Visual system and field detection device
Through the intelligent vision system combined with ordinary cameras, infrared cameras and ambient light sensors, deep learning algorithms are used to identify mango growth status and pests, solving the problems of traditional inefficiency in monitoring and the impact of light, and achieving high-quality image acquisition and accurate recognition.
Patent Information
- Application Number
- CN202411749314.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-02
- Publication Date
- 2025-05-06
AI Technical Summary
Traditional mango planting monitoring relies on manual observation, is inefficient and susceptible to personal experience and lighting conditions. The existing image recognition technology is insufficient in robustness and accuracy in complex environments.
Design an intelligent vision system, including acquisition module, processing module, analysis module and execution module, use ordinary cameras and infrared cameras, combine with ambient light sensors to automatically adjust exposure parameters, and use deep learning algorithms such as convolutional neural networks for image feature extraction and recognition.
It realizes the acquisition of high-quality images under different lighting conditions, improves the accuracy and efficiency of mango growth status and pest recognition, provides management decisions in real time, and improves response speed and decision quality.
Smart Images
Figure CN119935998A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to a visual system and an on-site detection device. Background Art
[0002] Traditional mango planting monitoring and management work mainly relies on manual observation and empirical judgment. This method is time-consuming and labor-intensive, and is easily affected by personal experience and conditions of the day, resulting in unstable monitoring results. This is not only inefficient, but also easily affected by human factors, resulting in inaccurate recognition. With the development of computer vision technology, although modern image recognition technology has been applied in many fields, its application in agriculture, especially in mango planting, is not mature enough. The robustness and accuracy of existing technologies under complex environmental conditions need to be improved. In particular, mango is a crop that is sensitive to light conditions, and its growth state and maturity are significantly affected by light. Existing image acquisition systems often have difficulty maintaining consistent image quality under different lighting conditions. Summary of the invention
[0003] In view of the above situation, in order to overcome the defects of the prior art, the purpose of the present invention is to provide a visual system and an on-site detection device. The above technical objectives are achieved through the following technical solutions: An intelligent vision system based on mango planting includes: a collection module for capturing image information during the growth of mangoes, a processing module for improving image quality and providing clear image data for subsequent feature extraction and recognition, an analysis module for analyzing image information, and an execution module for executing corresponding measures according to data analysis results; The acquisition module includes an image acquisition terminal, an acquisition card connected to the image acquisition terminal, a dual-channel communication interface built into the acquisition card, a temporary storage device for periodically temporarily storing image data, and an information terminal for collecting acquisition card information. The information terminal and the temporary storage device are respectively connected to the dual-channel communication interface on the acquisition card by wired connection to receive data information from the acquisition card. The processing module is connected to the information terminal, which receives information from the acquisition card. The processing module sequentially performs noise reduction, illumination correction, contrast enhancement, and image sharpening on the image information. The noise reduction step removes random noise in the image by median filtering, and the illumination correction reduces the influence of uneven illumination in the image information. The contrast enhancement adjusts the local area of the image information by adaptive histogram equalization to adapt to the contrast adjustment of different areas. The image sharpening process uses a high-pass filter to enhance the high-frequency part of the image information and increase the image clarity. The analysis module performs statistical analysis on the recognition results, evaluates the growth status of mangoes, extracts and classifies the on-site image information features through a convolutional neural network, obtains characteristic information such as the size, shape, and color of mangoes, evaluates the growth status of mangoes, and then determines the anthrax status of mangoes during their growth according to the distribution of brown and black dead spots on the surface of mangoes and the shrinkage degree of mango leaves; The execution module includes an irrigation system, a fertilization system, a pesticide spraying system and an environmental control system placed on site.
[0004] Furthermore, the acquisition module also includes an adaptive camera unit, which includes a controller, an ambient light sensor and an infrared camera. The ambient light sensor monitors the light intensity around the mango growth environment in real time. The ambient light sensor transmits the monitoring data to the controller, and the monitoring signal is transmitted to the image acquisition terminal through the controller, so that the image acquisition terminal can adjust its own aperture, shutter, ISO and other camera information according to the lighting information. At the same time, it is also independently analyzed by the controller. When the controller determines that the current lighting environment is insufficient and the image acquisition terminal cannot capture high-quality image information, the infrared camera also arranged on the scene is controlled to perform the shooting work at the same time.
[0005] Furthermore, the analysis module outputs the result content according to the recognition result, and gives adjustment suggestions using the data model. The above content is presented through a display control terminal for fruit farmers to review. The display control terminal has a touch function, and the display control terminal integrates a system for controlling the execution module, so that fruit farmers can remotely control the operation of the execution module.
[0006] A field detection device based on an intelligent visual system for mango planting, comprising a hardware support unit connected and installed with a collection module and used to support the movement and adjustment of an image collection terminal, the hardware support unit comprising a device base, a support pole, an upper cross bar, and a linear servo module, the device base is arranged on one side of a mango tree on site, the entire device base is embedded in the soil for fixation, the support pole is fixed to the upper part of the device base, the upper cross bar is fixed to the upper part of the support pole, the linear servo module is fixed to the bottom of the upper cross bar, the image collection terminal and an infrared camera are installed on the linear servo module and are controlled to perform lateral movement.
[0007] Furthermore, the image acquisition terminal and the infrared camera are provided with a waterproof and dustproof cover on the outside. The waterproof and dustproof cover is a transparent structure, which enhances the shooting quality of the image acquisition terminal and the infrared camera without affecting their normal ability to obtain image information. The entire waterproof and dustproof cover moves horizontally synchronously with the image acquisition terminal and the infrared camera.
[0008] Furthermore, the base of the device is annular in structure as a whole, with a fastening groove for soil burial inside, a circle of counterweight grooves is provided at the upper end of the base of the device, and counterweight blocks are provided inside for weight increase, and a stabilizing chassis embedded in the soil is provided at the lower end of the base of the device, and the stabilizing chassis is connected to the base of the device by a connecting rod.
[0009] Furthermore, it comprises a control box, which is buried underground and is completely closed except for wiring.
[0010] In summary, the present invention has the following beneficial effects: The present invention uses a combination of an ordinary camera and an infrared camera to capture image information of the growth of mango trees in an orchard, and uses an ambient light sensor to automatically control the camera aperture, shutter, and ISO value, that is, automatically adjust the camera's exposure parameters to ensure that the best high-quality images can be obtained under different lighting conditions; The present invention adopts a deep learning algorithm, especially a convolutional neural network, to automatically extract and identify mango features in image information, thereby improving the accuracy and efficiency of recognition. The system of the present invention can analyze image data in real time and quickly provide targeted management decisions, such as irrigation, fertilization, and pest control, thereby improving response speed and decision quality. Combined with the intelligent environmental control system, it can automatically adjust the environmental conditions of the greenhouse or growth room according to the growth needs of mangoes to achieve precise environmental control; By collecting and analyzing a large amount of mango growth data, the system can provide data-driven planting optimization suggestions to help growers improve the scientific nature of crop management. The system has self-learning capabilities and can continuously optimize the recognition algorithm based on new data to adapt to mangoes of different varieties and growth environments. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application, but do not constitute an improper limitation of the present invention. In the drawings: Figure 1 It is a working flow diagram of the system of the present invention; Figure 2 is a system connection diagram of the present invention; Figure 3 It is a device connection diagram of the present invention.
[0012] In the figure, 1. device base; 2. support pole; 3. upper cross bar; 4. linear servo module; 5. waterproof and dustproof cover; 6. counterweight block; 7. stabilization chassis. DETAILED DESCRIPTION
[0013] The above and other technical contents, features and effects of the present invention are described in detail below with reference to the attached Figure 1 To Attachment Figure 3 The detailed description of the embodiments will clearly show that the structural contents mentioned in the following embodiments are all based on the drawings in the specification.
[0014] Exemplary embodiments of the present invention will be described below with reference to the accompanying drawings. Example
[0015] An intelligent vision system based on mango planting includes: a collection module for capturing image information during the growth process of mangoes, a processing module for improving image quality and providing clear image data for subsequent feature extraction and recognition, an analysis module for analyzing image information, and an execution module for executing corresponding measures according to data analysis results.
[0016] The acquisition module is the basis of the entire intelligent visual recognition system. It is responsible for capturing image information during the growth process of mangoes. It mainly includes an image acquisition terminal, an acquisition card connected to the image acquisition terminal, a dual-channel communication interface built into the acquisition card, a temporary storage device for periodically temporarily storing image data, and an information terminal for collecting acquisition card information. The image acquisition terminal uses an ordinary camera to access the pan-tilt head to realize rotation and video operations. It can be remotely controlled manually or automatically controlled through program settings. According to the selected video mode, the information terminal and the temporary storage device are respectively connected to the dual-channel communication interface on the acquisition card through wired connections to receive data information from the acquisition card.
[0017] The processing module is connected to the information terminal and is responsible for improving the image quality and providing clear image data for subsequent feature extraction and recognition. The information terminal receives information from the acquisition card, and the processing module performs noise reduction, illumination correction, contrast enhancement, and image sharpening on the image information in turn. The noise reduction step removes random noise in the image through median filtering, and illumination correction reduces the impact of uneven illumination in the image information. Contrast enhancement adjusts the local area of the image information through adaptive histogram equalization to adapt to the contrast adjustment of different areas. Image sharpening uses a high-pass filter to enhance the high-frequency part of the image information and increase image clarity.
[0018] The analysis module is the core part of the intelligent visual recognition system. It is responsible for extracting useful information from the preprocessed images and conducting further analysis to identify the growth status, maturity, and pests and diseases of mangoes. The analysis module conducts statistical analysis on the recognition results, evaluates the growth status of mangoes, extracts and classifies the on-site image information features through convolutional neural networks, obtains characteristic information such as mango size, shape, and color, and evaluates the growth status of mangoes. Secondly, the anthracnose situation during the growth process of mangoes is judged based on the distribution of brown and black spots on the surface of mangoes and the degree of shrinkage of mango leaves.
[0019] The execution module includes an on-site irrigation system, a fertilization system, a pesticide spraying system, and an environmental control system. The execution module executes corresponding measures based on the data information of the analysis module, or provides suggestions for fruit farmers to manually execute some measures. For example, the switch and irrigation amount of the irrigation system can be automatically adjusted according to the growth status and environmental conditions of the mango. Fertilization suggestions, including fertilizer type and amount, can be provided according to the nutritional status of the mango. After identifying pests and diseases, the spraying system can be automatically triggered to spray pesticides at a fixed point, or suggestions for manual spraying can be provided. The temperature and light conditions of the mango growth environment are monitored to meet the needs of mango growth, and finally the maturity of the mango is monitored and a harvesting plan is formulated to ensure that the mango is harvested at the best time.
[0020] The acquisition module also includes an adaptive camera unit, which includes a controller, an ambient light sensor and an infrared camera. The ambient light sensor monitors the light intensity around the mango growth environment in real time. The ambient light sensor transmits the monitoring data to the controller, and the monitoring signal is transmitted to the image acquisition terminal through the controller, so that the image acquisition terminal can adjust its own aperture, shutter, ISO and other camera information according to the lighting information. At the same time, it is also independently analyzed by the controller. When the controller determines that the current lighting environment is insufficient and the image acquisition terminal cannot capture high-quality image information, the infrared camera also arranged on the scene is controlled to perform shooting at the same time. Under extremely poor lighting conditions, because infrared light can penetrate certain obstacles that visible light cannot penetrate, infrared cameras can be used for image acquisition, thereby making up for the poor ability of ordinary cameras to capture image information in this environment.
[0021] The analysis module outputs the result content based on the recognition result and gives adjustment suggestions using the data model. The above content is presented through a display control terminal for fruit farmers to review. The display control terminal has a touch function, and the display control terminal integrates a system for controlling the execution module, so that fruit farmers can remotely control the execution module.
[0022] A field detection device based on an intelligent visual system for mango planting, comprising a hardware support unit connected to an acquisition module and used to support the movement and adjustment of an image acquisition terminal, the hardware support unit comprising a device base, a support pole, an upper cross bar, and a linear servo module. The device base is arranged on one side of a mango tree on site, the entire device base is embedded in the soil for fixation, the support pole is fixed to the upper part of the device base, the upper cross bar is fixed to the upper part of the support pole, the linear servo module is fixed to the bottom of the upper cross bar, the image acquisition terminal and an infrared camera are installed on the linear servo module and are controlled to perform lateral movement.
[0023] The image acquisition terminal and the infrared camera are provided with waterproof and dustproof covers on the outside. The waterproof and dustproof covers are transparent structures. While enhancing the shooting quality of the image acquisition terminal and the infrared camera, they do not affect their normal ability to obtain image information. The entire waterproof and dustproof cover moves horizontally synchronously with the image acquisition terminal and the infrared camera.
[0024] The base of the device is annular in structure as a whole, with a fastening groove inside for burying soil. A circle of counterweight grooves is provided at the upper end of the base of the device, and counterweight blocks are provided inside for adding weight. The lower end of the base of the device is provided with a stabilizing chassis embedded in the soil, and the stabilizing chassis and the base of the device are connected by a connecting rod.
[0025] It includes a control box which is buried underground and is completely closed except for wiring.
[0026] The above is a further detailed description of the present invention in combination with a specific implementation method, and it cannot be determined that the specific implementation of the present invention is limited to this; for technical personnel in the technical field to which the present invention belongs and related technical fields, based on the technical solution of the present invention, the expansion and replacement of operating methods and data should all fall within the protection scope of the present invention.
Claims
1. An intelligent visual system based on mango planting includes, characterized in that: It includes an acquisition module for capturing image information during the growth of mangoes, a processing module for improving image quality and providing clear image data for subsequent feature extraction and recognition, an analysis module for analyzing image information, and an execution module for executing corresponding measures according to the data analysis results; The acquisition module includes an image acquisition terminal, an acquisition card connected to the image acquisition terminal, a dual-channel communication interface built into the acquisition card, a temporary storage device for periodically temporarily storing image data, and an information terminal for collecting acquisition card information. The information terminal and the temporary storage device are respectively connected to the dual-channel communication interface on the acquisition card by wired connection to receive data information from the acquisition card. The processing module is connected to the information terminal, which receives information from the acquisition card. The processing module sequentially performs noise reduction, illumination correction, contrast enhancement, and image sharpening on the image information. The noise reduction step removes random noise in the image by median filtering, and the illumination correction reduces the influence of uneven illumination in the image information. The contrast enhancement adjusts the local area of the image information by adaptive histogram equalization to adapt to the contrast adjustment of different areas. The image sharpening process uses a high-pass filter to enhance the high-frequency part of the image information and increase the image clarity. The analysis module performs statistical analysis on the recognition results, evaluates the growth status of mangoes, extracts and classifies the on-site image information features through a convolutional neural network, obtains characteristic information such as the size, shape, and color of mangoes, evaluates the growth status of mangoes, and then determines the anthrax status of mangoes during their growth according to the distribution of brown and black dead spots on the surface of mangoes and the shrinkage degree of mango leaves; The execution module includes an irrigation system, a fertilization system, a pesticide spraying system and an environmental control system placed on site.
2. The intelligent visual system based on mango planting according to claim 1, characterized in that: The acquisition module also includes an adaptive camera unit, which includes a controller, an ambient light sensor and an infrared camera. The ambient light sensor monitors the light intensity around the mango growth environment in real time. The ambient light sensor transmits the monitoring data to the controller, and the monitoring signal is transmitted to the image acquisition terminal through the controller, so that the image acquisition terminal can adjust its own aperture, shutter, ISO and other camera information according to the lighting information. At the same time, it is also independently analyzed by the controller. When the controller determines that the current lighting environment is insufficient and the image acquisition terminal cannot capture high-quality image information, the infrared camera also arranged on the scene is controlled to perform shooting at the same time.
3. The intelligent visual system based on mango planting according to claim 1, characterized in that: The analysis module outputs the result content according to the recognition result, and gives adjustment suggestions using the data model. The above content is presented through a display control terminal for fruit farmers to review. The display control terminal has a touch function, and the display control terminal integrates a system for controlling the execution module, so that fruit farmers can remotely control the operation of the execution module.
4. An on-site detection device based on an intelligent visual system for mango planting according to any one of claims 1 to 3, characterized in that: It includes a hardware support unit connected to the acquisition module and used to support the movement and adjustment of the image acquisition terminal. The hardware support unit includes a device base, a support pole, an upper cross bar, and a linear servo module. The device base is arranged on one side of the mango tree on site, and the entire device base is embedded in the soil for fixation. The support pole is fixed to the upper part of the device base, the upper cross bar is fixed to the upper part of the support pole, and the linear servo module is fixed to the bottom of the upper cross bar. The image acquisition terminal and the infrared camera are installed on the linear servo module and are controlled to perform lateral movement.
5. An on-site detection device according to claim 4, characterized in that: The image acquisition terminal and the infrared camera are provided with a waterproof and dustproof cover on the outside. The waterproof and dustproof cover is a transparent structure. While enhancing the image acquisition terminal and the infrared camera's shooting quality, it does not affect their normal ability to obtain image information. The entire waterproof and dustproof cover moves laterally synchronously with the image acquisition terminal and the infrared camera.
6. The on-site detection device according to claim 4, characterized in that: The base of the device is annular in structure as a whole, with a fastening groove inside for burying soil, a circle of counterweight grooves is provided at the upper end of the base of the device, and counterweight blocks are provided inside for adding weight, and a stabilizing chassis embedded in the soil is provided at the lower end of the base of the device, and the stabilizing chassis is connected to the base of the device by a connecting rod.
7. The on-site detection device according to claim 4, characterized in that: It includes a control box which is buried underground and is completely closed except for wiring.