Agricultural machinery harvesting operation safety monitoring system and method

Abstract

The invention discloses an agricultural machine harvesting operation safety monitoring system and method, and relates to the field of safety monitoring, and the system comprises a topographic measurement module which is used for measuring the topographic information of an agricultural machine operation area in real time; the target detection module is used for collecting an image of an agricultural machine operation area in real time, performing target identification based on the image, and determining target information in the agricultural machine operation area; the boundary generation module is used for generating electronic fence boundary information of the agricultural machinery operation area based on a satellite map and a geographic information system; the position acquisition module is used for acquiring the position information of the agricultural machinery in real time; and the safety monitoring module is used for monitoring the working environment of the agricultural machine in real time according to the terrain information, the target information, the electronic fence boundary information and the position information. The operation environment of the agricultural machine can be accurately monitored in real time, and then the harvesting operation safety and reliability of the agricultural machine in the high-plant complex environment are improved.

Description

Technical Field

[0001] The present application relates to the field of safety monitoring, and in particular to a system and method for monitoring the safety of agricultural machinery harvesting operations. Background Art

[0002] With the rapid development of agricultural mechanization and intelligence, harvesting machinery is increasingly used in modern agriculture. However, safety issues have gradually become an important research direction in the operation of harvesting machinery. Especially in the harvesting environment of high plants (such as corn), the complex operating conditions put forward higher requirements on the driver's operating ability and the environmental perception ability of the equipment. The high plant environment often limits the driver's observation ability because the height of the plant exceeds the line of sight of the cockpit. At the same time, interference factors such as dust and noise further reduce the perception efficiency and reaction ability. In such an operation scenario, harvesting machinery is prone to collision with obstacles, accidental injury to personnel in the operation area, or mistakenly entering inaccessible areas due to the inability to accurately perceive the environment. In addition, related technologies use radar and machine vision to identify the boundary of the ridge in front of the agricultural machinery in the direction of travel, and measure the distance from the agricultural machinery to the boundary of the ridge. However, in actual unstructured farmland, due to the complexity and diversity of types and the lack of refined management of the operation area, the boundary of the ridge is blurred. The method of using vehicle-mounted sensors such as lidar to detect and obtain the inaccessible area of ​​the harvesting machinery in real time is difficult to ensure real-time performance. Summary of the invention

[0003] The purpose of this application is to provide a safety monitoring system and method for agricultural machinery harvesting operations, which can accurately monitor the operating environment of agricultural machinery in real time, thereby improving the safety and reliability of agricultural machinery harvesting operations in a complex environment with high plants.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides an agricultural machinery harvesting operation safety monitoring system, comprising:

[0006] Topographic measurement module, used to measure the terrain information of the agricultural machinery operation area in real time;

[0007] The target detection module is used to collect images of the agricultural machinery operation area in real time, and perform target recognition based on the images to determine the target information in the agricultural machinery operation area;

[0008] The boundary generation module is used to generate the electronic fence boundary information of the agricultural machinery operation area based on satellite maps and geographic information systems;

[0009] Position collection module, used to collect the position information of agricultural machinery in real time;

[0010] The safety monitoring module is respectively connected to the terrain measurement module, the target detection module, the boundary generation module and the position acquisition module, and is used to monitor the working environment of the agricultural machinery in real time according to the terrain information, the target information, the electronic fence boundary information and the position information.

[0011] In a second aspect, the present application provides a method for monitoring the safety of agricultural machinery harvesting operations, comprising:

[0012] Real-time measurement of terrain information in agricultural machinery operation areas;

[0013] Collect location information and images of agricultural machinery operation areas in real time;

[0014] Performing target recognition based on the image to determine target information within the agricultural machinery operation area;

[0015] Generate electronic fence boundary information of agricultural machinery operation areas based on satellite maps and geographic information systems;

[0016] The operating environment of the agricultural machinery is monitored in real time according to the terrain information, the target information, the electronic fence boundary information and the location information.

[0017] According to the specific embodiments provided by this application, this application has the following technical effects: This application provides a system and method for monitoring the safety of agricultural machinery harvesting operations. By measuring the terrain information of the agricultural machinery operation area in real time, possible operation obstacles can be predicted. It is particularly suitable for safety hazards caused by ground unevenness changes, steep slopes of terraces, cliffs, etc. in the harvesting of high-plant crops. Combined with satellite maps and electronic fence boundary information of agricultural machinery operation areas generated by geographic information systems, it overcomes the problem of insufficient perception of dangerous factors such as the boundaries of the operation plots, negative obstacles, and obscured obstacles in the scene of harvesting high-plant crops. The recognition and detection of obstacles and personnel targets are realized through image processing, thereby improving the safety and reliability of agricultural machinery harvesting operations in complex environments with high plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A schematic diagram of functional modules of a safety monitoring system for harvesting agricultural machinery provided in one embodiment of the present application;

[0020] Figure 2A detailed module diagram of a safety monitoring system for harvesting agricultural machinery provided in one embodiment of the present application;

[0021] Figure 3 Schematic diagram of a target detection module in one embodiment of the present application;

[0022] Figure 4 A logical diagram of a safety determination mechanism in an embodiment of the present application;

[0023] Figure 5 A flowchart of a method for monitoring the safety of harvesting operations of agricultural machinery provided in one embodiment of the present application;

[0024] Figure 6 This is a flowchart of target recognition in one embodiment of the present application.

[0025] Figure numerals: 101-topographic measurement module, 102-target detection module, 103-boundary generation module, 104-position acquisition module, 105-safety monitoring module, 201-microwave ranging sensor, 202-electromagnetic wave ranging sensor, 203-multi-mode imaging submodule, 204-vehicle positioning device, 205-communication module, 206-main control unit, 207-safety decision unit, 208-execution unit, 209-operation terminal, 301-slide rail, 302-slide rail control submodule. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0028] In an exemplary embodiment, Figure 1 As shown, a safety monitoring system for agricultural machinery harvesting operation is provided, including: a terrain measurement module 101, a target detection module 102, a boundary generation module 103, a position acquisition module 104 and a safety monitoring module 105. The position and function of each module are introduced below.

[0029] (I) The terrain measurement module 101 is used to measure the terrain information of the agricultural machinery operation area in real time. The terrain information includes the ground change trend and the side terrain.

[0030] In an exemplary embodiment, Figure 2As shown, the terrain measurement module 101 includes: a microwave ranging sensor 201 and an electromagnetic wave ranging sensor 202. There are multiple microwave ranging sensors 201 and multiple electromagnetic wave ranging sensors 202.

[0031] The microwave ranging sensor 201 is arranged at the bottom of the crop divider of the agricultural machinery harvesting platform. The microwave ranging sensor 201 is used to measure the distance between the bottom of the agricultural machinery and the ground in real time to determine the trend of terrain changes.

[0032] The height of the harvesting platform from the ground is generally less than 40 cm. The present application uses a high-precision, low-power, short-range microwave ranging sensor 201 to measure the terrain changes at the bottom of the harvesting platform divider, thereby realizing rapid measurement of the terrain undulations in the direction of travel of the agricultural machinery, forming a trend map of ground changes ahead, and predicting possible operational obstacles through terrain changes, thereby adjusting the impact of ground unevenness changes on harvesting safety during the harvesting of high-plant crops.

[0033] The emission angle of the microwave ranging sensor 201 is perpendicular to the ground and slightly tilted forward to ensure that the terrain below the harvesting platform can be detected from a lower position during the harvesting operation. The microwave ranging sensor 201 transmits microwave signals at a high frequency to measure the ground height difference in front of the agricultural machinery and convert the echo signal into distance data. The distance data is output in the form of a time series. After pre-processing such as filtering and denoising the distance data, the relationship between the terrain height and the position of the agricultural machinery is established in combination with the speed of the agricultural machinery at the time of distance data collection. The sliding window algorithm is used to extract the terrain change trend from the data within the nearest distance, and the ground height of the specified target point in the direction of the agricultural machinery is predicted by fitting the trend to achieve terrain prediction.

[0034] The electromagnetic wave ranging sensor 202 is arranged on the side of the agricultural machine, and the electromagnetic wave ranging sensor 202 is used to measure the side terrain of the agricultural machine in real time.

[0035] By setting an electromagnetic wave ranging sensor 202 on the side of the agricultural machinery, dangerous terrain such as terraced steep slopes and cliffs can be scanned and identified. It is preferred to use a medium- and long-range electromagnetic wave ranging sensor 202 with a longer wavelength and stronger penetration ability, with a ranging range of less than 150m, to overcome the obstruction problem of laser radar and optical imaging systems caused by dust. Among them, the electromagnetic wave ranging sensor 202 is also called a radio frequency ranging sensor or a radio wave sensor, which calculates the distance based on the time difference from signal transmission to reception.

[0036] (ii) The target detection module 102 is used to collect images of the agricultural machinery operation area in real time, and perform target recognition based on the images to determine target information within the agricultural machinery operation area.

[0037] The present application utilizes multi-mode and multi-spectral scanning reconstruction imaging to achieve de-obstruction (crop obstruction effect) and reconstruction imaging of personnel and obstacle targets within a specific distance range ahead, thereby realizing the detection of obstacles or personnel targets within the working area in the direction of travel.

[0038] In an exemplary embodiment, Figure 3 As shown, the target detection module 102 includes: a multi-mode imaging submodule 203, a slide rail 301, a slide rail control submodule 302 and an image processing submodule (not shown in the figure).

[0039] The slide rail 301 is arranged at the front end of the agricultural machine. The slide rail 301 is 1 meter long and can realize the periodic left and right movement of the multi-mode imaging submodule 203 to cover a larger working area.

[0040] The multi-mode imaging submodule 203 is arranged on the slide rail 301, and the multi-mode imaging submodule 203 is used to collect images of the agricultural machinery operation area in real time. The images include visible light images, infrared images and polarization images. Specifically, the multi-mode imaging submodule 203 is installed in a safe area at the front end of the agricultural machinery, such as the front part above the cockpit. Its position needs to ensure a wide field of view, cover the front operation area of ​​the agricultural machinery and be unobstructed, and be equipped with a dust cover and a wiper assembly to adapt to the complex farmland harvesting operation environment.

[0041] In an exemplary embodiment, the multi-mode imaging submodule 203 includes: a visible light camera, an infrared camera, and a polarized light camera.

[0042] Visible light cameras are used to collect visible light images of agricultural machinery operation areas in real time.

[0043] Infrared cameras are used to collect infrared images of agricultural machinery operation areas in real time.

[0044] Polarized light cameras are used to collect polarized images of agricultural machinery operation areas in real time.

[0045] The slide rail control submodule 302 is connected to the multi-mode imaging submodule 203 , and is used to drive the multi-mode imaging submodule 203 to move on the slide rail 301 , so that the multi-mode imaging submodule 203 can capture images at different viewing angles.

[0046] Furthermore, the target detection module 102 further includes a trigger controller, which simultaneously triggers the visible light camera, the infrared camera and the polarized light camera to take photos at a designated position of the slide rail 301 through a trigger signal line.

[0047] The image processing submodule is connected to the multi-mode imaging submodule 203, and the image processing submodule is used to perform target recognition on the image and determine the target information in the agricultural machinery operation area.

[0048] In an exemplary embodiment, the image processing submodule includes: an image preprocessing unit, an image fusion unit, an image reconstruction unit and a target detection unit.

[0049] The image preprocessing unit is connected to the visible light camera, the infrared camera and the polarized light camera respectively, and is used to perform distortion correction, noise suppression and image enhancement processing on the visible light image, the infrared image and the polarized image in sequence to obtain a preprocessed visible light image, a preprocessed infrared image and a preprocessed polarized image.

[0050] Firstly, the three image data are distorted according to the camera's intrinsic parameters to eliminate the image deviation caused by lens distortion. Then, Gaussian filtering and non-uniformity correction are used to suppress noise. Finally, histogram equalization and edge enhancement are used to improve image contrast and clarity, providing high-quality input for subsequent processing.

[0051] The image fusion unit is connected to the image preprocessing unit, and is used to fuse the preprocessed visible light image, the preprocessed infrared image and the preprocessed polarization image to obtain a multimodal fusion image.

[0052] First, the projection matrix is ​​calculated based on the internal and external parameter information of the camera calibration and the camera position information of different viewing angles. Then, the images of different viewing angles and different modalities are geometrically transformed to ensure that all images are aligned to a unified coordinate system. Then, the weighted average method is used to fuse the images of different modalities to obtain a multimodal fusion image that can reflect more information. Among them, different modalities include visible light, infrared and polarized light. The fusion process is: the corresponding pixel values ​​of the images of different modalities are weighted averaged according to certain weights. The weight is determined according to the image quality. When the resolution is inconsistent, the interpolation method is used to ensure the uniform resolution size of the image.

[0053] The image reconstruction unit is connected to the image fusion unit, and is used to reconstruct the blocked area in the multimodal fusion image to obtain a reconstructed fusion image.

[0054] In combination with the multi-perspective images collected during the periodic scanning process of the multi-mode imaging submodule 203, an image reconstruction algorithm is used to achieve image reconstruction of the occluded area at a specified distance. First, based on the images from different perspectives, the depth information of the scene is calculated by parallax. Then, a threshold is set based on the depth information, and the depth difference is used to separate the foreground occluded plant marks of the closer parts of each image, that is, the pixels at the positions occluded by the marks in each perspective image are removed. Next, in a unified coordinate system, the multi-perspective images are fused to the maximum value to form a refocus of the specified area to achieve de-occlusion reconstruction.

[0055] The target detection unit is connected to the image reconstruction unit, and is used to perform target detection on the reconstructed fusion image using a deep learning target detection algorithm to determine target information within the agricultural machinery operation area.

[0056] Specifically, the detected obstacles are classified through a multimodal classifier to identify obstacles such as people, telephone poles, fences, tree trunks, etc., and the target location, category, and status data are obtained.

[0057] (III) The boundary generation module 103 is used to generate the electronic fence boundary information of the agricultural machinery operation area based on satellite maps and Geographic Information System (GIS).

[0058] In an exemplary embodiment, the boundary generation module 103 is a server. By combining high-resolution satellite maps and GIS information, the electronic fence boundary of the operation area is set, and the electronic fence boundary information can be further modified through the vehicle terminal to limit the operation area, and the real-time location information of the agricultural machinery is combined to prevent the agricultural machinery from entering the unsafe area, thereby ensuring the safety of the operation.

[0059] When the agricultural machinery is in the farmland to be worked, the location information of the agricultural machinery is obtained based on the high-precision positioning equipment on the vehicle and sent to the server. The server generates the electronic fence boundary information of the target work plot based on the high-resolution satellite map and GIS information and sends it to the vehicle terminal of the agricultural machinery. At the same time, the user can modify and set the electronic fence through the vehicle terminal interaction according to the work site environment to ensure flexible adjustment of the fence range.

[0060] (iv) The location acquisition module 104 is used to acquire the location information of the agricultural machinery in real time.

[0061] In an exemplary embodiment, the position acquisition module 104 is a vehicle-mounted positioning device 204, specifically a dual-antenna real-time dynamic carrier phase differential (Real-time kinematic, RTK) high-precision positioning device, which detects the position, driving direction, speed and other information of the agricultural machinery in real time through the vehicle-mounted positioning device 204.

[0062] The position acquisition module 104 exchanges information with the boundary generation module 103 through the communication module 205, uploads the longitude and latitude information of the agricultural machinery, and downloads high-resolution maps, electronic fences and other information.

[0063] (V) The safety monitoring module 105 is respectively connected to the terrain measurement module 101, the target detection module 102, the boundary generation module 103 and the position acquisition module 104. The safety monitoring module 105 is used to monitor the working environment of the agricultural machinery in real time according to the terrain information, the target information, the electronic fence boundary information and the position information.

[0064] Furthermore, the safety monitoring module 105 is also used to provide a safety warning and / or generate a brake control instruction for the working environment of the agricultural machinery based on the terrain information, the target information, the electronic fence boundary information and the location information, based on a pre-established safety identification mechanism, so as to ensure the safety of the operation. The safety identification mechanism can be adjusted dynamically.

[0065] In an exemplary embodiment, Figure 2 As shown, the security monitoring module 105 includes a main control unit 206 , a security decision unit 207 and an execution unit 208 .

[0066] The main control unit 206 is used to fuse and analyze the data collected by various sensors, and to monitor the working environment of the agricultural machinery in real time according to the terrain information, the target information, the electronic fence boundary information and the location information.

[0067] The safety decision unit 207 is used to provide safety warnings and / or generate brake control instructions for the working environment of the agricultural machinery based on the terrain information, the target information, the electronic fence boundary information and the location information, based on a pre-established safety judgment mechanism.

[0068] Specifically, the current operating environment of the agricultural machinery is analyzed and evaluated based on the safety judgment mechanism, and the comprehensive evaluation information of the safety status of the current operation of the agricultural machinery is output according to the analysis results, including the following three states:

[0069] (1) Safety: The working environment is risk-free and can operate normally.

[0070] (2) Warning: There is a medium to low risk, but it will not interfere with the operation for the time being and is only a warning.

[0071] (3) Danger: There are high-risk targets or they are in a dangerous state and immediate measures must be taken.

[0072] like Figure 4 As shown, the safety judgment mechanism includes the following processing logic:

[0073] Determine whether the agricultural machinery is facing the boundary of the electronic fence and is less than the safety threshold distance. If so, a warning is generated; if not, there are two judgment logics:

[0074] ① Determine whether there are obstacles in the working area. If not, a safe state is generated. If so, determine whether the obstacle is a human target. If so, a dangerous state is generated. If not, a warning is generated.

[0075] ② Determine whether there are dangerous terrains such as steep slopes, protrusions, gullies, etc. around the agricultural machinery. If not, a safe state is generated. If yes, determine whether there are protrusions or gullies in the working area ahead. If yes, a dangerous state is generated. If not, a warning is generated.

[0076] The execution unit 208 includes an alarm device and a brake control device. The alarm device is used to give an alarm according to the safety warning generated by the safety decision unit 207 , and the brake control device is used to brake the agricultural machinery according to the brake control instruction generated by the safety decision unit 207 .

[0077] Furthermore, the harvesting operation safety monitoring system further includes an operation terminal 209. The operation terminal 209 is installed in the cockpit of the agricultural machine and is used for human-machine interaction and real-time display of sensor data, warning information and operation area status.

[0078] Based on the same inventive concept, the embodiment of the present application also provides a method for monitoring the safety of agricultural machinery harvesting operations using the agricultural machinery harvesting operation safety monitoring system mentioned above, such as Figure 5 As shown, the agricultural machinery harvesting operation safety monitoring method includes the following steps 501 to 505.

[0079] Step 501: measure the terrain information of the agricultural machinery operation area in real time.

[0080] Step 502: collect the location information and images of the agricultural machinery operation area in real time. The images include visible light images, infrared images and polarization images.

[0081] Step 503: Perform target recognition based on the image to determine target information within the agricultural machinery operation area.

[0082] In an exemplary embodiment, Figure 6 As shown, step 503 includes the following steps 601 to 604.

[0083] Step 601 , respectively performing distortion correction, noise suppression and image enhancement processing on the visible light image, the infrared image and the polarization image in sequence to obtain a preprocessed visible light image, a preprocessed infrared image and a preprocessed polarization image.

[0084] Step 602: Fusing the preprocessed visible light image, the preprocessed infrared image, and the preprocessed polarization image to obtain a multimodal fused image.

[0085] Step 603: reconstruct the occluded area in the multimodal fusion image to obtain a reconstructed fusion image.

[0086] Step 604: Use a deep learning target detection algorithm to perform target detection on the reconstructed fused image to determine target information within the agricultural machinery operation area.

[0087] Step 504: Generate electronic fence boundary information of the agricultural machinery operation area based on the satellite map and the geographic information system. In addition, the electronic fence boundary information can also be sent to the vehicle terminal, and the electronic fence boundary information can be modified by the vehicle terminal.

[0088] Step 505 , real-time monitoring of the working environment of the agricultural machinery is performed based on the terrain information, the target information, the electronic fence boundary information and the location information.

[0089] This application combines forward microwave ranging sensors and lateral electromagnetic wave ranging sensors to perform real-time scanning and prediction of terrain in high-plant harvesting scenarios, predict possible operating obstacles through terrain changes, and is particularly suitable for safety hazards caused by ground unevenness changes, steep slopes of terraces, cliffs, etc. in high-plant crop harvesting. In combination with electronic fences generated by high-resolution maps and GIS maps, it overcomes the problem of insufficient perception of dangerous factors such as operating plot boundaries, negative obstacles (such as gullies), and obscured obstacles (such as ground protrusions) in high-plant crop harvesting scenarios. Further utilizing the advantages of multi-mode multi-spectral sensor fusion technology combined with visible light images, infrared images, and polarized images, through image de-occlusion and reconstruction technology, it effectively solves the problems of missed detection or false detection caused by factors such as occlusion, complex lighting, and dust in the harvesting environment of high plants (such as corn), and realizes the identification and detection of obstacles and personnel targets. In addition, the multi-mode imaging submodule uses periodic scanning to obtain images from multiple perspectives, which can effectively reduce hardware costs compared to deploying imaging device arrays at multiple points. This effectively improves the safety and reliability of agricultural machinery harvesting operations in complex environments with high plants.

[0090] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0091] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0092] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0093] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A safety monitoring system for agricultural machinery harvesting operations, characterized in that: The agricultural machinery harvesting operation safety monitoring system comprises: Topographic measurement module, used to measure the terrain information of the agricultural machinery operation area in real time; The target detection module is used to collect images of the agricultural machinery operation area in real time, and perform target recognition based on the images to determine the target information in the agricultural machinery operation area; The boundary generation module is used to generate the electronic fence boundary information of the agricultural machinery operation area based on satellite maps and geographic information systems; Position collection module, used to collect the position information of agricultural machinery in real time; The safety monitoring module is respectively connected to the terrain measurement module, the target detection module, the boundary generation module and the position acquisition module, and is used to monitor the working environment of the agricultural machinery in real time according to the terrain information, the target information, the electronic fence boundary information and the position information.

2. The agricultural machinery harvesting operation safety monitoring system according to claim 1 is characterized in that: The terrain information includes ground change trends and side terrain; The topographic measurement module includes: The microwave distance sensor is installed at the bottom of the crop divider of the agricultural machinery harvester to measure the distance between the bottom of the agricultural machinery and the ground in real time and determine the trend of terrain changes; The electromagnetic wave ranging sensor is installed on the side of the agricultural machinery and is used to measure the side terrain of the agricultural machinery in real time.

3. The agricultural machinery harvesting operation safety monitoring system according to claim 1 is characterized in that: The target detection module comprises: A slide rail is provided at the front end of the agricultural machine; A multi-mode imaging submodule, arranged on the slide rail, for collecting images of the agricultural machinery operation area in real time; A slide rail control submodule, connected to the multi-mode imaging submodule, and used to drive the multi-mode imaging submodule to move on the slide rail so that the multi-mode imaging submodule can capture images at different viewing angles; The image processing submodule is connected to the multi-mode imaging submodule and is used to perform target recognition on the image and determine target information within the agricultural machinery operation area.

4. The agricultural machinery harvesting operation safety monitoring system according to claim 3 is characterized in that: The images include visible light images, infrared images and polarization images; The multi-mode imaging submodule comprises: Visible light camera, used to collect visible light images of the agricultural machinery operation area in real time; Infrared camera, used to collect infrared images of agricultural machinery operation areas in real time; Polarized light camera, used to collect polarized images of agricultural machinery operation areas in real time.

5. The agricultural machinery harvesting operation safety monitoring system according to claim 4 is characterized in that: The image processing submodule comprises: an image preprocessing unit, connected to the visible light camera, the infrared camera and the polarized light camera, respectively, for sequentially performing distortion correction, noise suppression and image enhancement processing on the visible light image, the infrared image and the polarized image, respectively, to obtain a preprocessed visible light image, a preprocessed infrared image and a preprocessed polarized image; an image fusion unit, connected to the image preprocessing unit, and used to fuse the preprocessed visible light image, the preprocessed infrared image and the preprocessed polarization image to obtain a multimodal fused image; An image reconstruction unit, connected to the image fusion unit, and configured to reconstruct the occluded area in the multimodal fusion image to obtain a reconstructed fusion image; The target detection unit is connected to the image reconstruction unit and is used to perform target detection on the reconstructed fused image using a deep learning target detection algorithm to determine target information within the agricultural machinery operation area.

6. The agricultural machinery harvesting operation safety monitoring system according to claim 1, characterized in that: The boundary generation module is a server.

7. The agricultural machinery harvesting operation safety monitoring system according to claim 1, characterized in that: The position acquisition module is a vehicle-mounted locator.

8. The agricultural machinery harvesting operation safety monitoring system according to claim 1, characterized in that: The safety monitoring module is also used to provide safety warnings and / or generate braking control instructions for the working environment of the agricultural machinery based on the terrain information, the target information, the electronic fence boundary information and the location information, based on a pre-established safety judgment mechanism.

9. A method for monitoring the safety of agricultural machinery harvesting operations, using the agricultural machinery harvesting operation safety monitoring system according to any one of claims 1 to 8, characterized in that: The agricultural machinery harvesting operation safety monitoring method comprises: Real-time measurement of terrain information in agricultural machinery operation areas; Collect location information and images of agricultural machinery operation areas in real time; Performing target recognition based on the image to determine target information within the agricultural machinery operation area; Generate electronic fence boundary information of agricultural machinery operation areas based on satellite maps and geographic information systems; The operating environment of the agricultural machinery is monitored in real time according to the terrain information, the target information, the electronic fence boundary information and the location information.

10. The agricultural machinery harvesting operation safety monitoring method according to claim 9, characterized in that: The agricultural machinery harvesting operation safety monitoring method further comprises: The electronic fence boundary information is sent to the vehicle-mounted terminal, and the electronic fence boundary information is modified by the vehicle-mounted terminal.

Images