A target detection optimization method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of target detection, in particular to a target detection optimization method and device, electronic equipment and a storage medium, which comprises the following steps: S1, image acquisition: preparing an image acquisition module, the image acquisition module is used for image acquisition and analysis of images; S2, environment acquisition: preparing an environment acquisition module, the environment acquisition module is used for collecting radar images, environment temperature and environment humidity; S3, comprehensive analysis: preparing a comprehensive analysis module, the comprehensive analysis module is used for receiving and processing image information, thermal imaging and target cloud points collected by the image acquisition module and the environment acquisition module, and real-time monitoring of a vehicle driving state; S4, driving scheme analysis: using a deep learning module, according to historical driving records of the vehicle in different driving environments, a vehicle automatic driving model is constructed based on a convolution layer algorithm. The application is perfect in function, can effectively detect targets and assist vehicle driving.

Description

Technical Field

[0001] This invention relates to the field of target detection technology, and specifically to a target detection optimization method, apparatus, electronic device, and storage medium. Background Technology

[0002] Object detection optimization methods are primarily used in autonomous driving technology. Autonomous driving technology is a technology that deeply integrates the automotive industry with next-generation information technologies such as artificial intelligence, high-performance computing, big data, and the Internet of Things. It has significant advantages in reducing traffic congestion, minimizing traffic accidents, and improving fuel economy. Furthermore, the development of autonomous driving technology can help cities build safe and efficient future mobility infrastructure, which is of significant strategic importance for the cross-industry integration and development of the automotive sector.

[0003] In existing technologies, target detection optimization methods typically use visible light cameras for image acquisition, which offers advantages such as high resolution, low cost, and stable performance. Visible light cameras can capture rich semantic, color, and texture information from vehicle targets; however, these features are easily affected by adverse weather conditions, leading to a significant reduction in the accuracy of image acquisition under such conditions.

[0004] In summary, addressing the issue that existing target detection optimization methods are easily affected by adverse weather conditions, leading to a significant reduction in image acquisition accuracy under such conditions, has become a pressing problem in this field. Therefore, it is necessary to propose a target detection optimization method, apparatus, electronic device, and storage medium that can improve image acquisition accuracy under adverse weather conditions. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a target detection optimization method, apparatus, electronic device, and storage medium. Through the design of an image acquisition module and an environment acquisition module, the environment in which the vehicle is located can be effectively identified. Further analysis through a comprehensive analysis module improves the accuracy of target detection.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: a target detection optimization method, comprising the following steps:

[0007] S1, Image Acquisition: Prepare the image acquisition module, which is used to acquire and analyze images of the target to be detected; use the image acquisition module to acquire image information of the vehicle's surroundings, analyze the weather and road conditions based on the image information, and select pedestrians, vehicles, traffic facilities and obstacles in the image.

[0008] S2, Environmental Acquisition: Prepare the environmental acquisition module, which is used to acquire radar images, ambient temperature, and ambient humidity to form thermal images and target cloud points; use the environmental acquisition module to collect information about the environment around the vehicle; the environmental acquisition module collects the temperature of each detected target and displays the temperature image and temperature profile of the detected target in the form of thermal imaging; compare and analyze the temperature image and temperature profile of the detected target to determine the pedestrians, vehicles, traffic facilities, road conditions, and obstacles in the environment around the vehicle.

[0009] Simultaneously, the environmental acquisition module transmits electromagnetic wave signals to the vehicle's surrounding environment and receives echo signals. The echo signals are analyzed and processed to extract the distance, size, shape, direction, and speed of the detected targets. Pedestrians, vehicles, traffic facilities, road conditions, and obstacles in the vehicle's surrounding environment are identified and displayed in the form of target cloud points.

[0010] The environmental acquisition module collects ambient humidity data to assist the image acquisition module in determining the current weather conditions.

[0011] S3, Comprehensive Analysis: Prepare the comprehensive analysis module. The comprehensive analysis module is used to receive and process image information, thermal imaging, and target cloud points collected by the image acquisition module and the environment acquisition module, and monitor the vehicle's driving status in real time. Using the comprehensive analysis module to monitor the vehicle's driving status in real time, it combines image information and environmental information to comprehensively analyze pedestrians, vehicles, traffic facilities, road conditions, and obstacles around the vehicle, determine the current driving environment of the vehicle, and record the vehicle's driving environment and driving status each time to form a historical driving record.

[0012] S4, Driving Plan Analysis: Prepare a deep learning module, which is used to perform deep learning based on historical driving records to optimize driving plans; use the deep learning module to build an autonomous driving model of the vehicle based on the vehicle's historical driving records in different driving environments and convolutional layer algorithms; based on the autonomous driving model, combined with the current vehicle driving state and driving environment, simulate different driving plans; based on the driving conditions simulated by different driving plans, analyze the driving plans with low accident rates and compliance with road traffic regulations based on the vehicle driving state, the surrounding environment of the vehicle, and road traffic regulations.

[0013] Furthermore, in S1, the image acquisition module identifies and distinguishes pedestrians, vehicles, and traffic facilities based on their overall shape and outline features.

[0014] Furthermore, in S1, road conditions include road type, road material, road markings, and road surface unevenness.

[0015] Furthermore, in S1, the road surface types include straight sections, turning sections, uphill sections, and downhill sections.

[0016] Furthermore, in S1, the driving status includes the vehicle's speed, turning angle, and lighting status.

[0017] Furthermore, in S1, obstacles include objects that appear on the road surface other than pedestrians, vehicles, traffic facilities, and the road surface itself.

[0018] Furthermore, in S2, the ambient temperature includes the ground temperature and the interior temperature of the vehicle.

[0019] The technical principles of the above solution are as follows:

[0020] First, the image acquisition module undergoes extensive image recognition training. After training, the image acquisition module collects image information around the vehicle. During the image acquisition process, the image acquisition module selects and displays pedestrians, vehicles, traffic facilities, and obstacles in the image using different colored boxes, thereby reminding the driver of the pedestrians, vehicles, traffic facilities, and obstacles around the vehicle that require attention.

[0021] Simultaneously with image acquisition, the environmental acquisition module collects the temperature information of each detected target. Since the temperatures of pedestrians, vehicles, traffic facilities, road surfaces, and obstacles are all different—the temperature of a vehicle engine is much higher than the body temperature of a pedestrian, while the temperature of the road surface and obstacles is generally lower in normal weather—the environmental acquisition module uses thermal imaging to display the temperature image and temperature profile of the detected target. The lower the temperature, the bluer the image color; the higher the temperature, the redder the image color. As the temperature rises, the image changes from dark blue to dark red. The environmental acquisition module then compares and analyzes the temperature image and temperature profile of the detected target with the temperature images and temperature profiles of pedestrians, vehicles, traffic facilities, road surfaces, and obstacles under normal conditions, thereby effectively determining the presence of pedestrians, vehicles, traffic facilities, road surface conditions, and obstacles in the vehicle's surrounding environment.

[0022] Because different objects have different distances, speeds, shapes, and sizes, their reception and reflection of waves will also differ. Furthermore, since the propagation speed of electromagnetic waves is known (approximately the speed of light in a vacuum), radar can calculate the distance between the target and the radar by measuring the time delay between the transmitted signal and the received echo. Therefore, the environmental acquisition module also transmits electromagnetic wave signals from the vehicle's surrounding environment and receives echo signals. It analyzes and processes the echo signals to extract the distance, size, shape, direction, and speed of the detected target, identifying pedestrians, vehicles, traffic facilities, road conditions, and obstacles in the vehicle's surrounding environment. The environmental acquisition module uses multiple antennas to scan simultaneously, determining the azimuth and elevation angles of the detected target relative to the radar, thus determining the target's position in three-dimensional space and displaying the detected target as a target cloud, allowing the driver to more intuitively understand the vehicle's surrounding environment.

[0023] The environmental acquisition module collects ambient humidity data and transmits it to the image acquisition module to help the image acquisition module determine the current weather conditions.

[0024] After image and environmental data acquisition is completed, the comprehensive analysis module extracts the data from both, prioritizing the current weather conditions and selecting the primary criteria for judgment. In hot weather, the temperature difference between pedestrians, road surfaces, and obstacles may decrease during environmental data acquisition, resulting in similar colors and reduced clarity of temperature contours in thermal imaging. In this case, the comprehensive analysis module uses the image information acquired by the image acquisition module and the target cloud points acquired by radar as the primary criteria to determine the condition of pedestrians, vehicles, traffic facilities, road surfaces, and obstacles around the vehicle, further supplemented by thermal imaging. In rainy weather, image acquisition is prone to blurring, and radar is also affected. In this situation, the vehicle engine temperature is still much higher than the pedestrian's body temperature, while the temperatures of traffic facilities, road surfaces, and obstacles are lower. The comprehensive analysis module uses the temperature images and contours acquired by thermal imaging as the primary criteria, supplemented by radar and image information. This allows for comprehensive detection of the vehicle's surrounding environment under various weather conditions.

[0025] The deep learning module constructs an autonomous driving model for the vehicle based on the vehicle's historical driving records in different driving environments and using convolutional layer algorithms. The autonomous driving model combines the current vehicle driving status and driving environment to simulate different driving schemes. Based on the driving conditions simulated by different driving schemes, and considering the vehicle driving status, the surrounding environment, and road traffic regulations, it analyzes driving schemes with low accident rates and compliance with road traffic regulations.

[0026] The above approach has the following beneficial effects:

[0027] 1. This invention utilizes an image acquisition module for detailed image recognition of the vehicle's surrounding environment, particularly the real-time selection and display of pedestrians, vehicles, traffic facilities, and obstacles. This significantly improves the driver's perception of their surroundings and reduces the risk of accidents caused by blind spots or negligence. Simultaneously, the environmental acquisition module employs thermal imaging technology to effectively identify potential hazards, such as puddles and obstacles on the road, even under complex or severe weather conditions, further enhancing driving safety.

[0028] 2. The integrated analysis module of this invention is designed to automatically adjust its judgment criteria according to different weather conditions, ensuring effective detection of the vehicle's surrounding environment under various adverse weather conditions. For example, in hot weather, when thermal imaging technology is limited by reduced temperature differences, the integrated analysis module will rely more on image information and radar data for judgment; while in rainy weather, when image and radar data are interfered with, the integrated analysis module will rely more on thermal imaging data. This flexible switching mechanism greatly enhances the system's environmental adaptability.

[0029] 3. The deep learning module of this invention analyzes the vehicle's historical driving records in different driving environments to construct an autonomous driving model. The deep learning module can simulate different driving schemes, evaluate their accident risks and legality, and ultimately select the optimal driving scheme. This decision-making approach based on big data and machine learning enables vehicles to more intelligently cope with complex traffic environments, improving driving safety and efficiency.

[0030] Furthermore, a target detection device includes a mounting bar, on the side wall of which a temperature sensor, a humidity sensor, an image acquisition device, and a radar are fixedly connected.

[0031] Beneficial effects: The mounting bar design allows the temperature sensor, humidity sensor, image acquisition unit, and radar to be integrated into a single unit, enabling better installation of the target detection device on the vehicle and thus providing better target detection services for the vehicle.

[0032] Furthermore, an electronic device for target detection includes a controller for receiving information acquired by a temperature sensor, a humidity sensor, an image acquisition device, and radar.

[0033] Beneficial effects: During vehicle operation, the controller can assist the vehicle in adjusting driving operations such as speed, steering, and braking based on information such as ambient temperature, humidity, road conditions, and the driving status of vehicles ahead, thereby ensuring driving safety.

[0034] Furthermore, a target detection storage medium includes a storage device for storing image information, environmental information, and historical driving records of the vehicle in different driving environments.

[0035] Beneficial effects: The storage device can store image information, environmental information, and historical driving records of the vehicle in different driving environments, which facilitates the comprehensive analysis module to extract the required information for comprehensive analysis, and also facilitates the deep learning module to perform deep learning.

[0036] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating a target detection optimization method according to an embodiment of the present invention.

[0038] Figure 2 This is a schematic diagram illustrating the steps of a target detection optimization method in an embodiment of the present invention.

[0039] Figure 3 This is an isometric view of a target detection device according to an embodiment of the present invention.

[0040] The reference numerals in the accompanying drawings include: 1. Mounting rod; 2. Temperature sensor; 3. Humidity sensor; 4. Image acquisition device; 5. Radar. Detailed Implementation

[0041] The following detailed description illustrates the specific implementation method:

[0042] Example 1:

[0043] As attached Figures 1-3 As shown, an object detection optimization method includes the following steps:

[0044] S1, Image Acquisition: Prepare the image acquisition module, which is used to acquire and analyze images of the target to be detected; use the image acquisition module to acquire image information of the vehicle's surroundings, analyze the weather and road conditions based on the image information, and select pedestrians, vehicles, traffic facilities, and obstacles in the image with different colored boxes; the image acquisition module identifies and distinguishes pedestrians, vehicles, and traffic facilities based on their overall shape and contour features.

[0045] Road surface conditions include road type, road material, road markings, and road surface unevenness. Road type includes straight sections, curved sections, uphill sections, and downhill sections. Driving conditions include vehicle speed, turning angle, and lighting status. Obstacles include objects on the road surface other than pedestrians, vehicles, traffic facilities, and the road surface itself.

[0046] S2, Environmental Acquisition: Prepare the environmental acquisition module, which is used to acquire radar images, ambient temperature, and ambient humidity to form thermal images and target cloud points; use the environmental acquisition module to collect information about the environment around the vehicle; the environmental acquisition module collects the temperature of each detected target and displays the temperature image and temperature profile of the detected target in the form of thermal imaging; compare and analyze the temperature image and temperature profile of the detected target to determine the pedestrians, vehicles, traffic facilities, road conditions, and obstacles in the environment around the vehicle;

[0047] Simultaneously, the environmental acquisition module transmits electromagnetic wave signals to the vehicle's surrounding environment and receives echo signals. The echo signals are analyzed and processed to extract the distance, size, shape, direction, and speed of the detected targets. Pedestrians, vehicles, traffic facilities, road conditions, and obstacles in the vehicle's surrounding environment are identified and displayed in the form of target cloud points.

[0048] The environmental acquisition module collects ambient humidity data to assist the image acquisition module in determining the current weather conditions.

[0049] The ambient temperature includes the ground temperature and the vehicle interior temperature. The environmental data acquisition module also collects the vehicle interior temperature and adjusts it to the temperature value set by the driver.

[0050] S3, Comprehensive Analysis: Prepare the comprehensive analysis module. The comprehensive analysis module is used to receive and process image information, thermal imaging, and target cloud points collected by the image acquisition module and the environment acquisition module, and monitor the vehicle's driving status in real time. Using the comprehensive analysis module to monitor the vehicle's driving status in real time, it combines image information and environmental information to comprehensively analyze pedestrians, vehicles, traffic facilities, road conditions, and obstacles around the vehicle, determine the current driving environment of the vehicle, and record the vehicle's driving environment and driving status each time to form a historical driving record.

[0051] S4, Driving Plan Analysis: Prepare a deep learning module, which is used to perform deep learning based on historical driving records to optimize driving plans; use the deep learning module to build an autonomous driving model of the vehicle based on the vehicle's historical driving records in different driving environments and convolutional layer algorithms; based on the autonomous driving model, combined with the current vehicle driving state and driving environment, simulate different driving plans; based on the driving conditions simulated by different driving plans, analyze the driving plans with low accident rates and compliance with road traffic regulations based on the vehicle driving state, the surrounding environment of the vehicle, and road traffic regulations.

[0052] The specific implementation process is as follows: First, the image acquisition module undergoes extensive image recognition training. After training, the image acquisition module is used to collect image information around the vehicle. During the image acquisition process, the image acquisition module will select and display pedestrians, vehicles, traffic facilities, and obstacles in the image using different colored boxes, thereby reminding the driver of the pedestrians, vehicles, traffic facilities, and obstacles around the vehicle that require attention.

[0053] For example, if a pedestrian is crossing a zebra crossing in front of the driver's vehicle, the traffic light is red with 20 seconds remaining, and a white sedan is stationary in the second lane on the opposite side of the road; after the image recognition module detects the image, it will select the pedestrian with a red frame and move it in real time with the pedestrian's movement; the image recognition module will select the oncoming vehicle with a blue frame and move it in real time with the vehicle's movement; the image recognition module will select the traffic light with a green frame and highlight the remaining 20 seconds of the red light; at the same time, the image recognition module will also mark the lane outline, zebra crossing outline, and roadside outline with solid white lines.

[0054] Simultaneously with image acquisition, the environmental acquisition module collects the temperature information of each detected target. Since the temperatures of pedestrians, vehicles, traffic facilities, road surfaces, and obstacles are all different—the temperature of a vehicle engine is much higher than the body temperature of a pedestrian, while the temperature of the road surface and obstacles is generally lower in normal weather—the environmental acquisition module uses thermal imaging to display the temperature image and temperature profile of the detected target. The lower the temperature, the bluer the image color; the higher the temperature, the redder the image color. As the temperature rises, the image changes from dark blue to dark red. The environmental acquisition module then compares and analyzes the temperature image and temperature profile of the detected target with the temperature images and temperature profiles of pedestrians, vehicles, traffic facilities, road surfaces, and obstacles under normal conditions, thereby effectively determining the presence of pedestrians, vehicles, traffic facilities, road surface conditions, and obstacles in the vehicle's surrounding environment.

[0055] For example, there is a pedestrian crossing a zebra crossing in front of the driver's vehicle, a right-turn sign is painted on the road surface in the first lane, a car is stationary in the second lane, and there is a puddle 10 meters directly in front of the vehicle; because the zebra crossing and traffic signs are made of different materials than the road surface, and the puddle, manhole cover, and drain are different from the road surface in terms of structure and location, their temperatures are also different. Therefore, thermal imaging will clearly show the pedestrian, vehicle, zebra crossing, right-turn sign, and puddle on the road surface with different temperature images and temperature contours.

[0056] Because different objects have different distances, speeds, shapes, and sizes, their reception and reflection of waves will also differ. Furthermore, since the propagation speed of electromagnetic waves is known (approximately the speed of light in a vacuum), radar can calculate the distance between the target and the radar by measuring the time delay between the transmitted signal and the received echo. Specifically, distance (R) equals the time delay (τ) multiplied by half the electromagnetic wave propagation speed (c), i.e., R = 2cτ. Therefore, the environmental acquisition module also transmits electromagnetic wave signals to the vehicle's surrounding environment and receives echo signals. It analyzes and processes the echo signals to extract the distance, size, shape, direction, and speed of the detected target, determining the presence of pedestrians, vehicles, traffic facilities, road conditions, and obstacles in the vehicle's surrounding environment. The environmental acquisition module uses multiple antennas to scan together, determining the azimuth and elevation angles of the detected target relative to the radar, thereby determining the target's position in three-dimensional space and displaying the detected target as a target cloud, allowing the driver to more intuitively understand the vehicle's surrounding environment.

[0057] The environmental acquisition module collects ambient humidity data and transmits it to the image acquisition module to help the image acquisition module determine the current weather conditions.

[0058] After image and environmental data acquisition is completed, the comprehensive analysis module extracts the data from both, prioritizing the current weather conditions and selecting the primary criteria for judgment. In hot weather, the temperature difference between pedestrians, road surfaces, and obstacles may decrease during environmental data acquisition, resulting in similar colors and reduced clarity of temperature contours in thermal imaging. In this case, the comprehensive analysis module uses the image information acquired by the image acquisition module and the target cloud points acquired by radar as the primary criteria to determine the condition of pedestrians, vehicles, traffic facilities, road surfaces, and obstacles around the vehicle, further supplemented by thermal imaging. In rainy weather, image acquisition is prone to blurring, and radar is also affected. In this situation, the vehicle engine temperature is still much higher than the pedestrian's body temperature, while the temperatures of traffic facilities, road surfaces, and obstacles are lower. The comprehensive analysis module uses the temperature images and contours acquired by thermal imaging as the primary criteria, supplemented by radar and image information. This allows for comprehensive detection of the vehicle's surrounding environment under various weather conditions.

[0059] The deep learning module constructs an autonomous driving model for the vehicle based on its historical driving records in different driving environments and using convolutional layer algorithms. The autonomous driving model includes multiple convolutional layers, pooling layers, and fully connected layers. Convolutional layers are used to extract features from the driving environment; pooling layers are used to reduce feature dimensionality and computational cost; and fully connected layers are used to map the extracted features to simulate different driving behaviors. Based on the simulated driving conditions of different driving behaviors, the module analyzes the driving situation based on the vehicle's driving status, surrounding environment, and traffic regulations to determine driving strategies with low accident rates and compliance with traffic regulations.

[0060] This invention utilizes an image acquisition module for detailed image recognition of the vehicle's surrounding environment, particularly the real-time selection and display of pedestrians, vehicles, traffic facilities, and obstacles. This significantly improves the driver's perception of their surroundings and reduces the risk of accidents caused by blind spots or negligence. Simultaneously, the environmental acquisition module employs thermal imaging technology to effectively identify potential hazards, such as puddles and obstacles on the road, even under complex or severe weather conditions, further enhancing driving safety.

[0061] The integrated analysis module can automatically adjust its judgment criteria according to different weather conditions, ensuring effective detection of the vehicle's surrounding environment under various conditions. For example, in hot weather, when thermal imaging technology is limited by reduced temperature differences, the integrated analysis module will rely more on image information and radar data for judgment; while in rainy weather, when image and radar data are interfered with, the integrated analysis module will rely more on thermal imaging data. This flexible switching mechanism greatly enhances the system's environmental adaptability.

[0062] The deep learning module analyzes the vehicle's historical driving records in different driving environments to build an autonomous driving model. This module can simulate different driving strategies, assess their accident risks and legality, and ultimately select the optimal driving strategy. This decision-making approach based on big data and machine learning enables vehicles to more intelligently cope with complex traffic environments, improving driving safety and efficiency.

[0063] Example 2:

[0064] As attached Figure 1 , Figure 3 As shown, the difference from the above embodiment is that a target detection device includes a mounting bar 1, and a temperature sensor 2, a humidity sensor 3, an image acquisition device 4 and a radar 5 are fixedly connected to the side wall of the mounting bar 1 by bolts.

[0065] The specific implementation process is as follows: The design of mounting bar 1 can integrate temperature sensor 2, humidity sensor 3, image acquisition device 4 and radar 5 into a whole, so as to better install the target detection device on the vehicle and thus provide better target detection services for the vehicle.

[0066] Example 3:

[0067] As attached Figure 1 , Figure 3 As shown, the difference from the above embodiments is that an electronic device for target detection includes a controller for receiving information acquired by a temperature sensor, a humidity sensor, an image acquisition device, and radar.

[0068] The specific implementation process is as follows: During vehicle operation, the controller can assist the vehicle in adjusting driving operations such as speed, steering, and braking based on information such as ambient temperature, humidity, road conditions, and the driving status of vehicles ahead, to ensure driving safety.

[0069] Example 4:

[0070] As attached Figure 1 , Figure 3 As shown, the difference from the above embodiments is that a target detection storage medium includes a storage device for storing image information, environmental information, and historical driving records of the vehicle in different driving environments.

[0071] The specific implementation process is as follows: During the vehicle's operation, the storage device can store image information, environmental information, and historical driving records of the vehicle in different driving environments, which facilitates the comprehensive analysis module to extract the required information for comprehensive analysis, and also facilitates the deep learning module to perform deep learning.

[0072] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. An optimized method for target detection, characterized in that, Includes the following steps: S1, Image Acquisition: Prepare the image acquisition module, which is used to acquire and analyze images of the target to be detected; use the image acquisition module to acquire image information of the vehicle's surroundings, analyze the weather and road conditions based on the image information, and select pedestrians, vehicles, traffic facilities and obstacles in the image with different colored boxes; S2, Environmental Acquisition: Prepare the environmental acquisition module, which is used to acquire radar images, ambient temperature, and ambient humidity to form thermal images and target cloud points; use the environmental acquisition module to collect information about the environment around the vehicle; The environmental acquisition module collects the temperature information of each detection target and displays the temperature image and temperature profile of the detection target in the form of thermal imaging. Based on the comparison and analysis of the temperature image and temperature profile of the detection target, the system can determine the pedestrians, vehicles, traffic facilities, road conditions and obstacles in the environment around the vehicle. Simultaneously, the environmental acquisition module transmits electromagnetic wave signals to the vehicle's surrounding environment and receives echo signals. The echo signals are analyzed and processed to extract the distance, size, shape, direction, and speed of the detected targets. Pedestrians, vehicles, traffic facilities, road conditions, and obstacles in the vehicle's surrounding environment are identified and displayed in the form of target cloud points. The environmental acquisition module collects ambient humidity data to assist the image acquisition module in determining the current weather conditions. S3, Comprehensive Analysis: Prepare the comprehensive analysis module. The comprehensive analysis module is used to receive and process image information, thermal imaging and target cloud points collected by the image acquisition module and the environment acquisition module, and monitor the vehicle driving status in real time. The vehicle's driving status is monitored in real time using a comprehensive analysis module. At the same time, image information and environmental information are combined to comprehensively analyze pedestrians, vehicles, traffic facilities, road conditions and obstacles around the vehicle to determine the current driving environment of the vehicle. The driving environment and driving status of the vehicle each time are recorded to form a historical driving record. S4, Driving Plan Analysis: Prepare a deep learning module, which is used to perform deep learning based on historical driving records to optimize driving plans; Using a deep learning module, an autonomous driving model is constructed based on the vehicle's historical driving records in different driving environments and a convolutional layer algorithm. Based on the autonomous driving model, combined with the current vehicle driving state and driving environment, different driving schemes are simulated. Based on the driving conditions simulated by different driving schemes, and considering the vehicle driving state, the surrounding environment, and road traffic regulations, a driving scheme with a low accident rate and compliance with road traffic regulations is analyzed.

2. The target detection optimization method according to claim 1, characterized in that, In S1, the image acquisition module identifies and distinguishes pedestrians, vehicles, and traffic facilities based on their overall shape and outline features.

3. The target detection optimization method according to claim 1, characterized in that, In S1, road conditions include road type, road material, road markings, and road surface unevenness.

4. The target detection optimization method according to claim 1, characterized in that, In S1, the road surface types include straight sections, turning sections, uphill sections, and downhill sections.

5. The target detection optimization method according to claim 1, characterized in that, In S1, the driving status includes the vehicle's speed, turning angle, and lighting status.

6. The target detection optimization method according to claim 1, characterized in that, In S1, obstacles include objects on the road surface other than pedestrians, vehicles, traffic facilities, and the road surface itself.

7. The target detection optimization method according to claim 6, characterized in that, In S2, ambient temperature includes ground temperature and vehicle interior temperature.

8. A target detection apparatus based on the target detection optimization method according to any one of claims 1-7, characterized in that, It includes a mounting bar (1), and a temperature sensor (2), a humidity sensor (3), an image acquisition device (4), and a radar (5) are fixedly connected to the side wall of the mounting bar (1).

9. An electronic device for target detection based on the target detection optimization method according to any one of claims 1-7, characterized in that, It includes a controller, which receives information from temperature sensors, humidity sensors, image acquisition devices, and radar.

10. A target detection storage medium based on the target detection optimization method according to any one of claims 1-7, characterized in that, It includes a storage device for storing image information, environmental information, and the vehicle's historical driving records in different driving environments.

Images