Loading vehicle intelligent identification management and control system based on image identification

Abstract

The invention relates to the technical field of identification, management and control of loaded vehicles, in particular to an intelligent identification, management and control system for loaded vehicles based on image identification. Comprising a server, a stock analysis module, a loading distribution module, an image recognition module and a loading accumulation module, by analyzing the coal stock state of each loading point and the conveying value of the loading vehicle in real time, loading tasks can be scientifically and reasonably distributed, and distribution of the loading tasks is dynamically adjusted according to the emergency degree of the loading tasks, the requirements of the loading points and the current position of the vehicle; by detecting the coal pile height and the point cloud density of the vehicle-mounted coal pile, the uniformity and the compactness of the coal pile can be accurately judged, and if it is found that the coal pile of the loading vehicle is in a non-uniform state, coal leveling operation is triggered so as to ensure that the coal loading quality of each loading vehicle meets the requirement; automation of the loading quality detection process is achieved, the requirement for manual inspection is greatly reduced, and the loading process is more accurate and reliable.

Description

Technical Field

[0001] The present invention relates to the technical field of loading vehicle identification and control, and in particular to an intelligent loading vehicle identification and control system based on image recognition. Background Art

[0002] Coal washing loading vehicle identification and control refers to the use of identification and management technology to monitor and manage the loading vehicles entering, operating and leaving the washing area. Its main purpose is to ensure that the quantity and volume of coal loaded on each vehicle meet the loading requirements and guide the vehicle to smoothly transport the washed finished coal out of the washing plant. Therefore, for coal washing plants, loading vehicle identification and control is particularly important;

[0003] At present, the loading vehicles of coal washing plants adopt predetermined routes for loading and transportation. This loading and transportation method lacks flexibility and is inefficient, which will lead to long waiting time at some loading points or vacant loading points, as well as congestion of loading vehicles on some routes or vehicles on some routes, thus affecting the overall operating efficiency of the coal washing plant. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent identification and control system for loading vehicles based on image recognition to solve the problems mentioned in the above background technology.

[0005] The object of the present invention can be achieved by the following technical solutions: an intelligent identification and control system for loading vehicles based on image recognition, comprising: a server, a stock analysis module, a loading allocation module, an image recognition module and a loading accumulation module;

[0006] The server is used to obtain loading and transportation information, where the loading and transportation information includes the coal inventory at each loading point, the license plate number of each loading vehicle and the corresponding loading task when it enters the coal washing plant, and the surface image of the coal pile on the vehicle;

[0007] The inventory analysis module performs real-time analysis based on the inventory status of each load to obtain the inventory status value and sends it to the load allocation module;

[0008] The loading distribution module performs priority analysis based on the loading tasks of each loading vehicle to determine the priority of each loading vehicle, and scientifically distributes the loading tasks of each loading vehicle entering the coal washing plant accordingly;

[0009] The image recognition module performs image recognition on the loaded vehicle at the loading point to determine whether it meets the loading requirements. If it meets the requirements, the number of waiting vehicles at the corresponding loading point is reduced by one, thereby updating the number of waiting vehicles at the loading point and updating it to the loading allocation module;

[0010] The loading accumulation module communicates with the RFID radio frequency device installed at the exit of the coal washing plant to identify the license plate number of each loading vehicle, and retrieves the loading task of the loading vehicle entering the coal washing plant most recently based on the license plate number. When the loading volume leaving the factory this time is consistent with the loading volume of the loading task, the loading vehicle is allowed to leave the factory, and the loading volume leaving the factory this time is accumulated into the completion volume of the loading task; the updated completion volume of the loading task is sent to the loading allocation module.

[0011] Preferably, the specific process of real-time analysis based on the coal inventory status of each load is:

[0012] The coal inventory of each loading point at each collection time is retrieved and recorded as Ci, where i = 1, 2, 3 ... I, I is a positive integer, I represents the total number of collection times, and i represents the sequence number of any collection time; a two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and coal inventory as the vertical coordinate, the coal inventory is input into the coordinate axis according to its corresponding collection time, and the position of the coal inventory in the coordinate axis is recorded as the inventory point, and a smooth curve is used to connect the inventory points in sequence to obtain a curve chart of the coal inventory value changing with time; a tangent line is drawn at each inventory point, and a tangent expression is obtained by data fitting, and the tangent expression is derived to obtain the inventory derivative corresponding to each inventory point;

[0013] The coal stock increase degree is calculated by summing up the stock derivatives greater than zero, which is recorded as M1, and the coal stock decrease degree is calculated by summing up the stock derivatives less than zero, which is recorded as M2;

[0014] The stock derivative equal to zero is recorded as the equilibrium derivative, and the quantity of the equilibrium derivative is recorded as M3;

[0015] Substitute the coal stock increase M1, coal stock decrease M2, the number of equilibrium derivatives M3 and the coal stock Ci at each collection time into the set formula The inventory status value Mg of the loading point is obtained by calculation, wherein g1, g2, and g3 are respectively the set proportional coefficients, thereby the inventory status value of each loading point can be obtained.

[0016] Preferably, the specific process of scientifically allocating the loading tasks of each loading vehicle entering the coal washing plant is as follows:

[0017] Step 1: Use RFID radio frequency technology at the entrance of the washing plant to identify the license plate number of each loading vehicle. Each loading vehicle corresponds to a loading task, where the loading task includes the loading volume recorded as Z1, the total loading volume recorded as R1, and the delivery completion time;

[0018] Step 2: Get the current time of the system, and calculate the time difference between it and the delivery completion time of the loading task to obtain the delivery time, which is recorded as R2;

[0019] Step 3: Get the completed amount corresponding to the loading task and record it as R3, and substitute it, the total amount of coal R1 and the delivery time R2 into the set formula Calculate and obtain the delivery value Ry of each loading vehicle, where y1 and y2 are respectively the set proportional coefficients; sort the loading vehicles according to their corresponding delivery values, and select the loading vehicle with the largest delivery value as the calibration loading vehicle;

[0020] Step 4: Obtain the distance between the calibrated loading vehicle and each loading point, recorded as Q1, and the number of loading vehicles waiting at each loading point, recorded as Q2, and substitute it and the inventory status value Mg of each loading point into the set formula Calculate and obtain the loading value Qy between the calibration loading vehicle and each loading point, where y3, y4, and y5 are the set proportional coefficients respectively, select the loading point with the largest loading value as the calibration loading point, and assign the loading task of the calibration loading vehicle to the calibration loading point, thereby increasing the number of loading vehicles waiting at the calibration loading point by one;

[0021] Step 5: Repeat steps 1 to 4 to distribute the loading tasks of the loading vehicles entering the coal washing plant to each loading point in turn.

[0022] Preferably, the specific process of image recognition and judgment is:

[0023] A number of monitoring points are set on the surface of the coal pile on the vehicle. The surface image of the coal pile on the vehicle is detected by image recognition technology to identify the coal pile height at each monitoring point, and they are recorded as Gj, where j = 1, 2, 3...J, J is a positive integer, J represents the total number of monitoring points on the surface of the coal pile on the vehicle, and j represents the sequence number of any monitoring point; the maximum and minimum values ​​of the coal pile height corresponding to each monitoring point are selected and recorded as Gmax and Gmin respectively; Gj, Gmax and Gmin are substituted into the set formula The highly non-mean value Gf is calculated, where f1 and f2 are the set proportional coefficients respectively. is the average height of coal pile at each monitoring point;

[0024] Generate three-dimensional point cloud data of the coal pile using laser scanning or stereo vision technology, and construct a three-dimensional stereogram of the coal pile on the vehicle based on the data, divide the three-dimensional stereogram into several areas, identify and calculate the number of point clouds in each area, record it as V1; randomly select one of the areas, calculate the distance between two adjacent points to obtain the point interval, and compare and analyze them to obtain the packing density value of the area, thereby obtaining the packing density value of each area, record it as Vfp, where p = 1, 2, 3 ... P, P is a positive integer, P represents the total number of areas in the three-dimensional stereogram, and p represents the serial number of any area;

[0025] Select the maximum and minimum values ​​of the densely packed values ​​corresponding to each region and record them as Vfmax and Vfmin respectively; substitute Vfp, Vfmax and Vfmin into the set formula The calculation is performed to obtain the compact non-mean value Uf, where f6 and f7 are the set proportional coefficients respectively. is the mean of the densely packed values ​​in each region;

[0026] The height non-mean and the tight non-mean are weightedly calculated to obtain the loading non-mean. When the loading non-mean is greater than or equal to the set non-mean threshold, the coal leveling operation is performed; otherwise, the transferred coal meets the loading requirements and no coal leveling operation is required.

[0027] Preferably, the specific process of comparing and analyzing the point spacing within the region is:

[0028] Compare and analyze the point spacing with the set spacing interval to divide the point spacing into medium near point proximity and low near point proximity; calculate the volume of the area, and count the cumulative number of high near point proximity, medium near point proximity and low near point proximity in the area, and record them as V3, V4, and V5 respectively;

[0029] Substitute the area volume V1, the number of point clouds V2, the cumulative number of highly close points V3, the cumulative number of medium close points V4, and the cumulative number of low close points V5 into the set formula The packing density value Vf of the region is obtained by calculation, wherein f3, f4, f5, and f6 are respectively set proportional coefficients, and f3>f4>f5>0; thus, the packing density value of each region can be obtained.

[0030] Beneficial effects of the present invention:

[0031] 1. By establishing a curve chart showing the change of coal inventory over time, the trend of coal inventory changes at each loading point can be monitored in real time. According to the calculation results of the inventory status value (Mg), data support can be provided for dynamic adjustment of loading strategies, so that loading points can flexibly respond to different inventory conditions and improve loading efficiency;

[0032] 2. By analyzing the coal inventory status of each loading point and the delivery value of the loading vehicle in real time, the loading tasks can be scientifically and reasonably allocated. According to the urgency of the loading task, the demand of the loading point and the current location of the vehicle, the allocation of the loading task can be dynamically adjusted to maximize the loading efficiency and reduce the idle time of the loading point and the waiting time of the loading vehicle;

[0033] 3. By detecting the coal pile height and point cloud density of the vehicle-mounted coal pile, the uniformity and compactness of the coal pile can be accurately judged. If the coal pile of the loading vehicle is found to be uneven, the coal leveling operation can be triggered to ensure that the coal loading quality of each loading vehicle meets the requirements; it can improve the coal loading quality of the loading vehicle and avoid subsequent processing and transportation problems caused by uneven loading, realize the automation of the loading quality detection process, and greatly reduce the need for manual inspection, which not only improves work efficiency, but also reduces the risk of human error, making the loading process more accurate and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be further described below in conjunction with the accompanying drawings.

[0035] Figure 1 It is a schematic diagram of system module connection of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] See also Figure 1 As shown, the present invention is an intelligent identification and control system for loading vehicles based on image recognition, comprising: a server, an inventory analysis module, a loading allocation module, an image recognition module and a loading accumulation module;

[0038] The server communicates with various sensors in each coal washing plant (specific sensors are usually installed in key locations in the coal washing plant, such as loading points, entrances and exits, etc., and specific sensors include RFID, high-definition cameras, etc.) to obtain loading and transportation information in real time, where the loading and transportation information includes the coal inventory at each loading point, the license plate number of each loading vehicle and the corresponding loading task when it enters the coal washing plant, and the surface image of the coal pile on the vehicle;

[0039] The inventory analysis module performs real-time analysis based on the inventory status of each load to obtain the inventory status value, specifically:

[0040] The coal inventory at each loading point at each collection time is retrieved (coal inventory refers to the finished coal obtained after coal washing, which is placed at each loading point to wait for the loading truck to transport it out of the washing plant), and it is recorded as Ci, where i = 1, 2, 3...I, I is a positive integer, I represents the total number of collection times, and i represents the sequence number of any collection time; a two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and coal inventory as the vertical coordinate, the coal inventory is input into the coordinate axis according to its corresponding collection time, and the position of the coal inventory in the coordinate axis is recorded as the inventory point, and a smooth curve is used to connect the inventory points in sequence to obtain the coal inventory. A curve chart showing the change of the quantity value over time; a tangent line is drawn at each inventory point, and the tangent expression is obtained by data fitting. The derivative of the tangent expression is derived to obtain the inventory derivative corresponding to each inventory point; it should be noted that when the inventory derivative is greater than zero, it means that the coal inventory at the loading point is increasing, indicating that the loading speed of the coal can be appropriately increased to prevent the loading point from being full of coal; when the inventory derivative is less than zero, it means that the coal inventory at the loading point is decreasing, indicating that the loading speed of the coal needs to be appropriately reduced to prevent the loading point from having no coal inventory during the coal loading process, but there are still loading vehicles waiting at the loading point;

[0041] The stock derivatives greater than zero are summed up to obtain the coal stock increase, which is recorded as M1; the stock derivatives less than zero are summed up to obtain the coal stock decrease, which is recorded as M2;

[0042] The stock derivative equal to zero is recorded as the equilibrium derivative, and the number of the equilibrium derivative is calculated and recorded as M3; it should be noted that when the stock derivative is equal to zero, it means that the coal stock corresponding to the stock derivative remains unchanged, and the coal loading at the loading point is in a balanced state (the coal output to the loading point by the washing process and the coal transported from the loading point are in a dynamic equilibrium state);

[0043] The coal stock increase, coal stock decrease, the number of equilibrium derivatives and the coal stock at each collection time are calculated through the set formula The inventory status value Mg of the loading point is calculated, where g1, g2, and g3 are the set proportional coefficients respectively; it can be seen from the formula that when the coal inventory is larger and the increasing trend is greater, it is necessary to increase the loading speed and perform priority loading to prevent the loading point from being full and reducing the loading efficiency; thus, the inventory status value of each loading point can be obtained;

[0044] By establishing a curve chart showing the change of coal inventory over time, the changing trend of coal inventory at each loading point can be monitored in real time. According to the calculation results of the inventory status value (Mg), data support can be provided for the dynamic adjustment of the loading strategy, so that the loading point can flexibly respond to different inventory conditions and improve loading efficiency.

[0045] The loading allocation module performs priority analysis based on the loading tasks of each loading vehicle to determine the priority of each loading vehicle, and scientifically allocates the loading tasks of each loading vehicle entering the coal washing plant accordingly, so as to achieve the maximum scientific planning of loading tasks and improve loading efficiency. Specifically:

[0046] Step 1: Use RFID radio frequency technology at the entrance of the washing plant to identify the license plate number of each loading truck. Each loading truck corresponds to a loading task, where the loading task includes the loading volume recorded as Z1, the total loading volume recorded as R1, and the delivery completion time. It should be noted that the loading volume refers to the amount of coal that the loading truck needs to load when entering the washing plant this time; the total loading volume refers to the total amount of coal that needs to be loaded for the loading task, and the delivery completion time refers to the time specified for the completion of the loading task. Specifically, a loading task number is: B22, the total loading volume is 35 tons, the amount of coal that the loading truck needs to load when entering the washing plant this time is 5 tons (loading volume), and it is specified to be completed before 8 pm on September 5 (delivery completion time);

[0047] Step 2: Get the current time of the system, and calculate the time difference between it and the delivery completion time of the loading task to get the delivery time, which is recorded as R2. It should be noted that the shorter the delivery time, the more urgent the loading task is, and the loading task should be arranged first;

[0048] Step 3: Get the completed amount corresponding to the loading task (the completed amount here refers to the amount of coal that has been transported out of the washing plant corresponding to the loading task) and record it as R3; use the set formula Calculate and obtain the delivery value Ry of each loading vehicle, where y1 and y2 are the set proportional coefficients respectively; the longer the delivery time, the smaller the delivery value; the larger the total load volume and the smaller the completed volume, the larger the delivery value; sort the loading vehicles according to their corresponding delivery values, and select the loading vehicle with the largest delivery value as the calibration loading vehicle;

[0049] Step 4: Obtain the distance between the calibrated loading vehicle and each loading point, recorded as Q1, and the number of loading vehicles waiting at each loading point, recorded as Q2; through the set formula Calculate and obtain the loading value Qy between the calibrated loading vehicle and each loading point, where y3, y4, and y5 are respectively the set proportional coefficients; it can be seen from the formula that when the inventory state value at the loading point is larger, it means that the loading point needs more loading vehicles to load coal to prevent the coal warehouse at the loading point from being full, and the loading value between the loading point and the calibrated loading vehicle is larger; when the number of loading vehicles waiting at the loading point is larger, the loading value between the loading point and the calibrated loading vehicle is smaller; when the distance between the loading point and the calibrated loading vehicle is larger, the loading value between the loading point and the calibrated loading vehicle is smaller; select the loading point with the largest loading value as the calibrated loading point, and assign the loading task of the calibrated loading vehicle to the calibrated loading point, thereby increasing the number of loading vehicles waiting at the calibrated loading point by one;

[0050] Step 5: Repeat steps 1 to 4 to sequentially assign the loading tasks of the loading vehicles entering the coal washing plant to each loading point;

[0051] By real-time analysis of the coal inventory status of each loading point and the delivery value of the loading vehicle, the loading tasks can be allocated scientifically and reasonably. According to the urgency of the loading task, the demand of the loading point and the current location of the vehicle, the allocation of loading tasks can be dynamically adjusted to maximize loading efficiency and reduce the idle time of the loading point and the waiting time of the loading vehicle.

[0052] Whenever any loading vehicle at a loading point completes the loading task, the image recognition module performs image recognition to determine whether the loading meets the requirements. If it meets the requirements, the number of loading vehicles waiting at the corresponding loading point is reduced by one. The specific steps of image recognition judgment are as follows:

[0053] A number of monitoring points are set on the surface of the coal pile on the vehicle (the coal pile on the vehicle refers to the coal loaded on the vehicle). The surface image of the coal pile on the vehicle is detected by image recognition technology to identify the coal pile height at each monitoring point, and they are recorded as Gj, where j = 1, 2, 3...J, J is a positive integer, J represents the total number of monitoring points on the surface of the coal pile on the vehicle, and j represents the sequence number of any monitoring point; the maximum and minimum values ​​of the coal pile height corresponding to each monitoring point are selected, and they are recorded as Gmax and Gmin respectively; using the set formula The highly non-mean value Gf is calculated, where f1 and f2 are the set proportional coefficients respectively. is the mean value of the coal pile height at each monitoring point; from the formula, it can be seen that the greater the difference in coal height at each monitoring point in the vehicle-mounted coal pile, the greater the height non-mean value;

[0054] Use laser scanning or stereo vision technology to generate three-dimensional point cloud data of the coal pile, and construct a three-dimensional stereogram of the vehicle-mounted coal pile based on it. The three-dimensional stereogram is divided into several areas, and the number of point clouds in each area is identified and calculated, recorded as V1; if the point cloud density of a certain area is low (the distance between points is large), it indicates that the area is loose, otherwise it is tightly stacked; take any one of the areas, calculate the distance between two adjacent points to get the point interval, and compare and analyze the point interval with the set interval. When the point interval is greater than the maximum value in the set interval, a high degree of close point adjacency is accumulated; when the point interval is in the set interval, a medium degree of close point adjacency is accumulated; when the point interval is less than the minimum value in the set interval, a low degree of close point adjacency is accumulated; the volume of the calculated area is recorded as V2, and the cumulative number of high degree of close point adjacency, medium degree of close point adjacency and low degree of close point adjacency in the area are counted respectively, and recorded as V3, V4, and V5 respectively; use the set formula The calculation is performed to obtain the area's dense packing value Vf, where f3, f4, f5, and f6 are the set proportional coefficients, and f3>f4>f5>0; it can be seen from the formula that the more point clouds there are in the area and the smaller the point spacing between the points, the denser the coal distribution in the area; thus, the dense packing value of each area can be obtained and recorded as Vfp, where p=1,2,3...P, P is a positive integer, P represents the total number of areas in the three-dimensional stereogram, and p represents the serial number of any area; the maximum and minimum values ​​of the dense packing values ​​corresponding to each area are selected, and they are recorded as Vfmax and Vfmin respectively; using the set formula The calculation is performed to obtain the compact non-mean value Uf, where f6 and f7 are the set proportional coefficients respectively. is the mean of the densely packed values ​​in each region;

[0055] The height non-mean and the compact non-mean are weighted to obtain the loading non-mean. When the loading non-mean is greater than or equal to the set non-mean threshold, it means that the coal loaded on the loading vehicle is in a non-uniform state, the safety is not high, and the coal leveling operation is required; otherwise, it means that the loaded coal meets the loading requirements and no coal leveling operation is required.

[0056] By detecting the coal pile height and point cloud density of the vehicle-mounted coal pile, the uniformity and compactness of the coal pile can be accurately judged. If the coal pile of the loading vehicle is found to be uneven, the coal leveling operation can be triggered to ensure that the coal loading quality of each loading vehicle meets the requirements; it can improve the coal loading quality of the loading vehicle and avoid subsequent processing and transportation problems caused by uneven loading, realize the automation of the loading quality detection process, and greatly reduce the need for manual inspection, which not only improves work efficiency, but also reduces the risk of human error, making the loading process more accurate and reliable.

[0057] The loading accumulation module communicates with the RFID radio frequency device installed at the exit of the coal washing plant to identify the license plate number of each loading vehicle, and retrieves the loading task of the loading vehicle entering the coal washing plant most recently according to the license plate number. When the loading volume of this shipment is consistent with the loading volume of the loading task, the loading vehicle is allowed to leave the factory, and the loading volume of this shipment is accumulated into the completion volume of the loading task;

[0058] Through detailed data recording and verification, the accuracy and completeness of the loading task are ensured, providing reliable data support for subsequent operations.

[0059] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent identification and control system for loading vehicles based on image recognition, comprising: Server, inventory analysis module, load allocation module, image recognition module and load accumulation module; characterized in that: The server is used to obtain loading and transportation information, where the loading and transportation information includes the coal inventory at each loading point, the license plate number of each loading vehicle and the corresponding loading task when it enters the coal washing plant, and the surface image of the coal pile on the vehicle; The inventory analysis module performs real-time analysis based on the inventory status of each load of coal to obtain the inventory status value, and sends it to the load allocation module; The loading distribution module performs priority analysis based on the loading tasks of each loading vehicle to determine the priority of each loading vehicle, and scientifically distributes the loading tasks of each loading vehicle entering the coal washing plant accordingly; The image recognition module performs image recognition on the loaded vehicle at the loading point to determine whether it meets the loading requirements. If it meets the requirements, the number of waiting vehicles at the corresponding loading point is reduced by one, thereby updating the number of waiting vehicles at the loading point and updating it to the loading allocation module; The loading accumulation module communicates with the RFID radio frequency device installed at the exit of the coal washing plant to identify the license plate number of each loading vehicle, and retrieves the loading task of the loading vehicle entering the coal washing plant most recently based on the license plate number. When the loading volume leaving the factory this time is consistent with the loading volume of the loading task, the loading vehicle is allowed to leave the factory, and the loading volume leaving the factory this time is accumulated into the completion volume of the loading task; the updated completion volume of the loading task is sent to the loading allocation module.

2. According to claim 1, the intelligent identification and control system for loading vehicles based on image recognition is characterized in that: The specific process of real-time analysis based on the coal inventory status of each load is as follows: Retrieve the coal inventory of each loading point at each collection time; construct a two-dimensional rectangular coordinate system with time as the horizontal coordinate and coal inventory as the vertical coordinate, input the coal inventory into the coordinate axis according to its corresponding collection time, and record the position of the coal inventory in the coordinate axis as the inventory point, and use a smooth curve to connect the inventory points in sequence to obtain a curve chart of the coal inventory value changing with time; draw a tangent line at each inventory point, use data fitting to obtain a tangent expression, and derive the tangent expression to obtain the inventory derivative corresponding to each inventory point; The stock derivatives greater than zero are summed up to obtain the coal stock increase, and the stock derivatives less than zero are summed up to obtain the coal stock decrease; Record the stock derivative that is equal to zero as the equilibrium derivative, and calculate the quantity of the equilibrium derivative; The coal stock increase, coal stock decrease, the number of equilibrium derivatives and the coal stock at each acquisition time are normalized and their values ​​are taken. The stock status value of the loading point is obtained by numerical analysis, and thus the stock status value of each loading point can be obtained.

3. According to claim 1, the intelligent identification and control system for loading vehicles based on image recognition is characterized in that: The specific process of scientifically allocating the loading tasks of each loading vehicle entering the coal washing plant is as follows: Step 1: Use RFID radio frequency technology at the entrance of the washing plant to identify the license plate number of each loading vehicle. Each loading vehicle corresponds to a loading task, where the loading task includes the loading quantity, total loading amount and delivery completion time; Step 2: Get the current time of the system, and calculate the time difference between it and the delivery completion time of the loading task to get the delivery time; Step 3: Obtain the completion amount corresponding to the loading task, normalize it with the total amount of coal and the delivery time, and take their values. Analyze the values ​​to obtain the delivery value of each loading vehicle. Sort the loading vehicles according to their corresponding delivery values, and select the loading vehicle with the largest delivery value as the calibration loading vehicle. Step 4: Obtain the distance between the calibration loader and each loading point and the number of loading vehicles waiting at each loading point, normalize them with the inventory status value of each loading point and take their values, analyze the values ​​to obtain the loading value between the calibration loader and each loading point, select the loading point with the largest loading value as the calibration loading point, and assign the loading task of the calibration loader to the calibration loading point, thereby increasing the number of loading vehicles waiting at the calibration loading point by one; Step 5: Repeat steps 1 to 4 to distribute the loading tasks of the loading vehicles entering the coal washing plant to each loading point in turn.

4. The intelligent identification and control system for loading vehicles based on image recognition according to claim 1 is characterized in that: The specific process of image recognition is as follows: A number of monitoring points are set on the surface of the coal pile on the vehicle, and the surface image of the coal pile on the vehicle is detected by image recognition technology to obtain the coal pile height at each monitoring point, and the maximum and minimum values ​​of the coal pile height corresponding to each monitoring point are selected, and the maximum and minimum values ​​are normalized with the coal pile height of each monitoring point and the values ​​are taken, and the height non-mean value is obtained by numerical analysis; Generate three-dimensional point cloud data of the coal pile using laser scanning or stereo vision technology, and form a three-dimensional stereogram of the coal pile on the vehicle based on the data. Divide the three-dimensional stereogram into several areas, identify and calculate the number of point clouds in each area; randomly select one of the areas, calculate the distance between two adjacent points to obtain the point interval, and compare and analyze them to obtain the area's compactness value, thereby obtaining the compactness value of each area; The maximum and minimum values ​​of the compactness values ​​corresponding to each region are selected, and the compactness values ​​of each region are normalized and taken to obtain the compactness non-mean value through numerical analysis; The height non-mean and the tight non-mean are weightedly calculated to obtain the loading non-mean. When the loading non-mean is greater than or equal to the set non-mean threshold, the coal leveling operation is performed; otherwise, the transferred coal meets the loading requirements and no coal leveling operation is required.

5. The intelligent identification and control system for loading vehicles based on image recognition according to claim 4 is characterized in that: The specific process of comparative analysis of point spacing within a region is as follows: Compare and analyze the point spacing with the set spacing interval to divide the point spacing into medium near point proximity and low near point proximity; calculate the volume of the area, and count the cumulative number of high near point proximity, medium near point proximity and low near point proximity in the area respectively; The regional volume, the number of point clouds, the cumulative number of highly close point neighbors, moderately close point neighbors and low close point neighbors are normalized and their values ​​are taken, and the numerical analysis is performed to obtain the regional packing density value; thus, the packing density value of each region can be obtained.

Images