Method and device for determining covering state of shielding body

Abstract

The invention discloses a method and a device for determining a covering state of a shielding body. The method comprises the following steps: acquiring video stream data of a target vehicle; detecting the video stream data to obtain a first detection result; detecting photosensitive sensor data of the target vehicle to obtain a second detection result; and determining the covering state of the shielding body corresponding to the first detection result and the second detection result. According to the method and the device, the technical problem that the detection result is inaccurate due to the fact that a single visual detection mode is adopted to identify the tarpaulin covering state of the vehicle in the prior art and is easily influenced by factors such as illumination and weather is solved.

Description

Technical Field

[0001] This application relates to the field of obstruction detection technology, and more specifically, to a method and apparatus for determining the coverage state of an obstruction. Background Technology

[0002] Construction waste transport vehicles play a vital role in urban construction. However, the failure to properly cover these vehicles with tarpaulins during transport leads to the spillage of cargo, polluting the environment, affecting traffic safety, and increasing urban maintenance costs. Relevant technologies often rely on single detection methods to assess the covering status, such as traditional visual image recognition algorithms. These methods are prone to inaccurate identification or missed detections in the complex and changing environments of construction sites and roads, and are easily affected by factors such as lighting and weather, resulting in inaccurate detection results.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method and apparatus for determining the coverage status of an obstruction, thereby at least solving the technical problem that related technologies use a single visual detection method to identify the tarpaulin coverage status of a vehicle, which is easily affected by factors such as lighting and weather, leading to inaccurate detection results.

[0005] According to one aspect of the embodiments of this application, a method for determining the coverage state of an obstruction is provided, comprising: acquiring video stream data of a target vehicle, wherein the target vehicle includes a vehicle covered by an obstruction, the obstruction being used to obstruct materials carried by the target vehicle; detecting the video stream data to obtain a first detection result, wherein the first detection result includes a coverage state determined based on visual image information of the video stream; detecting photosensitive sensor data of the target vehicle to obtain a second detection result, wherein the photosensitive sensor that collects the photosensitive sensor data is disposed in the obstruction area corresponding to the obstruction, and the second detection result includes a coverage state of the obstruction area determined based on light information; and determining the obstruction coverage state corresponding to both the first detection result and the second detection result.

[0006] In some embodiments of this application, detecting video stream data to obtain a first detection result includes: using a deep learning model to detect the video stream data to obtain occlusion coverage information of the target vehicle, wherein the occlusion coverage information includes the coverage area of ​​the occlusion; determining the occlusion coverage rate corresponding to the coverage area; comparing the occlusion coverage rate with a first preset threshold to obtain a first comparison result; determining a first coverage state of the occlusion based on the first comparison result, and using the first coverage state as the first detection result, wherein the first coverage state includes the coverage state identified based on visual image information.

[0007] In some embodiments of this application, detecting the photosensitive sensor data of the target vehicle to obtain a second detection result includes: receiving photosensitive sensor data collected by the photosensitive sensor; determining the light intensity of the occluded area based on the photosensitive sensor data; comparing the light intensity with a second preset threshold to obtain a second comparison result; determining the second coverage state of the occluded body based on the second comparison result, and using the second coverage state as the second detection result, wherein the second coverage state includes the coverage state identified based on the light intensity information.

[0008] In some embodiments of this application, the method further includes: determining the light color of the occluded area based on data from a photosensitive sensor; comparing the light color with a color feature library to obtain a third comparison result, wherein the color feature library includes preset color representations of the occluded body under different lighting conditions and color features of the materials transported by the target vehicle in an uncovered state; and determining a second coverage state based on the second comparison result and the third comparison result.

[0009] In some embodiments of this application, determining the occlusion coverage state corresponding to both the first detection result and the second detection result includes: comparing the first detection result with the second detection result to obtain a fourth comparison result; if the fourth comparison result indicates that the first detection result and the second detection result are consistent, determining the first detection result or the second detection result as the occlusion coverage state; if the fourth comparison result indicates that the first detection result and the second detection result are inconsistent, fusing the video stream data and the photosensitive sensor data to obtain fused data, and determining the occlusion coverage state based on the fused data.

[0010] In some embodiments of this application, before comparing the first detection result with the second detection result to obtain the fourth comparison result, the method further includes: using a deep learning model to detect video stream data to obtain vehicle information of the target vehicle, wherein the vehicle information includes the first license plate number of the target vehicle; obtaining the second license plate number corresponding to the vehicle where the photosensitive sensor is located; comparing the first license plate number with the second license plate number to obtain the fifth comparison result; and comparing the first detection result with the second detection result if the fifth comparison result indicates that the first license plate number and the second license plate number are consistent.

[0011] In some embodiments of this application, the method further includes: determining the type of the obstruction coverage state, wherein the type includes a complete coverage state, a partial coverage state, and an uncovered state, wherein the coverage area of ​​the first obstruction corresponding to the complete coverage state is greater than the coverage area of ​​the second obstruction corresponding to the partial coverage state, and the coverage area of ​​the second obstruction is greater than the coverage area of ​​the third obstruction corresponding to the uncovered state; and generating alarm information when the type is not a complete coverage state.

[0012] According to another aspect of the embodiments of this application, a device for determining the coverage state of an obstruction is also provided, comprising: an acquisition module for acquiring video stream data of a target vehicle, wherein the target vehicle includes a vehicle covered by an obstruction, the obstruction being used to obstruct materials carried by the target vehicle; a first detection module for detecting the video stream data to obtain a first detection result, wherein the first detection result includes a coverage state determined based on visual image information of the video stream; a second detection module for detecting photosensitive sensor data of the target vehicle to obtain a second detection result, wherein the photosensitive sensor that collects the photosensitive sensor data is disposed in the obstruction area corresponding to the obstruction, and the second detection result includes a coverage state of the obstruction area determined based on light information; and a determination module for determining the coverage state of the obstruction corresponding to both the first detection result and the second detection result.

[0013] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute the method for determining the coverage state of the obstruction as described above.

[0014] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored computer program, wherein the device where the non-volatile storage medium is located executes the above-mentioned method for determining the coverage state of the obstruction by running the computer program.

[0015] According to another aspect of the embodiments of this application, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the method for determining the coverage state of the obstruction.

[0016] In this embodiment, a dual-modal detection fusion approach is adopted. By combining visual image information and light intensity information, a first detection result is generated based on the visual image information, reflecting the coverage status of the occluder. A second detection result is derived based on the light information to confirm the degree of light occlusion in the occluded area. The first and second detection results are then compared comprehensively to obtain a comprehensive and accurate assessment of the coverage status. This effectively overcomes the susceptibility of single-modal detection to environmental interference, thereby ensuring the accuracy of the detection results. Furthermore, it solves the technical problem that related technologies using a single visual detection method to identify the tarpaulin coverage status of vehicles are easily affected by factors such as lighting and weather, leading to inaccurate detection results. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a hardware structure block diagram of a computer terminal for a method of determining the coverage state of an obstruction according to an embodiment of this application. Figure 2 This is a flowchart of a method for determining the coverage state of an obstruction according to an embodiment of this application; Figure 3 This is a schematic diagram of a device for determining the coverage state of an obstruction according to an embodiment of this application. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] Dump trucks, as essential tools for transporting materials between construction sites and cities, play an indispensable role in urban construction. However, if dump trucks are not properly covered with tarpaulins during transport, the loaded construction waste and materials can scatter due to vibration and bumps during vehicle movement, causing serious environmental pollution. This not only affects the cleanliness of roads within construction sites but also poses a serious threat to the urban public road landscape and traffic safety. Furthermore, construction waste spillage increases the workload of sanitation departments, raises urban maintenance costs, and may even trigger dust pollution, affecting air quality.

[0021] Currently, relevant departments have issued a series of regulations requiring dump trucks to be fully covered with tarpaulins during transportation to prevent spillage. However, due to the complex environment of construction sites and the high mobility of vehicles, it is difficult to achieve real-time and comprehensive supervision of the tarpaulin coverage of dump trucks by relying solely on manual inspections.

[0022] While some computer vision-based vehicle detection and recognition technologies have been applied to traffic monitoring, there are still many challenges in accurately identifying the tarpaulin coverage status of dump trucks. For example, the diverse shapes of dump trucks and the varied colors and materials of their tarpaulins, coupled with the influence of factors such as lighting, weather, and occlusion, result in low accuracy of traditional template matching or simple feature extraction methods in practical applications. Furthermore, the high speed and significant changes in posture of dump trucks in surveillance videos place high demands on the stability and robustness of target tracking.

[0023] To address the aforementioned technical problems, this application provides corresponding solutions, which are detailed below.

[0024] The method for determining the coverage state of an obstruction provided in this application can be executed on a mobile terminal, a computer terminal, or a similar computing device. Figure 1 A hardware block diagram of a computer terminal for determining the coverage state of an obstruction is shown. Figure 1 As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions connected via wired and / or wireless networks. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0025] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0026] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for determining the coverage state of an obstruction in this embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned method for determining the coverage state of an obstruction. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0027] The transmission module 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 106 may be a radio frequency (RF) module, used for wireless communication with the Internet.

[0028] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0029] It should be noted here that, in some optional embodiments, the above... Figure 1 The computer terminal shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 1 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0030] In the above operating environment, this application provides an embodiment of a method for determining the coverage state of an obstruction. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] Figure 2 This is a flowchart of a method for determining the coverage state of an obstruction according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps: Step S202: Obtain video stream data of the target vehicle, wherein the target vehicle includes a vehicle covered by an obstruction, the obstruction being used to obstruct the materials carried by the target vehicle.

[0032] In step S202 above, the video stream data is a continuous sequence of video images, which is captured in real time by the camera and transmitted to the smart terminal or backend server. The video stream data contains dynamic image information of the target vehicle and its obstructions.

[0033] To address the issue that static images cannot reflect the dynamic changes of vehicles, especially in high-speed moving environments, capturing continuous video stream data can ensure real-time monitoring of the coverage status of obstructions. It should be noted that the video stream data can be captured by surveillance cameras set up on the road or in a specific area, or by cameras mounted on vehicles; there are no restrictions on this.

[0034] In some embodiments of this application, the dump truck image data in the video stream captured by the camera includes front, side, and rear images of the vehicle. Acquiring these images provides more comprehensive information about the vehicle's appearance, helping to identify its overall features, markings, license plate number, and overall condition. Furthermore, images from different angles provide multi-angle monitoring of the dump truck, which helps to identify potential problems in different parts of the vehicle, such as loading status, leaks, and damage. Moreover, images from multiple angles increase the accuracy of vehicle feature identification.

[0035] In addition, images from different angles can provide more details and features, which helps reduce the possibility of misidentification and missed identification. Images from multiple angles can monitor the operating status of dump trucks in real time. If abnormalities are found, such as vehicle tilting or cargo spillage, early warnings can be issued in time so that corresponding measures can be taken to avoid safety accidents. Multi-angle image data can be used for data analysis to understand the operating patterns and loading conditions of dump trucks.

[0036] Step S204: Detect the video stream data to obtain a first detection result, wherein the first detection result includes the coverage status determined based on the visual image information of the video stream.

[0037] In step S204 above, the first detection result is the occlusion coverage status information obtained based on the analysis of video stream data and combined with a deep learning model. In some embodiments of this application, the first detection result may include, but is not limited to, vehicle information such as the overall characteristics, identification, license plate number, and vehicle condition of the vehicle, as well as whether the occlusion (such as a tarpaulin) completely, partially, or not covers the materials loaded on the vehicle.

[0038] In some embodiments of this application, pre-trained deep learning models, such as YOLO (You Only Look Once) and Faster R-CNN, can be used to perform real-time target detection and recognition on the video stream of the target vehicle. Specifically, the deep learning model extracts visual features such as texture, contour, and color from the video frames using a convolutional neural network (CNN), and uses pre-trained weights to predict the position and category of the target (vehicle and occluder). Regarding the coverage state of the occluder, the model can further identify the contact area and degree of occlusion between the tarpaulin and the vehicle.

[0039] To accurately quantify the coverage of obstructions, the first detection result can be obtained as follows: A deep learning model is used to detect the video stream data to obtain the obstruction coverage information of the target vehicle, where the obstruction coverage information includes the coverage area of ​​the obstruction; the obstruction coverage rate corresponding to the coverage area is determined; the obstruction coverage rate is compared with a first preset threshold to obtain a first comparison result; based on the first comparison result, a first coverage state of the obstruction is determined, and the first coverage state is used as the first detection result, where the first coverage state includes the coverage state identified based on visual image information.

[0040] Specifically, a pre-trained deep learning model can be used to process each frame of video, identify the target vehicle and the obstruction areas on it, calculate the number of pixels in the obstruction areas, and convert them into actual physical areas. Then, the calculated obstruction coverage area is divided by the total area of ​​the vehicle's loading area to obtain the obstruction coverage rate, which is then compared with a first preset threshold. It should be noted that the first preset threshold is a minimum coverage standard pre-set according to relevant regulations or transportation specifications, used to determine whether the obstruction meets transportation requirements.

[0041] In some specific embodiments of this application, a deep learning model can be used to analyze the video stream captured by the camera in real time, detect vehicle information and tarpaulin coverage, and distinguish between three states: "complete coverage," "partial coverage," and "no coverage" by setting thresholds. Specifically, the following steps are included: (1) Use a deep learning model to analyze the video stream captured by the camera in real time, and extract the vehicle's feature information and the area covered by the tarpaulin; (2) Identify the vehicle type, license plate number, and other information based on the vehicle's characteristic information; (3) Calculate the coverage area and coverage rate of the tarpaulin based on the coverage area of ​​the tarpaulin; (4) Set a threshold and compare the coverage of the tarpaulin with the threshold to distinguish between the three states of “complete coverage”, “partial coverage” and “no coverage”.

[0042] The aforementioned solution can analyze video streams in real time, promptly detect vehicle information and tarpaulin coverage, and provide real-time monitoring and feedback. The deep learning model can extract vehicle feature information and tarpaulin coverage area with high accuracy and reliability. It can not only identify vehicle type and license plate number but also calculate tarpaulin coverage area and coverage rate, providing more comprehensive information. By setting thresholds, it can accurately distinguish between "complete coverage," "partial coverage," and "no coverage," facilitating the assessment and management of tarpaulin coverage. The automated detection and analysis process reduces manual intervention, improves work efficiency, lowers costs, and provides timely information on vehicle and tarpaulin status, helping to strengthen the management and supervision of the transportation process and ensure the safety and compliance of goods.

[0043] Step S206: Detect the photosensitive sensor data of the target vehicle to obtain a second detection result. The photosensitive sensor that collects the photosensitive sensor data is set in the occlusion area corresponding to the occlusion body. The second detection result includes the coverage state of the occlusion area determined based on light information.

[0044] In step S206 above, the photosensitive sensor is a device that converts the received light intensity into an electrical signal. The photosensitive sensor can be installed in the loading area under the shield to monitor the light intensity under the shield in real time. The second detection result is based on the photosensitive sensor data, and the result of analyzing the changes in light intensity in the loading area to determine the shield coverage status is used to supplement and verify the first detection result (based on visual image information).

[0045] In some embodiments of this application, a photosensitive sensor array can be deployed at key locations such as the floor or sidewalls of the vehicle loading area. Data from these sensors is read periodically or continuously, and changes in light intensity are recorded. Theoretically, the light intensity of the sensors will decrease to its minimum when the obstruction completely covers the loading area, and increase otherwise. By setting a reasonable light intensity threshold, the coverage status of the obstruction can be determined.

[0046] In some embodiments of this application, a photosensitive sensor array can be installed around the top of the target vehicle's (e.g., a dump truck's) cargo compartment. A smart terminal receives the data collected by the photosensitive sensors, determines the dump truck's coverage status based on the data, records the sensor detection results, and transmits the data wirelessly to a backend server. It should be noted that the smart terminal can be linked to vehicle information. Vehicle information can be entered into the smart terminal's internal database and a binding relationship established through magnetic card input and verification.

[0047] By installing photosensitive sensor arrays at key locations inside the dump truck compartment, changes in light transmission can be monitored in real time and accurately, thereby obtaining real-time data on the coverage status of the dump truck. This improves the accuracy and reliability of monitoring, avoids the subjectivity and errors of manual monitoring, and allows the intelligent terminal to record the detection results of the photosensitive sensors and transmit the data to the backend server wirelessly. This enables real-time data transmission and remote monitoring, allowing managers to understand the coverage status of the dump truck in a timely manner and take appropriate measures to improve management efficiency.

[0048] In addition, after receiving the video stream, the backend server can use a deep learning model to detect license plate numbers in each frame, which can improve the accuracy and efficiency of license plate number recognition. By analyzing the texture, color and other features of the image, the deep learning model can accurately identify the license plate number and associate it with the vehicle information (vehicle information bound to the smart terminal), which facilitates the management and tracking of dump trucks.

[0049] In some embodiments of this application, the photosensitive sensor array can also be installed around the top of the interior of the dump truck, including but not limited to the top of the four sides of the truck. By installing sensors in different locations, the light transmission inside the dump truck can be comprehensively monitored, ensuring accurate perception of the covering status. Whether it is the covering material on the top, the accumulation at the bottom, or leakage on the sides, it can be detected in a timely manner. The arrangement of sensors in multiple locations can provide richer data, which helps to improve the accuracy of monitoring.

[0050] Furthermore, changes in light at different locations can corroborate each other, reducing the possibility of misjudgments and omissions. Installing sensors at key locations can avoid the occurrence of blind spots in monitoring. The distribution and covering of materials inside the dump truck may change due to loading methods and shaking during travel. By installing sensors at multiple locations, it is possible to better adapt to these changes and ensure accurate monitoring of the covering status under various conditions. When the covering status inside the dump truck becomes abnormal, sensors at multiple locations can issue early warning signals more quickly. If one sensor malfunctions or is interfered with, other sensors can still provide effective monitoring information.

[0051] It should be noted that binding smart terminals with vehicle information may include the following steps: (1) Magnetic card entry and verification: Each target vehicle (taking a dump truck as an example) is equipped with a unique magnetic card. The magnetic card contains basic information such as the vehicle's license plate number and the unit to which it belongs. The smart terminal reads the magnetic card information through the built-in magnetic card reader and performs preliminary verification to ensure the authenticity and validity of the information.

[0052] (2) Smart terminal binding: After verification, the smart terminal will enter the vehicle information in the magnetic card into its internal database and establish a binding relationship with the vehicle.

[0053] Each dump truck is equipped with a unique magnetic card, ensuring the accuracy and uniqueness of vehicle information. Reading the magnetic card information through a magnetic card reader can avoid manual input errors, improve the reliability of information, and the smart terminal can quickly read the magnetic card information and perform preliminary verification, reducing the time and workload of manual verification.

[0054] After successful verification, vehicle information is automatically entered into the internal database and a binding relationship is established, improving operational efficiency. Once bound, the smart terminal can obtain relevant vehicle information in real time, such as license plate number and affiliated unit, facilitating real-time monitoring and management of dump trucks and timely understanding of their operational status. The vehicle information in the magnetic card is stored in a chip, making it difficult to tamper with, ensuring the security and integrity of vehicle information and preventing criminals from forging or altering it. Through the binding of the smart terminal with vehicle information, centralized management and dispatching of dump trucks can be achieved. Managers can use the terminal to understand the vehicle's location, route, and other information in real time, optimizing transportation arrangements and improving management efficiency. When it is necessary to trace vehicle information or make inquiries, relevant information can be quickly obtained simply through the smart terminal.

[0055] In some embodiments of this application, the second detection result can be determined by the following steps: receiving photosensitive sensor data collected by a photosensitive sensor; determining the light intensity of the occluded area based on the photosensitive sensor data; comparing the light intensity with a second preset threshold to obtain a second comparison result; determining the second coverage state of the occluded body based on the second comparison result, and using the second coverage state as the second detection result, wherein the second coverage state includes the coverage state identified based on the light intensity information.

[0056] Specifically, the smart terminal continuously receives data from these sensors, representing light intensity as an electrical signal. It then compares the light intensity of the obstructed area with a second preset threshold to obtain a second comparison result. For example, the smart terminal or backend server may pre-set a second preset threshold, which is determined based on the average light intensity when the obstruction completely covers the loading area. By comparing the real-time collected light intensity data with the threshold, if the light intensity is much lower than the threshold, it indicates that the obstruction completely covers the area; if it is close to or higher than the threshold, it may indicate that the obstruction partially covers or does not cover the area.

[0057] In some specific embodiments of this application, the smart terminal uses a photosensitive sensor to detect changes in light intensity in the area covered by a tarpaulin (a form of obstruction) in real time, and sets a threshold. When the tarpaulin is completely covered, the photosensitive sensor is blocked, and the light intensity detected by the photosensitive sensor should be lower than the threshold. When the tarpaulin is not covered or only partially covered, part of the photosensitive sensor is blocked, and the light intensity detected by the part of the photosensitive sensor should be higher than the threshold. Specifically, this includes the following steps: (1) Install photosensitive sensors in the area covered by the tarpaulin. The number and location of the photosensitive sensors shall be determined according to the actual situation. (2) Real-time detection of light intensity changes in the photosensitive sensor and comparison with a set threshold; (3) When the tarpaulin is completely covered, the photosensitive sensor is blocked, and the light intensity detected by the photosensitive sensor should be lower than the threshold. (4) When the tarpaulin is not covered or is partially covered, part of the photosensitive sensor is blocked, and the light intensity detected by part of the photosensitive sensor should be higher than the threshold.

[0058] The above solution combines deep learning models and photosensitive sensor detection to more accurately determine the coverage status of the tarpaulin. The deep learning model can provide preliminary analysis results, while the photosensitive sensor can provide real-time light intensity data. The two verify each other and reduce the possibility of misjudgment.

[0059] A photosensitive sensor can determine the tarpaulin's coverage status based on changes in light intensity, unaffected by factors such as vehicle color or shape. This method works effectively under various lighting conditions, improving the system's adaptability. Setting a threshold can avoid false alarms caused by changes in ambient light or other interference factors. Only when the light intensity change exceeds the set threshold will it be considered a change in the tarpaulin's coverage status, improving detection reliability. Even if the deep learning model malfunctions or makes a misjudgment, the photosensitive sensor can still provide a certain degree of assurance, ensuring accurate monitoring of the tarpaulin's coverage status. The light intensity data detected by the photosensitive sensor can be recorded for subsequent analysis and research.

[0060] To address the potential impact of different materials and colors of obstructions on the accuracy of light intensity detection, the following steps can be performed: determine the light color of the obstructed area based on data from the photosensitive sensor; compare the light color with a color feature library to obtain a third comparison result, wherein the color feature library includes preset color representations of the obstruction under different lighting conditions and the color characteristics of the materials transported by the target vehicle in an uncovered state; and determine a second coverage state based on the second and third comparison results.

[0061] In some embodiments of this application, in addition to detecting changes in light intensity, changes in light color can also be detected. Changes in light intensity can provide direct information about the degree of soil cover; when there is more soil cover, the light transmission intensity decreases, while when there is less cover, the light transmission intensity increases. Changes in light color can also provide additional information, such as the color and composition of the soil, helping to more accurately determine the cover status, as environmental light conditions may change, such as day and night, sunny and cloudy days.

[0062] Considering changes in light intensity and color allows monitoring systems to better adapt to different environmental conditions, improving system stability and reliability. By comprehensively considering these changes, the possibility of false alarms and missed detections can be reduced. For example, relying solely on light intensity may not be sufficient to distinguish the covering status under certain special circumstances, while combining light color information can provide a more accurate assessment. Anomalies in light transmission may indicate a malfunction or problem with the dump truck covering system. By monitoring changes in light intensity and color, these anomalies can be detected promptly, enabling early warnings and fault detection, thus avoiding potential safety risks and environmental pollution.

[0063] In some embodiments of this application, the color feature library stores the preset color representation of the obstruction under different lighting conditions, as well as the color features of the materials transported by the target vehicle in the uncovered state, which are used to compare the color of the light detected in real time and assist in determining the coverage status of the obstruction.

[0064] Specifically, photosensitive sensors with color perception capabilities can be deployed in the obstructed area. These sensors, in addition to measuring light intensity, can also sense the color information of the light. The smart terminal receives this data and analyzes the light color using algorithms. A color feature library is maintained in the smart terminal or backend server, which contains the preset color representation of the obstructing object and the color characteristics of the materials transported by the target vehicle under different lighting conditions. By using machine learning algorithms, the coverage of the obstructing object is determined by comparing the real-time received light color data with the color features in the library.

[0065] For example, the color feature data of the shading object and the material under different lighting conditions are stored in a color feature library. Each feature entry includes the color value under specific lighting conditions and the corresponding environmental conditions (such as light intensity, time, weather, etc.). The real-time received light color data is compared with the preset color representations in the color feature library. For example, machine learning algorithms are used to analyze whether the color feature of the shading object under the current lighting conditions is consistent with the preset color of the shading object in the library, as well as the color feature of the material when the shading object is not covered. The tarpaulin coverage is judged based on the comprehensive comparison results. For example, if the real-time light color data matches the color feature of the shading object in the library, and the light intensity is as expected (i.e., below the threshold), then the shading object is considered to be fully covered. If the color feature deviates, but the intensity data still meets the shading conditions, it means that the shading object is partially damaged or not completely covered. Conversely, if the color feature meets the expectations but the intensity data is abnormally high, it indicates that the shading object is not covered or partially uncovered.

[0066] It should be noted that the color characteristics of the material determined in the above steps when the obstruction is not covering it are used as a reference. When the color data of the light detected in real time does not match the expected color characteristics under the obstruction, but matches the color characteristics of the material, this can be used as a signal that the obstruction may not cover or partially cover the material.

[0067] Step S208: Determine the coverage status of the occlusion that corresponds to both the first and second detection results.

[0068] In step S208 above, the covering state of the shield is the tarpaulin covering state that is finally determined by combining the first detection result and the second detection result.

[0069] In some embodiments of this application, logical operation rules can be used to perform fusion analysis on the first detection result and the second detection result. If the two results are consistent (i.e., both visual and sensor data point to the same occlusion state), the coverage state of the occlusion is confirmed; if they are inconsistent, further data analysis and verification are performed.

[0070] Alternatively, a decision tree model can be created to comprehensively analyze the first and second detection results. Nodes in the decision tree represent different judgment conditions (such as whether the light intensity is lower than a preset threshold, or whether the light color matches the records in the feature library), and leaf nodes represent the final judgment of the occlusion coverage status. Judgments are made along the decision tree path based on real-time detection data, and the final output is the occlusion coverage status.

[0071] Specifically, decision tree nodes and leaf nodes can be defined as follows: (1) Decision nodes: whether the light intensity is lower than the preset threshold; whether the light color matches the preset features in the color feature library; whether the visual detection result is consistent with the sensor detection result.

[0072] (2) Leaf node: The occluder is completely covered; the occluder is partially covered; the occluder is not covered.

[0073] (3) Execution steps: Based on the first and second detection results, starting from the root node of the decision tree, determine in sequence whether the conditions of each decision node are met. If the light intensity is lower than the threshold and the light color matches the feature library, proceed along the "yes" path of the decision tree; if the light intensity is higher than the threshold or the light color does not match, determine whether the visual detection result is consistent with the sensor detection result. Based on the path of the decision tree, the system outputs the final occlusion coverage status judgment (leaf node).

[0074] To improve the accuracy of occlusion coverage determination, the occlusion coverage state corresponding to both the first and second detection results can be determined in the following way: compare the first and second detection results to obtain a fourth comparison result; if the fourth comparison result indicates that the first and second detection results are consistent, determine the first or second detection result as the occlusion coverage state; if the fourth comparison result indicates that the first and second detection results are inconsistent, fuse the video stream data and the photosensitive sensor data to obtain fused data, and determine the occlusion coverage state based on the fused data.

[0075] Specifically, the visual coverage state of the occluded object obtained by the deep learning model (first detection result) is compared with the real-time coverage state of the occluded object detected by the photosensor (second detection result). If the two are consistent, the consistent result is directly taken as the coverage state of the occluded object (no further processing is required); if they are inconsistent, the subsequent fusion data analysis stage is initiated.

[0076] When the fourth comparison result indicates that the two sets of detection results are inconsistent, the system initiates a data fusion algorithm. First, the video stream data acquired by the deep learning model is converted into feature vectors; second, the light intensity and color data from the photosensor are standardized and converted into digital features. Finally, these two feature vectors are combined, and a secondary analysis is performed using a machine learning or deep learning model to determine the occlusion coverage status with more comprehensive information.

[0077] Before comparing the first detection result with the second detection result to obtain the fourth comparison result, the following steps can be performed: using a deep learning model to detect the video stream data to obtain the vehicle information of the target vehicle, wherein the vehicle information includes the first license plate number of the target vehicle; obtaining the second license plate number corresponding to the vehicle where the photosensitive sensor is located; comparing the first license plate number with the second license plate number to obtain the fifth comparison result; if the fifth comparison result indicates that the first license plate number and the second license plate number are consistent, comparing the first detection result with the second detection result.

[0078] Specifically, the video stream data captured by the camera is first fed into a specially trained deep learning model. The model can identify and extract the features of the vehicle in the video, including the vehicle outline, brand, color, and most importantly, the first license plate number. The magnetic card reader built into the smart terminal reads the vehicle's magnetic card information. The magnetic card stores the vehicle's basic information, including the license plate number. The smart terminal or backend server then uses the magnetic card information to obtain the second license plate number of the vehicle bound to the photosensitive sensor.

[0079] The two license plate numbers are compared at the string level to check if they are completely identical, resulting in a fifth comparison result. If the fifth comparison result indicates that the first license plate number matches the second license plate number, the first detection result is compared with the second detection result.

[0080] When multiple vehicles enter the monitoring range simultaneously or there are multiple vehicles in the video stream, license plate number comparison is a key step to ensure that the detection results of the deep learning model correspond to the correct vehicle with the data from the photosensitive sensor. Only when the fifth comparison result shows that the two license plates match will the subsequent occlusion detection results be associated, thus improving the accuracy and reliability of the system.

[0081] In some embodiments of this application, the following steps may also be performed: determining the type of the obstruction coverage state, wherein the type includes a complete coverage state, a partial coverage state, and an uncovered state, wherein the coverage area of ​​the first obstruction corresponding to the complete coverage state is greater than the coverage area of ​​the second obstruction corresponding to the partial coverage state, and the coverage area of ​​the second obstruction is greater than the coverage area of ​​the third obstruction corresponding to the uncovered state; generating alarm information when the type is not a complete coverage state.

[0082] Taking a dump truck as an example, in some embodiments of this application, the backend platform can match the detection results of the deep learning model (first detection result) with the detection results of the smart terminal (second detection result) based on the license plate number to ensure that the two correspond to the same dump truck. The backend platform compares the output of the deep learning model with the detection results of the sensor. When the comparison results are consistent, the detection result is correct. At this time, if the detection results are consistent, three results will appear: complete coverage, partial coverage, and no coverage. When the detection result is partial coverage or no coverage, an alarm mechanism is triggered. If the comparison results are inconsistent, the detection result is incorrect. The incorrect result is analyzed more deeply through the video stream or image captured by the camera. The light intensity change detected by the photosensitive sensor is compared by fusing data from multiple modalities and the result is output. If the detection result is still incorrect, an alarm mechanism is triggered.

[0083] Through steps S202 to S208, a dual-modal detection fusion method is adopted. By combining visual image information and light intensity information, a first detection result is generated based on the visual image information, reflecting the coverage status of the occluder. A second detection result is derived based on the light information to confirm the degree of light occlusion in the occluded area. The first and second detection results are then compared comprehensively to obtain a comprehensive and accurate judgment of the coverage status. This effectively overcomes the susceptibility of single-modal detection to interference from environmental factors, thereby ensuring the accuracy of the detection results. It also solves the technical problem that related technologies using a single visual detection method to identify the tarpaulin coverage status of vehicles are easily affected by factors such as lighting and weather, leading to inaccurate detection results.

[0084] To facilitate understanding of the method for determining the coverage state of the obstruction described above, the following explanation is provided in conjunction with some specific embodiments.

[0085] (1) Example 1, including the following steps: S110: Binding of smart terminal and vehicle information: Vehicle information is entered into the internal database of the smart terminal through magnetic card input and verification, and a binding relationship is established.

[0086] S210: Uses a deep learning model to analyze the video stream captured by the camera in real time to detect vehicle information and tarpaulin coverage.

[0087] S310: Intelligent terminal secondary detection: Real-time detection of light intensity changes in the area covered by the tarpaulin using a photosensitive sensor array.

[0088] S410: Platform Data Matching and Result Comparison: The backend platform matches the detection results of the deep learning model with the detection results of the smart terminal based on the license plate number to ensure that they correspond to the same dump truck. The backend platform compares the output results of the deep learning model with the detection results of the sensors. When the comparison results are consistent, the detection result is correct. In this case, if the detection results are consistent, three results will appear: complete coverage, partial coverage, and no coverage. When the detection result is partial coverage or no coverage, an alarm mechanism is triggered. If the comparison results are inconsistent, the detection result is incorrect. The incorrect result is analyzed more deeply through the video stream or image captured by the camera. The light intensity change detected by the photosensitive sensor is compared by fusing data from multiple modalities before outputting the result. If the detection result is still incorrect, an alarm mechanism is triggered.

[0089] S510: Uses GPS to locate vehicles and outputs the status of tarpaulin coverage.

[0090] (2) Example 2 includes the following steps: S120: Smart Terminal and Vehicle Information Binding: Each dump truck is equipped with a unique magnetic card containing basic information such as the vehicle's license plate number and affiliated unit. The smart terminal reads the information from the magnetic card using its built-in card reader and performs preliminary verification to ensure the authenticity and validity of the information. After successful verification, the smart terminal enters the vehicle information from the magnetic card into its internal database and establishes a binding relationship with the vehicle.

[0091] S220: A deep learning model is used to analyze the video stream captured by the camera in real time, detecting vehicle information and tarpaulin coverage. Thresholds are set to distinguish between three states: "complete coverage," "partial coverage," and "no coverage." The specific steps are as follows: The video stream captured by the camera is analyzed in real time using a deep learning model to extract the vehicle's feature information and the area covered by the tarpaulin. Based on the vehicle's feature information, the vehicle type, license plate number, and other information are identified. Based on the tarpaulin's coverage area, the coverage area and coverage rate of the tarpaulin are calculated. A threshold is set, and the tarpaulin's coverage rate is compared with the threshold to distinguish between three states: "complete coverage", "partial coverage", and "no coverage".

[0092] S320: The intelligent terminal performs secondary detection, which utilizes a photosensitive sensor to detect changes in light intensity in the area covered by the tarpaulin in real time. A threshold is set: when the tarpaulin is fully covered, the photosensitive sensor is blocked, and the light intensity detected by the sensor should be below the threshold; when the tarpaulin is not covered or only partially covered, part of the photosensitive sensor is blocked, and the light intensity detected by the remaining photosensitive sensor should be above the threshold. The specific steps are as follows: Install photosensitive sensors in the tarpaulin-covered area. The number and location of the photosensitive sensors are determined according to the actual situation. Real-time detection of changes in light intensity by the photosensitive sensors is performed and compared with a set threshold. When the tarpaulin is fully covered, the photosensitive sensors are blocked, and the light intensity detected by the photosensitive sensors should be lower than the threshold. When the tarpaulin is not covered or is partially covered, part of the photosensitive sensors are blocked, and the light intensity detected by the partial photosensitive sensors should be higher than the threshold.

[0093] S420: Platform data matching and result comparison (refer to S410 above).

[0094] S520: Uses GPS to locate vehicles and outputs the status of tarpaulin coverage.

[0095] (3) Example 3: At a construction site in a city, several dump trucks are transporting construction waste. These trucks are equipped with a deep learning-based tarpaulin cover detection system, which includes a smart terminal, camera, photosensor, and backend platform.

[0096] The smart terminal is linked to vehicle information. Vehicle information is entered into the smart terminal's internal database and a binding relationship is established through magnetic card input and verification. When a dump truck enters the construction site, a camera captures the vehicle's video stream, and model inference is used to analyze the video stream in real time, detecting vehicle information and tarpaulin coverage. Simultaneously, a photosensor detects changes in light intensity in the tarpaulin-covered area in real time. The smart terminal records the photosensor's detection results and transmits the data wirelessly to the backend server.

[0097] After receiving the video stream, the backend server uses a deep learning model to detect the license plate number in each frame. The deep learning model accurately identifies the license plate number by analyzing features such as texture and color, and associates it with vehicle information. Then, the backend platform matches the model's inference results with the smart terminal's detection results based on the license plate number, ensuring that both correspond to the same dump truck. The backend platform compares the tarpaulin coverage status inferred by the deep learning model with the smart terminal's detection results. If the comparison results match, the detection result is correct. In this case, three results will occur: complete coverage, partial coverage, and no coverage. If the detection result is partial coverage or no coverage, an alarm mechanism is triggered. If the comparison results do not match, the detection result is incorrect. The incorrect result is further analyzed using the video stream or images captured by the camera, utilizing light intensity changes detected by a photosensor, and comparing data from multiple modalities before outputting the result. If the detection result is still incorrect, an alarm mechanism is triggered. Finally, GPS is used to locate the vehicle and output the tarpaulin coverage status.

[0098] (4) Example 4: On a city road, several dump trucks are driving. These dump trucks are all equipped with a deep learning-based tarpaulin cover detection system, which includes a smart terminal, camera, photosensor, and backend platform.

[0099] The smart terminal is linked to vehicle information. Vehicle information is entered into the smart terminal's internal database and a binding relationship is established through magnetic card input and verification. When the dump truck is driving on the road, the camera captures the video stream and uses model inference to analyze the video stream in real time, detecting vehicle information and tarpaulin coverage. Simultaneously, a photosensor detects changes in light intensity in the tarpaulin-covered area in real time. The smart terminal records the photosensor's detection results and transmits the data wirelessly to the backend server.

[0100] After receiving the video stream, the backend server uses a deep learning model to detect the license plate number in each frame. The deep learning model accurately identifies the license plate number by analyzing features such as texture and color, and associates it with vehicle information. Then, the backend platform matches the model's inference results with the smart terminal's detection results based on the license plate number, ensuring that both correspond to the same dump truck. The backend platform compares the tarpaulin coverage status inferred by the deep learning model with the smart terminal's detection results. If the comparison results match, the detection result is correct. In this case, three results will occur: complete coverage, partial coverage, and no coverage. If the detection result is partial coverage or no coverage, an alarm mechanism is triggered. If the comparison results do not match, the detection result is incorrect. The incorrect result is further analyzed using the video stream or images captured by the camera, utilizing light intensity changes detected by a photosensor, and comparing data from multiple modalities before outputting the result. If the detection result is still incorrect, an alarm mechanism is triggered. Finally, GPS is used to locate the vehicle and output the tarpaulin coverage status.

[0101] (5) Example 5: At a city landfill, several dump trucks are unloading garbage. These trucks are equipped with a deep learning-based tarpaulin cover detection system, which includes a smart terminal, cameras, photosensors, and a backend platform.

[0102] The smart terminal is linked to vehicle information. Vehicle information is entered into the smart terminal's internal database and a binding relationship is established through magnetic card input and verification. When a dump truck enters the waste disposal site, a camera captures the vehicle's video stream, and model inference is used to analyze the video stream in real time, detecting vehicle information and tarpaulin coverage. Simultaneously, a photosensor detects changes in light intensity in the tarpaulin-covered area in real time. The smart terminal records the photosensor's detection results and transmits the data wirelessly to the backend server.

[0103] After receiving the video stream, the backend server uses a deep learning model to detect the license plate number in each frame. The deep learning model accurately identifies the license plate number by analyzing the image's texture, color, and other features, and associates it with vehicle information. Then, the backend platform matches the model's inference results with the smart terminal's detection results based on the license plate number, ensuring that both correspond to the same dump truck. The backend platform compares the tarpaulin coverage status inferred by the deep learning model with the smart terminal's detection results. If the comparison results match, the detection result is correct. In this case, three results will occur: complete coverage, partial coverage, and no coverage. If the detection result is partial coverage or no coverage, an alarm mechanism is triggered. If the comparison results do not match, the detection result is incorrect. The incorrect result is further analyzed using the video stream or image captured by the camera, utilizing light intensity changes detected by a photosensor, and comparing data from multiple modalities before outputting the result. If the detection result is still incorrect, an alarm mechanism is triggered. Finally, GPS is used to locate the vehicle and output the tarpaulin coverage status.

[0104] Figure 3 This is a structural diagram of a device for determining the coverage state of an obstruction according to an embodiment of this application, as shown below. Figure 3 As shown, the device includes: The acquisition module 302 is used to acquire video stream data of the target vehicle, wherein the target vehicle includes a vehicle covered by an obstruction, the obstruction being used to obstruct the materials carried by the target vehicle. The first detection module 304 is used to detect video stream data and obtain a first detection result, wherein the first detection result includes a coverage status determined based on visual image information of the video stream; The second detection module 306 is used to detect the photosensitive sensor data of the target vehicle and obtain a second detection result. The photosensitive sensor that collects the photosensitive sensor data is set in the occlusion area corresponding to the occlusion body. The second detection result includes the coverage state of the occlusion area determined based on light information. The determination module 308 is used to determine the coverage state of the occlusion body that corresponds to both the first detection result and the second detection result.

[0105] It should be noted that, Figure 3 The device for determining the coverage state of the obstruction shown is used to perform... Figure 2 The method for determining the coverage state of the obstruction shown is therefore... Figure 2 The relevant explanations in the method for determining the coverage state of obstructions also apply to... Figure 3 The device for determining the coverage state of the obstruction shown will not be described in detail here.

[0106] This application also provides an electronic device, which includes a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute the steps of the method for determining the coverage state of the obstruction in various embodiments of this application.

[0107] This application also provides a non-volatile storage medium including a stored computer program, wherein the device containing the non-volatile storage medium executes the steps of the method for determining the coverage state of the obstruction in various embodiments of this application by running the computer program.

[0108] This application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the method for determining the coverage state of an obstruction in various embodiments of this application.

[0109] This application also provides a computer program that, when executed by a processor, implements the steps of the method for determining the coverage state of the occlusion in various embodiments of this application.

[0110] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0111] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0112] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0113] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0114] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0115] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0116] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining the coverage state of an obstruction, characterized in that, include: Acquire video stream data of a target vehicle, wherein the target vehicle includes a vehicle covered by an obstruction, the obstruction being used to obstruct the materials carried by the target vehicle; The video stream data is detected to obtain a first detection result, wherein the first detection result includes a coverage status determined based on the visual image information of the video stream; The photosensitive sensor data of the target vehicle is detected to obtain a second detection result, wherein the photosensitive sensor that collects the photosensitive sensor data is set in the occlusion area corresponding to the occlusion body, and the second detection result includes the coverage state of the occlusion area determined based on light information; Determine the coverage status of the obstruction that corresponds to both the first detection result and the second detection result.

2. The method according to claim 1, characterized in that, The video stream data is inspected to obtain a first inspection result, including: A deep learning model is used to detect the video stream data to obtain the occlusion coverage information of the target vehicle, wherein the occlusion coverage information includes the coverage area of ​​the occlusion. Determine the coverage rate of the obstruction corresponding to the coverage area; The coverage rate of the occluder is compared with a first preset threshold to obtain a first comparison result; The first coverage state of the occluder is determined based on the first comparison result, and the first coverage state is used as the first detection result, wherein the first coverage state includes the coverage state identified based on visual image information.

3. The method according to claim 1, characterized in that, The photosensitive sensor data of the target vehicle is detected to obtain a second detection result, including: Receive the photosensitive sensor data collected by the photosensitive sensor; The light intensity of the blocked area is determined based on the data from the photosensitive sensor. The light intensity is compared with a second preset threshold to obtain a second comparison result; The second coverage state of the occluder is determined based on the second comparison result, and the second coverage state is used as the second detection result, wherein the second coverage state includes the coverage state identified based on light intensity information.

4. The method according to claim 3, characterized in that, The method further includes: The color of light in the blocked area is determined based on the data from the photosensitive sensor. The light color is compared with the color feature library to obtain a third comparison result. The color feature library includes the preset color performance of the occluder under different lighting conditions and the color features of the material transported by the target vehicle in the uncovered state. The second coverage state is determined based on the second comparison result and the third comparison result.

5. The method according to claim 2, characterized in that, Determining the coverage status of the obstruction that corresponds to both the first detection result and the second detection result includes: The first detection result is compared with the second detection result to obtain a fourth comparison result; If the fourth comparison result indicates that the first detection result is consistent with the second detection result, the first detection result or the second detection result shall be determined as the coverage state of the obstruction. If the fourth comparison result indicates that the first detection result is inconsistent with the second detection result, the video stream data and the photosensitive sensor data are fused to obtain fused data, and the coverage status of the occlusion is determined based on the fused data.

6. The method according to claim 5, characterized in that, Before comparing the first detection result with the second detection result to obtain the fourth comparison result, the method further includes: The deep learning model is used to detect the video stream data to obtain the vehicle information of the target vehicle, wherein the vehicle information includes the first license plate number of the target vehicle; Obtain the second license plate number corresponding to the vehicle where the photosensitive sensor is located; The first license plate number is compared with the second license plate number to obtain the fifth comparison result; If the fifth comparison result indicates that the first license plate number matches the second license plate number, the first detection result is compared with the second detection result.

7. The method according to claim 5, characterized in that, The method further includes: The type of the coverage state of the obstruction is determined, wherein the type includes a complete coverage state, a partial coverage state, and an uncovered state, wherein the coverage area of ​​the first obstruction corresponding to the complete coverage state is greater than the coverage area of ​​the second obstruction corresponding to the partial coverage state, and the coverage area of ​​the second obstruction is greater than the coverage area of ​​the third obstruction corresponding to the uncovered state. An alarm message is generated when the type does not belong to the complete coverage state.

8. A device for determining the coverage state of an obstruction, characterized in that, include: The acquisition module is used to acquire video stream data of a target vehicle, wherein the target vehicle includes a vehicle covered by an obstruction, the obstruction being used to obstruct the materials carried by the target vehicle. A first detection module is used to detect the video stream data and obtain a first detection result, wherein the first detection result includes a coverage state determined based on the visual image information of the video stream; The second detection module is used to detect the photosensitive sensor data of the target vehicle and obtain a second detection result. The photosensitive sensor that collects the photosensitive sensor data is set in the occlusion area corresponding to the occlusion body. The second detection result includes the coverage state of the occlusion area determined based on light information. The determination module is used to determine the coverage status of the occlusion that corresponds to both the first detection result and the second detection result.

9. An electronic device, characterized in that, include: A memory and a processor, the memory being used to store program instructions; the processor being connected to the memory and used to execute the method for determining the coverage state of an obstruction as described in any one of claims 1 to 7.

10. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program, wherein the device containing the non-volatile storage medium executes the method for determining the coverage state of an obstruction as described in any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the method for determining the coverage state of the obstruction as described in any one of claims 1 to 7.

Images