Equipment counterweight detection method and device, computing equipment and machine readable storage medium

Abstract

The invention discloses an equipment counterweight detection method and device, computing equipment and a machine readable storage medium, and relates to the technical field of engineering machinery. The method comprises the steps that a counterweight area image of to-be-detected equipment is acquired, the counterweight area image is acquired by an image acquisition device fixedly installed on the to-be-detected equipment, and a stable data basis is provided for identification through image acquisition at a fixed visual angle; inputting the counterweight area image into a target detection model to obtain a plurality of different types of detection frames; screening out a balancing weight identification frame located in the area frame from the plurality of detection frames to serve as an effective balancing weight identification frame; and determining the equipment counterweight of the to-be-detected equipment based on the quantitative relationship between the effective counterweight block identification frame and the weight identification frame and the weight information corresponding to the weight identification frame. Based on a multi-source information fusion and cross validation mechanism, the reliability of a counterweight calculation result under various complex working conditions is ensured, reliable data support is provided for safe operation of equipment, and the safety accident risk is reduced.

Description

Technical Field

[0001] This application relates to the field of engineering machinery technology, specifically to a method, apparatus, computing device, and machine-readable storage medium for detecting equipment counterweight. Background Technology

[0002] In the field of crane safety monitoring and construction management, accurate and real-time acquisition of the entire machine's counterweight information is a crucial prerequisite for scientific load calculation and preventing overturning accidents. Currently, the industry generally relies on feature-matching-based visual algorithms or radio frequency identification (RFID) technology, which achieves counterweight safety monitoring by installing detected components or tags on the surface of the counterweight blocks. However, these methods exhibit weak anti-interference capabilities in complex engineering environments, and their recognition accuracy is easily affected by external factors such as changes in lighting, weather conditions, and partial obstruction, leading to a decrease in the reliability of counterweight recognition. Furthermore, existing devices often require the installation of dedicated detection components on each counterweight block, resulting in complex system deployment. Repeated equipment installation and debugging are necessary when replacing or adjusting the counterweight, increasing operating costs and causing inconvenience on-site, thus limiting the large-scale application and promotion of this technology in highly dynamic, multi-condition construction scenarios. Summary of the Invention

[0003] In view of the above-mentioned shortcomings of the prior art, the purpose of the embodiments of this application is to provide a method, apparatus, computing device and machine-readable storage medium for detecting equipment counterweight.

[0004] To achieve the above objectives, the first aspect of this application provides a method for detecting equipment counterweight, comprising: The image of the counterweight area of ​​the device to be tested is acquired by an image acquisition device that is fixedly installed on the device to be tested. The installation position and field of view of the image acquisition device are configured to cover the stacked area of ​​the counterweight blocks and acquire the identification information set on each counterweight block. The image of the counterweight region is input into the target detection model to obtain multiple detection boxes of different categories. The categories of detection boxes include region boxes, counterweight block identifier boxes, and weight identifier boxes. From multiple detection boxes, select the counterweight block identification boxes that are located within the area box as valid counterweight block identification boxes; Based on the quantitative relationship between the effective counterweight marking frames and the weight marking frames, as well as the weight information corresponding to the weight marking frames, the equipment counterweight of the device to be tested is determined.

[0005] In this embodiment of the application, the weight marking frame includes a first weight marking frame and a second weight marking frame with different marking information. Based on the quantitative relationship between the effective counterweight marking frames and the weight marking frames, and the weight information corresponding to the weight marking frames, the equipment counterweight of the device to be tested is determined, including: Determine whether the number of valid counterweight marker frames is consistent with the number of first weight marker frames; When the number of effective counterweight marking frames is the same as the number of first weight marking frames, the equipment counterweight of the device to be tested is determined based on the sum of the first weight information of each first weight marking frame. If the number of valid counterweight marking frames is inconsistent with the number of first weight marking frames, determine whether the number of valid counterweight marking frames is consistent with the number of second weight marking frames. When the number of effective counterweight marking frames is the same as the number of second weight marking frames, the equipment counterweight of the device to be tested is determined based on the sum of the second weight information of each second weight marking frame. If the number of valid counterweight marking frames is inconsistent with the number of second weight marking frames, an anomaly is identified.

[0006] In this embodiment of the application, the counterweight marking frame includes a top counterweight marking frame and a bottom counterweight marking frame, and the equipment counterweight detection method further includes: The upper boundary of the counterweight area is determined by the position of the top counterweight marking frame, and the lower boundary of the counterweight area is determined by the position of the bottom counterweight marking frame. The region box is adjusted based on the upper and lower boundaries to obtain the refined region box; From multiple detection frames, select the counterweight marker frames located within the region frame as valid counterweight marker frames, including: From multiple detection frames, select the counterweight block identification frames located within the fine-tuning area frame as valid counterweight block identification frames.

[0007] In this embodiment of the application, the equipment counterweight detection method further includes: Based on pixel position, the effective counterweight block identification boxes are grouped according to their adjacency relationship to obtain at least one group of adjacent counterweight block identification boxes; For each group of adjacent counterweight identification frames, a dynamic spacing threshold is determined based on the dimensions of the two adjacent counterweight identification frames. Determine whether the pixel distance between the center points of two adjacent counterweight block identifier frames is greater than the dynamic spacing threshold; If the pixel distance between the center points is greater than the dynamic spacing threshold, an anomaly is determined that there is a missed counterweight detection between adjacent counterweight block identification box groups.

[0008] In this embodiment of the application, the equipment counterweight detection method further includes: Obtain the preset counterweight value of the force limiter of the device to be tested; Compare the equipment counterweight with the preset counterweight value; If the equipment counterweight is not equal to the preset counterweight value, an anomaly of counterweight setting mismatch is identified.

[0009] In this embodiment of the application, the device to be tested includes multiple counterweight areas, and the number of image acquisition devices is multiple, and each of the multiple counterweight areas is set up in a one-to-one correspondence. Acquire images of the counterweight area of ​​the device to be tested, including: Obtain multiple sub-images of the counterweight regions, each corresponding to a different counterweight region; Determine the equipment counterweight of the device to be tested, including: For each counterweight region sub-image, the step of inputting the counterweight region sub-image into the target detection model to obtain multiple detection boxes of different categories is performed to obtain the sub-equipment counterweight of each counterweight region; The weights of all sub-devices are summed to obtain the weight of the device to be tested.

[0010] In this embodiment of the application, the equipment counterweight detection method further includes: Output data characterizing the equipment weight of the device under test; If an anomaly is detected during the equipment counterweight detection process, a prompt message indicating the type of anomaly will be output.

[0011] A second aspect of this application provides a device for detecting equipment counterweights, comprising: The image acquisition module is used to acquire images of the counterweight area of ​​the device under test. The images of the counterweight area are acquired by an image acquisition device that is fixedly installed on the device under test. The installation position and field of view of the image acquisition device are configured to cover the stacked area of ​​the counterweight blocks and acquire the identification information set on each counterweight block. The image recognition module is used to input the image of the counterweight area into the target detection model to obtain multiple detection boxes of different categories. The categories of detection boxes include region boxes, counterweight block identifier boxes, and weight identifier boxes. The detection frame filtering module is used to filter out the counterweight block identification frames located within the area frame from multiple detection frames, and use them as valid counterweight block identification frames. The counterweight determination module is used to determine the equipment counterweight of the device to be tested based on the quantitative relationship between the effective counterweight block identification frame and the weight identification frame, as well as the weight information corresponding to the weight identification frame.

[0012] A third aspect of this application provides a computing device, comprising: The memory is configured to store instructions; The processor is configured to retrieve instructions from memory and, when executing the instructions, to implement the device weight detection method as described in the above embodiments.

[0013] A fourth aspect of this application provides a machine-readable storage medium storing instructions that cause a machine to perform the equipment counterweight detection method as described in the above embodiments.

[0014] The above technical solution acquires images of the counterweight area of ​​the equipment under test. These images are captured by an image acquisition device fixedly installed on the equipment. The device's installation position and field of view are configured to cover the stacked counterweight area and capture the identification information on each counterweight. This fixed-viewpoint image acquisition provides a stable data foundation for identification. The counterweight area image is input into a target detection model, resulting in multiple detection boxes of different categories. A deep learning model is used to accurately locate the counterweight area elements, and spatial filtering effectively eliminates environmental interference. From these detection boxes, counterweight block identification boxes located within the area boxes are selected as valid counterweight block identification boxes. Based on the quantitative relationship between valid counterweight block identification boxes and weight identification boxes, as well as the weight information corresponding to the weight identification boxes, the equipment counterweight is determined. This multi-source information fusion and cross-validation mechanism ensures the accuracy and reliability of the counterweight calculation results under various complex working conditions, providing reliable data support for safe equipment operation and reducing the risk of safety accidents.

[0015] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 The illustration shows a schematic flowchart of a method for detecting equipment counterweight according to an embodiment of this application; Figure 2 A schematic diagram illustrating the identification of a counterweight block according to an embodiment of this application is shown. Figure 3 This schematic diagram illustrates the installation position of an image acquisition device according to an embodiment of this application; Figure 4 The illustration shows a structural schematic diagram of a device for detecting equipment counterweights according to an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0018] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0019] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0020] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0021] Figure 1 The illustration shows a schematic flowchart of a method for detecting equipment counterweights according to an embodiment of this application. Figure 1 As shown in the figure, this application provides a method for detecting equipment counterweight, which may include the following steps: Step 100: Obtain an image of the counterweight area of ​​the device to be tested. The image of the counterweight area is acquired by an image acquisition device that is fixedly installed on the device to be tested. The installation position and field of view of the image acquisition device are configured to cover the stacked area of ​​the counterweight blocks and acquire the identification information set on each counterweight block. In this embodiment, it should be noted that the equipment to be tested includes various heavy equipment requiring counterweight balancing, such as cranes, tower cranes, and gantry cranes. The image acquisition device typically refers to an industrial camera or webcam, but visual sensors with image acquisition capabilities can also be used. The installation position of the image acquisition device must ensure that the field of view completely covers the stacked area of ​​all counterweights and can clearly capture the marking information set on the side or surface of the counterweights. The specific installation position can be selected according to the structural characteristics of the equipment, such as the top of the machine shed or the side of the operator's cab, where it can clearly face the counterweight stacking area. The fixed installation position ensures that the spatial relationship between the image coordinate system and the physical counterweight area is constant, avoiding recognition errors caused by changes in viewing angle. The marking information may include numbers, letters, symbols, labels of specific colors or patterns, such as "5T" indicating a 5-ton counterweight; "β" indicating the manufacturer's exclusive pattern marking for the counterweight, etc. For example, on a port crane, the camera can be installed on a fixed bracket under the boom, with its field of view covering the counterweight stacking area at the rear of the crane body, clearly capturing the "10t" numerical marking on the side of each counterweight and the manufacturer's exclusive pattern marking "β" for the counterweight. In one embodiment, the image acquisition device may select a global shutter camera with a resolution of 1080P or higher, and configure a fill light or select a model with a wide dynamic range according to the ambient lighting conditions to ensure that clear images can still be obtained under complex lighting conditions such as backlight and shadow.

[0022] This embodiment provides high-quality input data for subsequent processing through a standardized image acquisition process, laying the physical foundation for the accuracy and stability of the entire recognition system.

[0023] Step 200: Input the counterweight region image into the target detection model to obtain multiple detection boxes of different categories, including region boxes, counterweight block identifier boxes, and weight identifier boxes. It's important to note that object detection models are artificial intelligence models built upon deep learning algorithms. Common examples include YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), and Faster R-CNN (Faster Region-based Convolutional Neural Networks). These models can automatically identify the location and category of multiple objects in an image simultaneously. The detection box is a rectangular region output by the object detection model used to select the target. Its coordinate information clearly defines the target's specific location in the image, and different categories of detection boxes correspond to different identified targets. The detection bounding boxes can be categorized into region boxes, counterweight identifier boxes, and weight identifier boxes. Region boxes are used to select the entire counterweight stack area, limiting the detection range to the target area to eliminate background interference. Counterweight identifier boxes are used to select the unique identifier on each counterweight, which can be a brand identifier, model identifier, or other markings used to distinguish the counterweight itself. Weight identifier boxes are used to select identifiers that directly or indirectly represent the weight of the counterweight, such as the numbers "5t" or "12.5t," the letters "Y" or "M" (corresponding to a preset weight), or weight QR codes. For example, when using the YOLOv5 model to process crane counterweight images, the model first identifies the region box containing all counterweights, and then identifies the counterweight identifier box corresponding to the "β" counterweight identifier on each counterweight within that region, as well as the weight identifier box corresponding to the "6t" numerical identifier. The Faster R-CNN model can also identify these three types of detection boxes, but there are differences in detection speed and accuracy. During the model training phase, a large number of weighted images containing different lighting, angles, and occlusion conditions need to be collected to form a training set. Various targets are then finely annotated using annotation tools so that the model can learn the visual features of various targets.

[0024] This embodiment uses the automatic recognition of the target detection model to quickly classify and extract key information in the image, replacing the tedious process of manual recognition and greatly improving detection efficiency. At the same time, the structured detection box output provides a clear and standardized data format for subsequent screening and calculation, ensuring the operability of subsequent steps.

[0025] Step 300: From multiple detection boxes, select the counterweight block identification boxes located within the area box as valid counterweight block identification boxes; It should be noted that all detection boxes obtained in the above steps are in the same image coordinate system. By calculating the positional relationship between each counterweight marker box and the region box, it can be determined whether it is located within a reasonable counterweight area. In specific implementation, the intersection-union ratio (IUR) of the two rectangles can be used for judgment. When the IUR is greater than a preset threshold, the counterweight marker box is considered valid. This effectively eliminates interference objects in the background that are similar to the counterweight marker, such as text on distant buildings or markings on tools. Another implementation method is to establish the boundary coordinate range of the region box and only retain the counterweight marker boxes whose center point coordinates fall within this range.

[0026] By introducing spatial prior knowledge for data cleaning, the risk of misidentification caused by environmental interference is reduced, and the robustness of the system in complex industrial scenarios is improved.

[0027] Step 400: Based on the quantitative relationship between the effective counterweight block identification frames and the weight identification frames, and the weight information corresponding to the weight identification frames, determine the equipment counterweight of the device to be tested.

[0028] It should be noted that determining the equipment counterweight based on the quantitative relationship between the effective counterweight marker frames and the weight marker frames, as well as the weight information corresponding to the weight marker frames, is the core output of the entire testing method. The final counterweight result is obtained through quantitative analysis, providing data support for safe equipment operation. The number of effective counterweight marker frames directly corresponds to the actual number of effective counterweights on the equipment under test. The weight information corresponding to the weight marker frames is the source of weight data for each counterweight. The quantitative relationship between the two must meet a reasonable matching principle, commonly a one-to-one correspondence, meaning one effective counterweight marker frame corresponds to one weight marker frame, ensuring that each counterweight has corresponding weight data. When a weight marker frame cannot be identified due to obstruction, blurring, or other reasons, it can be supplemented with calculations based on the preset weight information corresponding to the counterweight marker frame. If the number of weight marker frames is inconsistent with the number of effective counterweight marker frames, it can be corrected through reasonable verification rules before calculation, such as adjacent counterweight spacing analysis and marker integrity judgment verification rules. Weight information can be obtained by directly reading the numerical identifier in the weight identifier box, such as "10t" which directly corresponds to a weight of 10 tons, or by reading indirect identifiers such as letters and QR codes and matching them with a preset weight database, such as the letter "Y" which corresponds to 5 tons. The final equipment counterweight is the sum of the weights of all effective counterweights.

[0029] In one embodiment, when the number of valid counterweight marker frames and weight marker frames is inconsistent, a backup strategy can be activated. For example, priority can be given to weight marker frames with high recognition confidence, or weight can be inferred based on the spatial correspondence between counterweight marker frames and letter weight marker frames. The weight information corresponding to the weight marker frames can be implemented by establishing a mapping table; for example, detecting the letter "5T" maps to 5000 kg, and detecting the letter "M" maps to 5000 kg. In more complex embodiments, logical verification can also be performed by combining the spatial arrangement order of multiple weight marker frames; for example, adjacent counterweights usually have the same weight or follow a specific pattern.

[0030] This embodiment achieves fault-tolerant calculation in the event of missing or abnormal partial identification information through a multi-level information fusion and verification mechanism, ensuring the integrity and reliability of the counterweight statistics results.

[0031] In this embodiment, an image of the counterweight area of ​​the equipment under test is acquired. This image is captured by an image acquisition device fixedly installed on the equipment. The installation position and field of view of the image acquisition device are configured to cover the stacked area of ​​the counterweight blocks and acquire the identification information set on each counterweight block. Image acquisition from a fixed perspective provides a stable data foundation for identification. The counterweight area image is input into a target detection model to obtain multiple detection boxes of different categories. A deep learning model is used to achieve accurate localization of the counterweight area elements, and spatial filtering effectively eliminates environmental interference. From the multiple detection boxes, the counterweight block identification boxes located within the area boxes are selected as valid counterweight block identification boxes. Based on the quantitative relationship between the valid counterweight block identification boxes and the weight identification boxes, as well as the weight information corresponding to the weight identification boxes, the equipment counterweight of the equipment under test is determined. Based on the multi-source information fusion and cross-validation mechanism, the accuracy and reliability of the counterweight calculation results are ensured under various complex working conditions, providing reliable data support for safe equipment operation and reducing the risk of safety accidents.

[0032] In one embodiment, the weight marking frame includes a first weight marking frame and a second weight marking frame with different marking information. Based on the quantity relationship between the effective counterweight marking frames and the weight marking frames, and the weight information corresponding to the weight marking frames, the equipment counterweight of the device to be tested is determined, including: Determine whether the number of valid counterweight marker frames is consistent with the number of first weight marker frames; When the number of effective counterweight marking frames is the same as the number of first weight marking frames, the equipment counterweight of the device to be tested is determined based on the sum of the first weight information of each first weight marking frame. If the number of valid counterweight marking frames is inconsistent with the number of first weight marking frames, determine whether the number of valid counterweight marking frames is consistent with the number of second weight marking frames. When the number of effective counterweight marking frames is the same as the number of second weight marking frames, the equipment counterweight of the device to be tested is determined based on the sum of the second weight information of each second weight marking frame. If the number of valid counterweight marking frames is inconsistent with the number of second weight marking frames, an anomaly is identified.

[0033] In this embodiment, it should be noted that the weight marking frame includes a first weight marking frame and a second weight marking frame, and the marking information of the two is different. The first weight marking frame usually refers to the weight marking with the highest identification priority, such as a clearly identifiable numerical weight label like "5t" or "10t"; the second weight marking frame refers to auxiliary weight markings, such as letter labels like "Y" or "M" used as supplementary markings when the numerical label is blurred or obscured. (Reference) Figure 2 , Figure 2 In one embodiment, the individual counterweight is marked with a digital weight label “5t”, a letter label “Y”, and a manufacturer-specific graphic designation “β” for the counterweight.

[0034] Specifically, the system first compares the number of valid counterweight marker frames with the number of first weight marker frames. If the numbers match, the first weight information is summed directly, yielding the most accurate counterweight result. If the numbers do not match, a backup judgment process is automatically activated. This process verifies the relationship between the number of valid counterweight marker frames and the number of second weight marker frames. If they match, the system switches to summing the second weight information. This hierarchical judgment mechanism ensures that calculations can still be completed using auxiliary features even when the main identification features are missing. When both levels of verification fail, an anomaly is promptly reported to prevent erroneous data output.

[0035] In this embodiment, by establishing multiple verification channels, the risk of detection failure caused by blurry or obscured single identifiers is reduced, the fault tolerance and accuracy of counterweight calculation are improved, and the reliability of detection results is guaranteed.

[0036] In one embodiment, the counterweight marking frame includes a top counterweight marking frame and a bottom counterweight marking frame, and the equipment counterweight detection method further includes: The upper boundary of the counterweight area is determined by the position of the top counterweight marking frame, and the lower boundary of the counterweight area is determined by the position of the bottom counterweight marking frame. The region box is adjusted based on the upper and lower boundaries to obtain the refined region box; From multiple detection frames, select the counterweight marker frames located within the region frame as valid counterweight marker frames, including: From multiple detection frames, select the counterweight block identification frames located within the fine-tuning area frame as valid counterweight block identification frames.

[0037] In this embodiment, it should be noted that the counterweight identification frame includes a top counterweight identification frame and a bottom counterweight identification frame. The top counterweight identification frame specifically refers to the identification frame corresponding to the topmost counterweight in the counterweight stack, while the bottom counterweight identification frame refers to the identification frame of the bottommost counterweight. By extracting the upper boundary pixel coordinates of the top counterweight identification frame as the upper boundary of the counterweight region, and simultaneously extracting the lower boundary pixel coordinates of the bottom counterweight identification frame as the lower boundary of the counterweight region, the resulting refined region frame can more accurately reflect the vertical distribution range of the actual counterweight stack. Compared to the initial region frame, the refined region frame, by anchoring two counterweights with clear positional characteristics, effectively eliminates background interference and misidentified targets outside the stacked area. When filtering valid counterweight identification frames, replacing the filtering range from the initial region frame to the refined region frame ensures that only the identification results truly located within the counterweight stacked area are retained.

[0038] In this embodiment, the spatial refinement strategy improves the accuracy of the counterweight block identification frame selection, provides a cleaner dataset for subsequent calculations, and enhances the overall detection accuracy.

[0039] In one embodiment, the equipment counterweight detection method further includes: Based on pixel position, the effective counterweight block identification boxes are grouped according to their adjacency relationship to obtain at least one group of adjacent counterweight block identification boxes; For each group of adjacent counterweight identification frames, a dynamic spacing threshold is determined based on the dimensions of the two adjacent counterweight identification frames. Determine whether the pixel distance between the center points of two adjacent counterweight block identifier frames is greater than the dynamic spacing threshold; If the pixel distance between the center points is greater than the dynamic spacing threshold, an anomaly is determined that there is a missed counterweight detection between adjacent counterweight block identification box groups.

[0040] In this embodiment, it should be noted that, based on the vertical pixel coordinates of the counterweight identification frame in the image, adjacent counterweight identification frames are grouped in order from top to bottom or from bottom to top to form several adjacent counterweight identification frame groups.

[0041] Understandably, setting a dynamic spacing threshold is crucial for accurately identifying missed counterweights. This embodiment effectively improves detection accuracy through adaptive calculation. Specifically, the dynamic spacing threshold is not a fixed value, but a calculated result that is dynamically adjusted based on the real-time detection dimensions of adjacent counterweights. Specifically, the dynamic spacing threshold is equal to half the sum of the pixel heights of the two adjacent counterweight identification frames multiplied by a preset coefficient, which typically ranges from 1.2 to 1.4. Here, pixel height refers to the pixel span of the counterweight identification frame in the vertical direction of the image, reflecting the projected size of the actual physical thickness of the counterweight in the image. The dynamic calculation method fully considers the spatial relationship characteristics of combinations of counterweights with different thicknesses. For example, when two thick counterweights are adjacent, their reasonable spacing threshold will naturally be greater than that of two thin counterweights adjacent. In practical applications, the system calculates the dynamic spacing threshold for each pair of adjacent counterweights in real time and compares the calculated dynamic spacing threshold with the measured pixel distance between the center points.

[0042] In one embodiment, the target counterweights are first classified by pixel height, and the median of the pixel heights of all counterweights is calculated. If the pixel height of a counterweight (i.e., the difference between the upper and lower pixels, h) is greater than the median, it is determined to be a thick block; if it is less than or equal to the median, it is determined to be a thin block. Next, the desired spacing is determined based on the thickness combinations of adjacent counterweights. Thickness combinations include, for example, thin-thin, thin-thick, and thick-thick. The normal physical spacing corresponding to different combinations is different. In actual operation, the median of the actual pixel spacing under the same combination can be used as the desired spacing. Alternatively, the desired spacing can be simplified to 0.5×(h1+h2) by utilizing the characteristic that the relative distance between the camera and the counterweight is fixed, where h1 represents the difference between the upper and lower pixels of the upper counterweight and h2 represents the difference between the upper and lower pixels of the lower counterweight. Finally, a redundancy coefficient is introduced, with a value ranging from 1.2 to 1.4. The dynamic spacing threshold is set as the product of the desired spacing and the redundancy coefficient, i.e., 0.5 × (h1 + h2) × (1.2 ~ 1.4). The redundancy coefficient is set to accommodate slight installation deviations or image distortions. This embodiment reduces the risk of misjudgment due to differences in counterweight thickness by introducing a size-adaptive threshold calculation mechanism, making missed detection more consistent with the layout characteristics of actual physical space.

[0043] Specifically, by calculating the pixel distance between the center points of two adjacent counterweight block identification frames and comparing it with a dynamic spacing threshold, when the actual pixel distance is significantly greater than the expected threshold, it indicates that there may be an unidentified counterweight block between the two adjacent counterweight blocks, thus determining that there is a missed detection anomaly. Based on the prior knowledge that the counterweight blocks should be tightly stacked, it can effectively identify missed detections caused by occlusion, lighting, or limitations of the recognition algorithm.

[0044] In this embodiment, spatial relationship analysis provides an effective means of verifying the completeness of the identification results, enhances the system's self-diagnosis capabilities, avoids underestimation of counterweight due to missed detections, and further ensures the safety of equipment operation.

[0045] In one embodiment, the equipment counterweight detection method further includes: Obtain the preset counterweight value of the force limiter of the device to be tested; Compare the equipment counterweight with the preset counterweight value; If the equipment counterweight is not equal to the preset counterweight value, an anomaly of counterweight setting mismatch is identified.

[0046] In this embodiment, it should be noted that the force limiter is a safety protection device installed on the equipment under test, used to preset safe counterweight values ​​for different operating scenarios. The preset counterweight value in the force limiter is read through the equipment communication interface. This preset counterweight value is usually manually set by the operator according to the construction plan. The equipment counterweight value calculated by visual recognition is compared with the preset counterweight value of the force limiter in real time. When the two values ​​are found to be unequal, an anomaly of counterweight setting mismatch is determined. This mismatch may stem from various reasons such as operator input errors, failure to update settings in a timely manner after counterweight replacement, or visual recognition errors. In this embodiment, the comparison mechanism does not directly modify the force limiter settings, but rather returns the decision-making power to the operator through an anomaly prompt, ensuring that any counterweight change is manually confirmed.

[0047] In this embodiment, by establishing data verification between visual recognition and the equipment control system, the error in counterweight setting caused by human negligence or system error is effectively prevented, thereby improving the overall operational safety.

[0048] In one embodiment, the device to be tested includes multiple counterweight areas, and the number of image acquisition devices is multiple, with each of the multiple counterweight areas corresponding to one another. Acquire images of the counterweight area of ​​the device to be tested, including: Obtain multiple sub-images of the counterweight regions, each corresponding to a different counterweight region; Determine the equipment counterweight of the device to be tested, including: For each counterweight region sub-image, the step of inputting the counterweight region sub-image into the target detection model to obtain multiple detection boxes of different categories is performed to obtain the sub-equipment counterweight of each counterweight region; The weights of all sub-devices are summed to obtain the weight of the device to be tested.

[0049] In this embodiment, it should be noted that when the device under test has multiple physically isolated counterweight areas, each image acquisition device independently acquires a sub-image of the corresponding counterweight area. Each sub-image is then fed into the target detection model for processing, independently generating a set of detection boxes for that area and calculating the sub-device counterweight. Finally, all sub-device counterweights are merged into the total device counterweight through a summation operation. This distributed processing architecture ensures the independence of each area's recognition while achieving overall counterweight management through unified calculation. It is understood that there can be one or two counterweight areas; the number of counterweight areas is not limited. The number of image acquisition devices corresponds one-to-one with the counterweight areas, improving the accuracy of image acquisition. For example, if a device has counterweight stacking areas on both the left and right sides, then two image acquisition devices are configured to acquire images corresponding to each counterweight area. (Reference) Figure 3 Taking a camera as an image acquisition device as an example, in Figure 3 In the middle, there are counterweight stacking areas on both the left and right sides of the device, and camera 1 and camera 2 respectively capture the counterweight area sub-images of the left and right counterweight areas.

[0050] In this embodiment, a modular processing scheme enables the system to flexibly adapt to different equipment configurations and support the accurate detection requirements of complex counterweight layouts.

[0051] In one embodiment, the equipment counterweight detection method further includes: Output data characterizing the equipment weight of the device under test; If an anomaly is detected during the equipment counterweight detection process, a prompt message indicating the type of anomaly will be output.

[0052] In this embodiment, it should be noted that the output data representing the equipment's counterweight can be implemented in various forms, including visually displaying the counterweight value on a monitor, sending structured data to the control system via a communication bus, or generating a counterweight statistical report. When any abnormality is detected during the detection process, the system will simultaneously output prompt information including the type of abnormality. For example, a pop-up warning may indicate a missed abnormality, an audible alarm may indicate a counterweight setting mismatch, or a flashing status light may indicate a system abnormality. The output information can be distinguished into data output under normal conditions and warning output under abnormal conditions, ensuring that operators can quickly identify the system status and take appropriate measures.

[0053] In this embodiment, a robust result output mechanism ensures the effective transmission of test results, enabling users to quickly grasp the equipment's counterweight status, promptly address any abnormalities, and improve the practicality and safety of the testing method.

[0054] Figure 4 The diagram illustrates the structure of a device for detecting equipment counterweights according to an embodiment of this application. Figure 4As shown in the figure, this application provides a device for detecting equipment counterweight 1000, which may include: The image acquisition module 1001 is used to acquire the image of the counterweight area of ​​the device to be tested. The image of the counterweight area is acquired by an image acquisition device fixedly installed on the device to be tested. The installation position and field of view of the image acquisition device are configured to cover the stacked area of ​​the counterweight blocks and acquire the identification information set on each counterweight block. The image recognition module 1002 is used to input the image of the counterweight area into the target detection model to obtain multiple detection boxes of different categories, including region boxes, counterweight block identification boxes and weight identification boxes. The detection frame filtering module 1003 is used to filter out the counterweight block identification frames located within the area frame from multiple detection frames, and use them as valid counterweight block identification frames. The counterweight determination module 1004 is used to determine the equipment counterweight of the device to be tested based on the quantitative relationship between the effective counterweight block identification frame and the weight identification frame, as well as the weight information corresponding to the weight identification frame.

[0055] The equipment counterweight detection device provided in this application embodiment can respectively achieve… Figure 1 The various processes of the equipment counterweight detection method in the method embodiment can achieve the same technical effect, and will not be described again here to avoid repetition.

[0056] This application embodiment also provides a computing device, including: The memory is configured to store instructions; The processor is configured to retrieve instructions from memory and, when executing the instructions, to implement the device weight detection method as described in the above embodiments.

[0057] This application also provides a machine-readable storage medium storing instructions that cause a machine to perform the equipment counterweight detection method as described in the above embodiments.

[0058] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0059] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0060] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0061] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0062] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0063] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0064] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0065] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0066] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A device weight detection method, characterized by, The method comprises the following steps: acquiring an image of a weight region of a device to be detected, wherein the image of the weight region is acquired by an image acquisition device fixedly installed on the device to be detected, and the installation position and field of view of the image acquisition device are configured to cover the weight block stacking region and acquire identification information arranged on each weight block; inputting the image of the weight region into a target detection model to obtain a plurality of detection boxes of different categories, wherein the categories of the detection boxes include a region box, a weight block identification box, and a weight identification box; from the plurality of detection boxes, screening out the weight block identification boxes located in the region box as valid weight block identification boxes; determining the device weight of the device to be detected based on the quantity relationship between the valid weight block identification boxes and the weight identification boxes and the weight information corresponding to the weight identification boxes.

2. The apparatus weight detection method according to claim 1, wherein, The weight identification box includes a first weight identification box and a second weight identification box with different identification information, and the determination of the device weight of the device to be detected based on the quantity relationship between the valid weight block identification boxes and the weight identification boxes and the weight information corresponding to the weight identification boxes comprises: determining whether the number of the valid weight block identification boxes is consistent with the number of the first weight identification boxes; in the case where the number of the valid weight block identification boxes is consistent with the number of the first weight identification boxes, determining the device weight of the device to be detected based on the sum of the first weight information of each first weight identification box; in the case where the number of the valid weight block identification boxes is not consistent with the number of the first weight identification boxes, determining whether the number of the valid weight block identification boxes is consistent with the number of the second weight identification boxes; in the case where the number of the valid weight block identification boxes is consistent with the number of the second weight identification boxes, determining the device weight of the device to be detected based on the sum of the second weight information of each second weight identification box; in the case where the number of the valid weight block identification boxes is not consistent with the number of the second weight identification boxes, determining that there is a detection abnormality.

3. The apparatus weight detection method of claim 1, wherein, The weight block identification box includes a top weight identification box and a bottom weight identification box, and the device weight detection method further comprises: determining the upper boundary of the weight region according to the position of the top weight identification box and determining the lower boundary of the weight region according to the position of the bottom weight identification box; adjusting the region box based on the upper boundary and the lower boundary to obtain a refined region box; from the plurality of detection boxes, screening out the weight block identification boxes located in the refined region box as valid weight block identification boxes. Further comprising:

4. The apparatus weight detection method of claim 1, wherein, grouping the valid weight block identification boxes according to the adjacent relationship based on the pixel position to obtain at least one adjacent weight block identification box group; for each adjacent weight block identification box group, determining a dynamic spacing threshold value according to the size of two adjacent weight block identification boxes; determining whether the center pixel distance of the two adjacent weight block identification boxes is greater than the dynamic spacing threshold value; ​ In a case where the center pixel distance is greater than the dynamic interval threshold, it is determined that there is an abnormality of missing counterweights between the adjacent counterweight identification frame groups.

5. The apparatus weight detection method of claim 1, wherein, Further comprising: Obtaining a preset counterweight value of a force limiter of the device to be detected; Comparing the device counterweight with the preset counterweight value; In a case where the device counterweight is not equal to the preset counterweight value, it is determined that there is an abnormality of counterweight setting mismatch.

6. The apparatus weight detection method of claim 1, wherein, The device to be detected comprises a plurality of counterweight regions, and the number of the image acquisition devices is equal to the number of the plurality of counterweight regions and is arranged in one-to-one correspondence with the plurality of counterweight regions. The obtaining of the counterweight region image of the device to be detected comprises: Obtaining a plurality of counterweight region sub-images corresponding to the plurality of counterweight regions, respectively; The determination of the device counterweight of the device to be detected comprises: For each counterweight region sub-image, the step of inputting the counterweight region sub-image into a target detection model to obtain a plurality of detection frames of different categories is performed to obtain a sub-device counterweight of each counterweight region; Summing all the sub-device counterweights to obtain the device counterweight of the device to be detected.

7. The apparatus weight detection method of claim 1, wherein, Further comprising: Outputting data representing the device counterweight of the device to be detected; In a case where it is determined that there is an abnormality in the device counterweight detection process, outputting prompt information indicating the category of the abnormality.

8. An apparatus weight detection device, characterized by, Comprising: An image acquisition module configured to obtain a counterweight region image of a device to be detected, wherein the counterweight region image is acquired by an image acquisition device fixedly installed on the device to be detected, and the installation position and field of view of the image acquisition device are configured to cover a counterweight stack region and acquire identification information arranged on each counterweight; An image recognition module configured to input the counterweight region image into a target detection model to obtain a plurality of detection frames of different categories, wherein the categories of the detection frames include region frames, counterweight identification frames, and weight identification frames; A detection frame screening module configured to screen, from the plurality of detection frames, counterweight identification frames located in region frames as valid counterweight identification frames; A counterweight determination module configured to determine a device counterweight of the device to be detected based on a quantity relationship between the valid counterweight identification frames and the weight identification frames and weight information corresponding to the weight identification frames.

9. A computing device, comprising: Comprising: A memory configured to store instructions; A processor configured to call the instructions from the memory and implement the device counterweight detection method according to any one of claims 1 to 7 when the instructions are executed.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing a machine to execute the device counterweight detection method according to any one of claims 1 to 7.

Images