A method and system for visual identification of loading quantity under heavy dust working condition

Abstract

The application discloses a kind of methods and systems for visual identification loading quantity under heavy dust working condition, comprising: obtaining image data and video data;Calculate texture degradation index, dust movement density, contrast attenuation factor, and identify heavy dust working condition;In heavy dust working condition, the time that the movable arm is improved from lower limit to upper limit is obtained, and the angle change amount of movable arm lower limit to upper limit is obtained, the compensation coefficient is obtained according to the angle change amount, the movable arm rising angle is obtained by the update cycle of preset movable arm rising pilot pressure;Whether it is effective loading bucket number is judged by pump pressure, if yes, when unloading is finished, loading bucket number is counted, otherwise it is not counted;According to single vehicle loading bucket number, calculate reference bucket number, and based on single vehicle loading bucket number and reference bucket number, obtain the final loading quantity.The application solves the technical problem that video picture is seriously blocked under heavy dust working condition, which leads to the failure of visual identification function.

Description

Technical Field

[0001] This invention relates to the field of excavator equipment monitoring technology, and in particular to a method and system for visually identifying the quantity of goods loaded on a truck under heavy dust conditions. Background Technology

[0002] The statistics of the number of buckets and the quantity loaded on large excavators are core data supports for production scheduling, cost accounting, and efficiency evaluation. Traditional manual measurement methods not only consume a lot of manpower but are also prone to statistical distortion due to human error, making it difficult to meet the requirements of modern operations for data accuracy and real-time performance. With the development of automation technology, automatic calculation solutions based on pilot pressure monitoring and image recognition are gradually replacing manual labor. However, under harsh working conditions such as heavy dust, existing technologies still face problems of functional failure or insufficient accuracy, and there is an urgent need for targeted optimization solutions.

[0003] In existing technologies, a collaborative logic of "pilot pressure + image recognition" is adopted. The pilot pressure of bucket digging, stick digging, boom raising and unloading is monitored by pressure sensors. Once the target is reached, the AI ​​camera is triggered to identify the bucket material, ore clips and ore clip material status, and count the effective number and quantity of buckets.

[0004] However, the study found that existing technologies rely on AI cameras to identify the material in the bucket and the status of the mining truck. However, under heavy dust conditions, the video image is severely obscured, blurred, and the contrast is reduced. Edge details in the image are lost, and dynamic particles (dust) interfere with the image, making it impossible for the camera to accurately identify the target. This leads to the failure of the visual recognition function, which directly causes the interruption or misjudgment of the number of buckets loaded and the number of trucks loaded. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a method and system for visually recognizing the number of loads on a vehicle under heavy dust conditions. Through steps such as visually recognizing heavy dust conditions, calculating the number of loads based on vehicle messages, and calibrating the number of loads based on the number of loads, this invention solves the technical problem of severe obstruction of video footage under heavy dust conditions, which leads to the failure of the visual recognition function.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for visually identifying the quantity of goods loaded onto a vehicle under heavy dust conditions, comprising: Acquire image and video data of the loading operation scene; Based on the image data, calculate the texture degradation index; based on the video data, calculate the dust movement density; calculate the contrast attenuation factor based on the contrast attenuation model; and identify heavy dust conditions based on the texture degradation index, dust movement density, and contrast attenuation factor. The time it takes for the boom to rise from its lower limit to its upper limit is obtained, and the angle change from the lower limit to the upper limit is obtained. The compensation coefficient is obtained based on the angle change. The boom rising angle is obtained by setting the update cycle of the boom rising pilot pressure. After the material transfer is completed, the pump pressure is used to determine whether it is an effective loading bucket. If it is, it is counted as a loading bucket after the unloading is completed; otherwise, it is not counted. The baseline number of buckets is calculated based on the number of buckets loaded per vehicle, and the final loading quantity is obtained based on the number of buckets loaded per vehicle and the baseline number of buckets.

[0007] A further technical solution is as follows: First, the image data is scaled down proportionally to obtain image data at different scales. Then, based on the image data at each scale, the Sobel operator is used to calculate the horizontal and vertical gradients to obtain the gradient value of each pixel. The average gradient value of the image data at each scale is then calculated. Finally, the minimum and maximum average gradient values ​​are selected from these values ​​to calculate the texture degradation index.

[0008] A further technical solution is that the method for calculating dust motion density is as follows: First, the difference between two consecutive frames of video data is calculated to obtain a difference image. Then, the difference image is binarized. The binarization process includes setting a difference threshold. If the difference between the difference images is greater than the difference threshold, it is marked as a moving pixel and a moving region is obtained. Finally, the moving region is filtered to obtain the number of moving pixels. Combined with the total number of pixels in the image, the dust motion density is calculated.

[0009] A further technical solution is that the method for calculating the contrast attenuation factor is as follows: first, the image data is processed into grayscale to obtain a grayscale image, and the actual contrast of the grayscale image is calculated. Then, the calibrated contrast is obtained, and finally, the contrast attenuation factor is calculated based on the actual contrast and the calibrated contrast.

[0010] A further technical solution, the premise for obtaining the time for the boom to rise from the lower limit to the upper limit, is to maintain the slewing pilot pressure at its maximum value and the boom rising pilot pressure at its maximum value; the compensation coefficient obtained based on the angle change is calculated using the following formula: ;in, The compensation coefficients are for different working speeds. For working gear, This represents the change in angle between the lower and upper limits of the boom. This represents the maximum pilot pressure for boom lifting. The time it takes for the boom to rise from its lower limit to its upper limit.

[0011] A further technical solution, when the slewing pilot pressure remains at its maximum value, the boom lifting angle is expressed as: ;in, The boom lifting angle, The update cycle for the boom lifting pilot pressure. The compensation coefficients are for different working speeds. For the first The value of the boom lifting pilot pressure per second. For the 2nd The value of the boom lifting pilot pressure per second. For the 3rd The value of the boom lifting pilot pressure per second. For the first The value of the boom lifting pilot pressure per second.

[0012] A further technical solution is that when the boom lifting angle is greater than the preset lifting angle threshold, it means that the material transfer action is over; if the pump pressure reaches the preset pump pressure threshold, it means that there is material in the bucket, which is a valid loading bucket number; if the pump pressure does not reach the preset pump pressure threshold, it means that there is no material in the bucket, which is an invalid loading bucket number.

[0013] Secondly, the present invention provides a visual recognition system for vehicle loading quantity under heavy dust conditions, comprising the following modules: The data acquisition module is configured to acquire image and video data of the loading operation scene; The working condition recognition module is configured to: calculate the texture degradation index based on the image data; calculate the dust movement density based on the video data; calculate the contrast attenuation factor based on the contrast attenuation model; and identify heavy dust working conditions based on the texture degradation index, dust movement density, and contrast attenuation factor. The loading bucket count calculation module is configured to: obtain the time it takes for the boom to rise from the lower limit to the upper limit, and obtain the angle change from the lower limit to the upper limit of the boom; obtain a compensation coefficient based on the angle change; obtain the boom rising angle by setting the update cycle of the boom rising pilot pressure; after the material transfer is completed, determine whether it is a valid loading bucket count by the pump pressure; if so, it is counted in the loading bucket count after the unloading is completed; otherwise, it is not counted. The calibration module is configured to calculate the baseline number of buckets based on the number of buckets loaded per vehicle, and to obtain the final loading quantity based on the number of buckets loaded per vehicle and the baseline number of buckets.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention effectively avoids the interference of heavy dust on visual recognition by visually recognizing heavy dust conditions and calculating the number of loading buckets based on vehicle-wide reports. By calculating the comprehensive dust index using texture degradation index, dust movement density, and contrast attenuation factor, and combining this with calculation logic based on vehicle-wide reports, the boom lifting angle is calculated using the boom lifting pilot pressure. Combined with pump pressure judgment, the number of loading buckets is statistically analyzed. This ensures the accuracy of bucket count calculation even in heavy dust scenarios, avoiding statistical interruptions or misjudgments caused by image occlusion in existing technologies. Attached Figure Description

[0015] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0016] Figure 1 This is a schematic diagram of the visual recognition method for the quantity of goods loaded onto a vehicle according to the present invention. Detailed Implementation The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0017] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0018] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0019] Example 1 This embodiment provides a method for visually identifying the quantity of goods loaded onto a vehicle under heavy dust conditions, such as... Figure 1 The method flowchart is shown below, and the specific steps are as follows: S1: Acquire image and video data of the loading operation scene, and identify heavy dust conditions based on the acquired image and video data. The specific method is as follows: Since dust can blur images and reduce the amount of edges and details, this embodiment first performs multi-scale texture degradation analysis based on the acquired image data to calculate the texture degradation index. Specifically: First, the image data is scaled down proportionally to the original image (1x), reduced by 2x, and reduced by 4x, resulting in three different scales. Then, based on the image data at each scale, the Sobel operator is used to calculate the horizontal and vertical gradients. The horizontal gradient... =Image × [[-1,0,1],[-2,0,2],[-1,0,1]]; Vertical gradient =Image × [[-1,-2,-1],[0,0,0],[1,2,1]]; The gradient value of each pixel is calculated based on the horizontal and vertical gradients, and the gradient value = The average gradient value of the image data at each scale is calculated. The average gradient value is the average of the gradient values ​​of all pixels. Finally, the minimum and maximum average gradient values ​​are selected to calculate the texture degradation index. The calculation formula is as follows: Texture degradation index = 1 - (minimum gradient average / maximum gradient average).

[0020] The physical meaning of calculating the texture degradation index: When there is no dust: the average gradient values ​​at different scales differ greatly → the texture degradation index is close to 0; When dust is severe: images at all scales are blurred → the average gradient is close to 1 → the texture degradation index is close to 1.

[0021] Since dust particles float in the air, they can cause randomly moving white dots to appear in the video. Therefore, in this embodiment, the difference between two consecutive frames of video data is calculated to obtain a difference image, where the difference image = |current frame - previous frame|.

[0022] The difference images are then binarized, which includes setting a difference threshold. If the difference between the images is greater than the difference threshold If the difference is less than or equal to the difference threshold, it is marked as a moving pixel, and the moving region is obtained. If the area is large, it is marked as a non-moving pixel. Finally, the moving areas are filtered out, removing large moving areas (which may be vehicles or other objects, not dust), and retaining moving areas with an area between 2 and 50 pixels. This gives the number of moving pixels. Combined with the total number of pixels in the image (referring to the total number of pixels in the original video frame (the current frame)), the dust motion density is calculated using the following formula: Dust movement density = (number of moving pixels) / (total number of pixels in the image).

[0023] The physical meaning of calculating dust kinematic density: When there is no dust: the kinematic density is approximately 0; when there is severe dust: the kinematic density of dust is close to 1.

[0024] Furthermore, since dust can cause images to appear grayish-white, reducing the difference between the brightest and darkest parts, this embodiment calculates the contrast attenuation factor based on a contrast attenuation model. First, the image data is processed into grayscale to obtain a grayscale image, and then the actual contrast of the grayscale image is calculated. Actual contrast = ;in, In a grayscale image, 95% of the pixels have a grayscale value less than its own. 5% of the pixels in a grayscale image have a grayscale value that is less than or equal to its grayscale value.

[0025] Next, obtain the calibration contrast ratio, which is the contrast value measured in a clean environment. Finally, based on the actual contrast ratio and the calibration contrast ratio, calculate the contrast attenuation factor using the following formula: Contrast attenuation factor = 1 - (actual contrast / rated contrast).

[0026] The physical meaning of calculating the contrast attenuation factor: When there is no dust: the actual contrast ratio is approximately equal to the calibrated contrast ratio, and the attenuation factor is approximately 0. When there is severe dust: the actual contrast ratio decreases, and the attenuation factor approaches 1.

[0027] Finally, based on the texture degradation index, dust movement density, and contrast attenuation factor, an overall score for the severity of dust, namely the comprehensive dust index, is given, as follows: Dust Comprehensive Index =0.5×Texture Degradation Index + 0.3×Dust Movement Density + 0.2×Contrast Attenuation Factor.

[0028] According to the comprehensive dust index The heavy dust condition was identified, specifically: Dust Comprehensive Index <0.3 → Dust-free; 0.3≤Dust Comprehensive Index <0.6 → Light dust; Dust Comprehensive Index ≥0.6→Severe dust.

[0029] S2: When a heavy dust condition is identified using the method in step S1, the time it takes for the boom to rise from the lower limit to the upper limit is obtained, and the angle change from the lower limit to the upper limit of the boom is obtained. The compensation coefficient is obtained based on the angle change, and the boom rising angle is obtained by preset the update cycle of the boom rising pilot pressure.

[0030] In this embodiment, the premise for obtaining the time for the boom to rise from the lower limit to the upper limit is to maintain the slewing pilot pressure and the boom rising pilot pressure at their maximum values. When both the slewing pilot pressure and the boom rising pilot pressure are at their maximum values, the time for the boom to rise from the lower limit to the upper limit is obtained, and the unit is seconds. The angular change from the lower limit to the upper limit of the boom is also obtained. The unit is degrees, and it is rounded to one decimal place.

[0031] The compensation coefficient is obtained based on the change in angle, and the formula is as follows: ;in, The compensation coefficients are for different working speeds. These are the 1-11 working gears. This represents the change in angle between the lower and upper limits of the boom. This represents the maximum pilot pressure for boom lifting. The time it takes for the boom to rise from its lower limit to its upper limit.

[0032] To further ensure accuracy without increasing the load on the controller, a preset update cycle for the boom lifting pilot pressure is used. The boom lifting angle is obtained. In this embodiment, Set to 0.1s, when the slewing pilot pressure remains at its maximum value, the boom rise angle is expressed as: ;in, The boom lifting angle, The update cycle for the boom lifting pilot pressure. The compensation coefficients are for different working speeds. For the first The value of the boom lifting pilot pressure per second. For the 2nd The value of the boom lifting pilot pressure per second. For the 3rd The value of the boom lifting pilot pressure per second. For the first The value of the boom lifting pilot pressure per second.

[0033] S3: After the material transfer is completed, the pump pressure is used to determine whether it is a valid loading bucket number. If it is, it is counted as a loading bucket number after the unloading is completed; otherwise, it is not counted.

[0034] As can be seen, the loading action is divided into digging, moving, and unloading. First, the digging action is judged by the bucket digging pilot pressure and the boom digging pilot pressure (the specified values ​​and holding time are to filter out accidental operation by the operator). When the boom lifting pilot pressure reaches the set value and the set time (the boom lifting pilot pressure value is greater than 5 and the holding time is greater than 0.2s), it means that the digging action is over.

[0035] When the slewing pilot pressure reaches the set value (the slewing pilot pressure value is greater than 20 and the holding time is greater than 0.5s), it means that the material transfer action has started. At this time, the boom lifting angle is calculated according to the method in step S2. The accuracy requirement of the boom lifting angle is not high here, and adding an angle sensor will increase the cost. The boom lifting angle is used to filter out non-loading operations such as sorting materials.

[0036] When the boom rises at an angle greater than the preset rise angle threshold, the material transfer action is complete. If the pump pressure reaches the preset pump pressure threshold, it means there is material in the bucket, which is a valid loading count. If the pump pressure does not reach the preset pump pressure threshold, it means there is no material in the bucket, which is an invalid loading count.

[0037] For a valid loading bucket count, the unloading action ends when the bucket unloading pilot pressure reaches the set value and the set time (the bucket unloading pilot pressure value is greater than 10 and the holding time is greater than 0.5s), and the loading bucket count is incremented by 1. This algorithm can calculate the loading bucket count in blind spots without increasing any hardware costs. The algorithm is simple and has high calculation accuracy.

[0038] S4: Calculate the baseline number of buckets based on the number of buckets loaded per vehicle, and obtain the final loading quantity based on the number of buckets loaded per vehicle and the baseline number of buckets.

[0039] The calculation results obtained in step S3 are uploaded to the cloud platform via TBOX, and the baseline number of dou (a unit of volume) is calculated in different time periods: the baseline number of dou for the day shift (8:00-20:00) and the baseline number of dou for the night shift (20:00-8:00 the next day).

[0040] The baseline bucket number is the mode of the number of buckets loaded on a single vehicle within this time period. If there are n modes, the baseline bucket number is (mode 1 + mode 2 + ... + mode n) / n, rounded up.

[0041] If the number of buckets loaded on a single vehicle is greater than or equal to 1.5 of the base number of buckets, the final loading quantity is increased by 1 for each occurrence; if the number of buckets loaded on a single vehicle is greater than or equal to 2n of the base number of buckets, the final loading quantity is increased by n for each occurrence; if the number of buckets loaded on a single vehicle is less than or equal to 0.5 of the base number of buckets, the final loading quantity is decreased by 1 for each occurrence.

[0042] Example 2 This embodiment provides a visual recognition system for determining the quantity of goods loaded onto a vehicle under heavy dust conditions, including the following modules: The data acquisition module is configured to acquire image and video data of the loading operation scene; The working condition recognition module is configured to: calculate the texture degradation index based on the image data; calculate the dust movement density based on the video data; calculate the contrast attenuation factor based on the contrast attenuation model; and identify heavy dust working conditions based on the texture degradation index, dust movement density, and contrast attenuation factor. The loading bucket count calculation module is configured to: obtain the time it takes for the boom to rise from the lower limit to the upper limit, and obtain the angle change from the lower limit to the upper limit of the boom; obtain a compensation coefficient based on the angle change; obtain the boom rising angle by setting the update cycle of the boom rising pilot pressure; after the material transfer is completed, determine whether it is a valid loading bucket count by the pump pressure; if so, it is counted in the loading bucket count after the unloading is completed; otherwise, it is not counted. The calibration module is configured to calculate the baseline number of buckets based on the number of buckets loaded per vehicle, and to obtain the final loading quantity based on the number of buckets loaded per vehicle and the baseline number of buckets.

[0043] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0044] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for visually identifying the quantity of goods loaded onto a vehicle under heavy dust conditions, characterized in that, include: Acquire image and video data of the loading operation scene; Based on the image data, the texture degradation index is calculated as follows: First, the image data is scaled down to obtain image data at different scales. Then, based on the image data at each scale, the Sobel operator is used to calculate the horizontal and vertical gradients to obtain the gradient value of each pixel. The average gradient value of the image data at each scale is then calculated. Finally, the minimum and maximum average gradient values ​​are selected to calculate the texture degradation index. Based on the video data, the dust motion density is calculated as follows: First, the difference between two consecutive frames in the video data is calculated to obtain a difference image. Then, the difference image is binarized. The binarization process includes setting a difference threshold. If the difference between the difference images is greater than the difference threshold, it is marked as a moving pixel, and a moving region is obtained. Finally, the moving region is filtered to obtain the number of moving pixels. Combined with the total number of pixels in the image, the dust motion density is calculated. The contrast attenuation factor is calculated based on the contrast attenuation model. The method for calculating the contrast attenuation factor is as follows: First, the image data is processed into grayscale to obtain a grayscale image, and the actual contrast of the grayscale image is calculated. Then, the calibrated contrast is obtained. Finally, the contrast attenuation factor is calculated based on the actual contrast and the calibrated contrast. Heavy dust conditions were identified based on texture degradation index, dust movement density, and contrast attenuation factor. The time it takes for the boom to rise from its lower limit to its upper limit is obtained, and the angle change from the lower limit to the upper limit is obtained. The compensation coefficient is obtained based on the angle change. The boom rising angle is obtained by setting the update cycle of the boom rising pilot pressure. After the material transfer is completed, the pump pressure is used to determine whether it is an effective loading bucket. If it is, it is counted as a loading bucket after the unloading is completed; otherwise, it is not counted. The premise for obtaining the time required for the boom to rise from its lower limit to its upper limit is that the slewing pilot pressure and the boom rising pilot pressure are both at their maximum values; the compensation coefficient obtained based on the angle change is calculated using the following formula: ;in, The compensation coefficients are for different working speeds. For working gear, This represents the change in angle between the lower and upper limits of the boom. This represents the maximum pilot pressure for boom lifting. The time it takes for the boom to rise from its lower limit to its upper limit; When the slewing pilot pressure remains at its maximum value, the boom rise angle is expressed as: ;in, The boom lifting angle, The update cycle for the boom lifting pilot pressure. The compensation coefficients are for different working speeds. For the first The value of the boom lifting pilot pressure per second. For the 2nd The value of the boom lifting pilot pressure per second. For the 3rd The value of the boom lifting pilot pressure per second. For the first The value of the boom lifting pilot pressure per second; The baseline number of buckets is calculated based on the number of buckets loaded per vehicle, and the final loading quantity is obtained based on the number of buckets loaded per vehicle and the baseline number of buckets.

2. The method for visually identifying the quantity of goods loaded onto a vehicle under heavy dust conditions as described in claim 1, characterized in that, When the boom rises at an angle greater than the preset rise angle threshold, the material transfer action is complete. If the pump pressure reaches the preset pump pressure threshold, it means there is material in the bucket, which is a valid loading count. If the pump pressure does not reach the preset pump pressure threshold, it means there is no material in the bucket, which is an invalid loading count.

3. A visual recognition system for determining the quantity of goods loaded onto trucks under heavy dust conditions, characterized in that, Includes the following modules: The data acquisition module is configured to acquire image and video data of the loading operation scene; The working condition recognition module is configured to: calculate the texture degradation index based on the image data. The method for calculating the texture degradation index is as follows: first, the image data is scaled down to obtain image data of different scales; then, based on the image data of each scale, the Sobel operator is used to calculate the horizontal and vertical gradients to obtain the gradient value of each pixel, and the average gradient value of the image data at each scale is calculated; finally, the minimum and maximum average gradient values ​​are selected to calculate the texture degradation index; and calculate the dust movement density based on the video data. The method for calculating the dust movement density is as follows: first, the difference between two consecutive frames of the video data is calculated to obtain a difference image; then, the difference image is binarized, including setting a difference threshold; if the difference between the difference images is greater than the difference threshold, it is marked as a moving pixel, and a moving region is obtained; finally, the moving region is filtered to obtain the number of moving pixels, and combined with the total number of pixels in the image, the dust movement density is calculated. The contrast attenuation factor is calculated based on the contrast attenuation model. The method for calculating the contrast attenuation factor is as follows: First, the image data is processed into grayscale to obtain a grayscale image, and the actual contrast of the grayscale image is calculated. Then, the calibrated contrast is obtained. Finally, the contrast attenuation factor is calculated based on the actual contrast and the calibrated contrast. Heavy dust conditions were identified based on texture degradation index, dust movement density, and contrast attenuation factor. The loading bucket count calculation module is configured to: obtain the time it takes for the boom to rise from the lower limit to the upper limit, and obtain the angle change from the lower limit to the upper limit of the boom; obtain a compensation coefficient based on the angle change; obtain the boom rising angle by setting the update cycle of the boom rising pilot pressure; after the material transfer is completed, determine whether it is a valid loading bucket count by the pump pressure; if so, it is counted in the loading bucket count after the unloading is completed; otherwise, it is not counted. The premise for obtaining the time required for the boom to rise from its lower limit to its upper limit is that the slewing pilot pressure and the boom rising pilot pressure are both at their maximum values; the compensation coefficient obtained based on the angle change is calculated using the following formula: ;in, The compensation coefficients are for different working speeds. For working gear, This represents the change in angle between the lower and upper limits of the boom. This represents the maximum pilot pressure for boom lifting. The time it takes for the boom to rise from its lower limit to its upper limit; When the slewing pilot pressure remains at its maximum value, the boom rise angle is expressed as: ;in, The boom lifting angle, The update cycle for the boom lifting pilot pressure. The compensation coefficients are for different working speeds. For the first The value of the boom lifting pilot pressure per second. For the 2nd The value of the boom lifting pilot pressure per second. For the 3rd The value of the boom lifting pilot pressure per second. For the first The value of the boom lifting pilot pressure per second; The calibration module is configured to calculate the baseline number of buckets based on the number of buckets loaded per vehicle, and to obtain the final loading quantity based on the number of buckets loaded per vehicle and the baseline number of buckets.

4. A computer-readable storage medium having a program stored thereon, characterized in that, When executed by the processor, the program implements the steps of the visual recognition method for the quantity of goods loaded under heavy dust conditions as described in any one of claims 1-2.

5. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the visual recognition method for loading quantity under heavy dust conditions as described in any one of claims 1-2.

Images