touch images to obtain a processed image set includes: The set of touch images is converted to grayscale to obtain a set of grayscale images; The grayscale image set is subjected to image filtering processing to obtain a filtered image set; The filtered image set is binarized to obtain the processed image set.

3. The relay actuation time analysis method based on target matching and tracking according to claim 1, characterized in that, The step of analyzing the touch point positions of the processed image set using the bounding box position information of each processed image to determine the touch point position information in each processed image of the processed image set includes: For each processed image, the position information of the bounding box is analyzed to determine the center point position parameter set. By analyzing the touch point positions of the processed image set using the set of position parameters, the touch point position information in each processed image of the processed image set is determined.

Images