Image content identification system for water source sites

Abstract

The present invention relates to an image content identification system for a water source site, comprising: a day and night capture mechanism, which is used to perform a bird's-eye view action facing the water surface of the water source to obtain a water surface capture image; a processing and analysis device, which is used to use an AI judgment model to intelligently identify the types of image processing that the water surface capture image has undergone based on various on-site information of the water source of the agricultural irrigation system, the distribution area of ​​the photoelectric sensor components, the number of pixels, the exposure time and the shutter speed, the multiple visual contents of the water surface capture image and the multiple visual contents of the BAYER image corresponding to the water surface capture image. Through the present invention, it is possible to use an AI judgment model with a targeted structural design to intelligently identify the types of image processing that the water surface capture image has undergone from the BAYER image corresponding to the water surface capture image to the multiple image processing that the water surface capture image has undergone based on the screened basic data, thereby providing an effective basis for the identification of the true content of the capture image at the water source site of agricultural irrigation.

Description

Technical Field

[0001] The invention relates to the field of agricultural irrigation, and in particular to an image content identification system used at a water source site. Background Art

[0002] Agricultural irrigation methods can generally be divided into traditional surface irrigation, ordinary sprinkler irrigation and micro-irrigation. Traditional surface irrigation includes furrow irrigation, furrow irrigation, flooding irrigation and overwater irrigation, but these irrigation methods often consume a lot of water and have low water utilization efficiency, which is a very unreasonable agricultural irrigation method. In addition, ordinary sprinkler irrigation technology is a more common irrigation method in agricultural production. However, the water utilization efficiency of ordinary sprinkler irrigation technology is not high. Modern agricultural micro-irrigation technologies include micro-sprinkler irrigation, drip irrigation, infiltration irrigation, etc. These irrigation technologies generally have good water-saving performance and higher water utilization rate than traditional irrigation modes.

[0003] Water sources are important monitoring objects of agricultural irrigation systems. The water quality, on-site scenes and water ripple fluctuations of water sources directly determine the irrigation quality and irrigation effect of agricultural irrigation systems. Therefore, a special monitoring mechanism, such as a visual monitoring mechanism, is generally designed to monitor and process water sources of agricultural irrigation systems. However, due to the fixed structure of visual monitoring equipment and the need for video transmission, the monitoring images output by the visual monitoring mechanism are often not the original image content of the water source of the agricultural irrigation system, but the image content obtained through more than one image processing, which may cause monitoring personnel to misjudge the water quality, on-site scenes and water ripple fluctuations of the water source of the agricultural irrigation system. Summary of the invention

[0004] In order to solve the technical problems in the related fields, the present invention provides an image content identification system for water source sites, which uses the water source area, average water source depth and wind force value at the water source site of the water source of the agricultural irrigation system, the distribution area of ​​the photoelectric sensor components, the number of pixels, the exposure time and the shutter speed, multiple visual contents of the water surface snapshot image and the multiple visual contents of the BAYER image corresponding to the water surface snapshot image as the input content of the AI ​​judgment model one by one, so as to run the AI ​​judgment model, obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model, and convert the multiple image locations through which the water surface snapshot image passes. The processing type is determined, thereby providing an effective basis for the identification of the real content of the captured images at the water source site of agricultural irrigation. Among them, an AI judgment model with a targeted structural design is used to perform intelligent judgment on the types of image processing from the BAYER image corresponding to the water surface captured image to the multiple copies of the water surface captured image. The targeted structural design of the artificial intelligence model is manifested in the introduction of a target training device for performing each training operation on the deep neural network to obtain the deep neural network after each training operation is performed and output as the AI ​​judgment model. The number of training operations performed by the deep neural network is monotonically positively correlated with the water source area of ​​the water source of the agricultural irrigation system.

[0005] According to the present invention, there is provided an image content identification system for a water source site, the system comprising:

[0006] On-site detection institutions are used to obtain various on-site information of the water source of the agricultural irrigation system. The various on-site information of the water source of the agricultural irrigation system includes the water source area, average water source depth and wind force value at the water source site;

[0007] The day and night capture mechanism is used to perform a downward shooting action facing the water surface of the water source to obtain and output corresponding water surface capture images;

[0008] A parameter extraction mechanism, used to extract the distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components of the day and night capture mechanism as various sensing parameter outputs of the day and night capture mechanism;

[0009] A target training device is used to perform each training operation on the deep neural network to obtain the deep neural network after each training operation is performed and output as an AI judgment model, wherein the number of training operations performed by the deep neural network is monotonically positively correlated with the water source area of ​​the water source of the agricultural irrigation system;

[0010] The processing and analysis device is respectively connected to the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to use the water source area, average water source depth and wind force value at the water source site of the agricultural irrigation system, the distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components, multiple visual contents of the water surface capture image and multiple visual contents of the BAYER image corresponding to the water surface capture image as the input content of the AI ​​judgment model item by item, so as to run the AI ​​judgment model and obtain the BAYER image corresponding to the water surface capture image output by the AI ​​judgment model to the multiple image processing types that the water surface capture image has undergone;

[0011] Among them, the water source area, average water source depth and wind force value at the water source site, distribution area, number of pixels, exposure time and shutter speed of photoelectric sensor components, multiple visual contents of water surface snapshot images and multiple visual contents of BAYER images corresponding to the water surface snapshot images of the water source of the agricultural irrigation system are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model, and the multiple image processing types from the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model to the water surface snapshot image include: the numerical representation of the multiple image processing types from the BAYER image corresponding to the water surface snapshot image to the water surface snapshot image is a numerical combination of ASCII codes of the multiple type names corresponding to the multiple image processing types from the BAYER image corresponding to the water surface snapshot image to the water surface snapshot image;

[0012] Among them, the water source area of ​​the water source of the agricultural irrigation system, the average depth of the water source and the wind speed value at the water source, the distribution area of ​​the photoelectric sensor components, the number of pixels, the exposure time and the shutter speed, the multiple visual contents of the water surface snapshot image and the multiple visual contents of the BAYER image corresponding to the water surface snapshot image are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model to obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model. The multiple image processing types that the water surface snapshot image undergoes also include: each image processing type is one of distortion processing, curve processing, color level adjustment, point image restoration, filtering processing, enhancement processing and exponential image enhancement processing.

[0013] The present invention has at least the following three important invention points:

[0014] Important invention point 1: The water source area, average water source depth and wind force value at the water source site, distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components, multiple visual contents of the water surface snapshot image and multiple visual contents of the BAYER image corresponding to the water surface snapshot image of the water source of the agricultural irrigation system are used as the input content of the AI ​​judgment model item by item, so as to run the AI ​​judgment model and obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model to obtain multiple types of image processing to the water surface snapshot image, thereby providing an effective basis for the identification of the true content of the snapshot image at the water source site of agricultural irrigation;

[0015] Important invention point 2: Specifically, the multiple visual contents of the BAYER image corresponding to the water surface snapshot image are the color component values ​​of each pixel point of the BAYER image corresponding to the water surface snapshot image in the CMYK color space, and the BAYER image corresponding to the water surface snapshot image is the image directly output by the photoelectric sensor component of the day and night snapshot mechanism, thereby providing reliable and sufficient basic information for intelligent identification;

[0016] Important invention point three: An AI judgment model with targeted structural design is used to perform intelligent judgment on the types of multiple image processing from the BAYER image corresponding to the water surface snapshot image to the water surface snapshot image. The targeted structural design of the artificial intelligence model is manifested in the introduction of a target training device for performing each training operation on the deep neural network to obtain a deep neural network after performing each training operation and output it as the AI ​​judgment model. The number of training operations performed by the deep neural network is monotonically positively correlated with the water source area of ​​the water source of the agricultural irrigation system. DETAILED DESCRIPTION

[0017] The following is a detailed description of the implementation scheme of the image content identification system for water source sites of the present invention.

[0018] The image content identification system for water source site shown in embodiment A of the present invention includes:

[0019] On-site detection institutions are used to obtain various on-site information of the water source of the agricultural irrigation system. The various on-site information of the water source of the agricultural irrigation system includes the water source area, average water source depth and wind force value at the water source site;

[0020] For example, the on-site detection mechanism is used to obtain various on-site information of the water source of the agricultural irrigation system, and the various on-site information of the water source of the agricultural irrigation system is the water source area, the average water source depth and the wind force value at the water source of the agricultural irrigation system. The on-site detection mechanism has various built-in value detection units, which are used to respectively detect the water source area, the average water source depth and the wind force value at the water source of the agricultural irrigation system;

[0021] The day and night capture mechanism is used to perform a downward shooting action facing the water surface of the water source to obtain and output corresponding water surface capture images;

[0022] A parameter extraction mechanism, used to extract the distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components of the day and night capture mechanism as various sensing parameter outputs of the day and night capture mechanism;

[0023] A target training device is used to perform each training operation on the deep neural network to obtain the deep neural network after each training operation is performed and output as an AI judgment model, wherein the number of training operations performed by the deep neural network is monotonically positively correlated with the water source area of ​​the water source of the agricultural irrigation system;

[0024] The processing and analysis device is respectively connected to the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to use the water source area, average water source depth and wind force value at the water source site of the agricultural irrigation system, the distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components, multiple visual contents of the water surface capture image and multiple visual contents of the BAYER image corresponding to the water surface capture image as the input content of the AI ​​judgment model item by item, so as to run the AI ​​judgment model and obtain the BAYER image corresponding to the water surface capture image output by the AI ​​judgment model to the multiple image processing types that the water surface capture image has undergone;

[0025] Among them, the water source area, average water source depth and wind force value at the water source site, distribution area, number of pixels, exposure time and shutter speed of photoelectric sensor components, multiple visual contents of water surface snapshot images and multiple visual contents of BAYER images corresponding to the water surface snapshot images of the water source of the agricultural irrigation system are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model, and the multiple image processing types from the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model to the water surface snapshot image include: the numerical representation of the multiple image processing types from the BAYER image corresponding to the water surface snapshot image to the water surface snapshot image is a numerical combination of ASCII codes of the multiple type names corresponding to the multiple image processing types from the BAYER image corresponding to the water surface snapshot image to the water surface snapshot image;

[0026] The water source area, average water source depth and wind force value at the water source site, distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components, multiple visual contents of the water surface snapshot image and multiple visual contents of the BAYER image corresponding to the water surface snapshot image of the water source of the agricultural irrigation system are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model, and the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model is obtained. The multiple image processing types to which the water surface snapshot image is subjected also include: each image processing type is one of distortion processing, curve processing, color scale adjustment, point image restoration, filtering processing, enhancement processing and exponential image enhancement processing;

[0027] The water source area, average water source depth and wind force value at the water source site, distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components, multiple visual contents of the water surface snapshot image and multiple visual contents of the BAYER image corresponding to the water surface snapshot image of the water source of the agricultural irrigation system are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model, and the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model is obtained. The multiple image processing types of the water surface snapshot image also include: the multiple visual contents of the water surface snapshot image are the values ​​of each color component of each pixel point of the water surface snapshot image in the CMYK color space;

[0028] Among them, the water source area of ​​the water source of the agricultural irrigation system, the average depth of the water source and the wind force value at the water source site, the distribution area of ​​the photoelectric sensor component, the number of pixels, the exposure time and the shutter speed, the multiple visual contents of the water surface snapshot image and the multiple visual contents of the BAYER image corresponding to the water surface snapshot image are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model to obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model. The types of image processing that the water surface snapshot image undergoes also include: the BAYER image corresponding to the water surface snapshot image is the image directly output by the photoelectric sensor component of the day and night capture mechanism.

[0029] Compared with embodiment A, the image content identification system for water source site shown in embodiment B may also include the following components:

[0030] A power detection mechanism, comprising a plurality of power detection units, for respectively detecting the current input power of the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device;

[0031] Wherein, the power detection mechanism includes a plurality of power detection units, and is used to respectively detect the current input power of the processing and analysis device, the field detection mechanism, the day and night capture mechanism, the parameter extraction mechanism, and the target training device, including: calculating the current input power of each of the processing and analysis device, the field detection mechanism, the day and night capture mechanism, the parameter extraction mechanism, and the target training device by respectively measuring the real-time power supply voltage and the real-time power supply current for the processing and analysis device, the field detection mechanism, the day and night capture mechanism, the parameter extraction mechanism, and the target training device;

[0032] The calculation of the respective current input powers is completed by measuring the real-time power supply voltage and the real-time power supply current for the processing and analysis device, the field detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device respectively, including: for any device of the processing and analysis device, the field detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, its current input power is the product of its real-time power supply voltage and its real-time power supply current;

[0033] Wherein, the power detection mechanism includes a plurality of power detection units, which are used to respectively detect the current input power of the processing and analysis device, the field detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and further includes: the plurality of power detection units respectively used for the processing and analysis device, the field detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device have the same power measurement upper limit threshold and power measurement lower limit threshold;

[0034] Among them, the power detection mechanism includes multiple power detection units, which are used to respectively detect the current input power of the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device. It also includes: the internal structures of the multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device are the same.

[0035] Compared with implementation A, the image content identification system for water source sites shown in implementation C may also include the following components:

[0036] An on-site notification mechanism is respectively connected to a plurality of power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to perform corresponding power notification operations based on power measurement results of the plurality of power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device;

[0037] Wherein, the on-site notification mechanism is respectively connected to a plurality of power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to perform corresponding power notification operations based on the power measurement results of the plurality of power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, including: the on-site notification mechanism includes a static storage unit for storing a power notification threshold;

[0038] And wherein, the on-site notification mechanism is respectively connected to multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to perform corresponding power notification operations based on the power measurement results of the multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device. It also includes: the on-site notification mechanism performs corresponding power notification operations for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device based on the voice playback mode.

[0039] In addition, in the image content identification system for water source site, the water source area of ​​the water source of the agricultural irrigation system, the average depth of the water source and the wind speed value at the water source site, the distribution area of ​​the photoelectric sensor components, the number of pixels, the exposure time and the shutter speed, the multiple visual contents of the water surface snapshot image and the multiple visual contents of the BAYER image corresponding to the water surface snapshot image are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model to obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model. The types of image processing that the water surface snapshot image undergoes also include: the multiple visual contents of the BAYER image corresponding to the water surface snapshot image are the color component values ​​of each pixel point of the BAYER image corresponding to the water surface snapshot image in the CMYK color space.

[0040] The image content identification system for water source sites of the present invention is used to address the technical problem in the prior art that the true content of captured images of water sources for agricultural irrigation is difficult to accurately identify. By adopting an AI judgment model with a targeted structural design, the system intelligently identifies the types of multiple image processing from the BAYER image corresponding to the water surface captured image to the water surface captured image based on the screened basic data, thereby providing an effective basis for the identification of the true content of the captured images of water sources for agricultural irrigation, and solving the above-mentioned technical problem.

[0041] While the present invention has been described in conjunction with the specific exemplary embodiments, it should be understood that the invention is not limited to the disclosed embodiments, but on the contrary, the invention is intended to cover various modifications and equivalent structures included within the spirit and scope of the appended claims and their equivalents.

Claims

1. An image content identification system for water source sites, characterized in that: The system comprises: On-site detection institutions are used to obtain various on-site information of the water source of the agricultural irrigation system. The various on-site information of the water source of the agricultural irrigation system includes the water source area, average water source depth and wind force value at the water source site; The day and night capture mechanism is used to perform a downward shooting action facing the water surface of the water source to obtain and output corresponding water surface capture images; A parameter extraction mechanism, used to extract the distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components of the day and night capture mechanism as various sensing parameter outputs of the day and night capture mechanism; A target training device is used to perform each training operation on the deep neural network to obtain the deep neural network after each training operation is performed and output as an AI judgment model, wherein the number of training operations performed by the deep neural network is monotonically positively correlated with the water source area of ​​the water source of the agricultural irrigation system; The processing and analysis device is respectively connected to the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to use the water source area, average water source depth and wind force value at the water source site of the agricultural irrigation system, the distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components, multiple visual contents of the water surface capture image and multiple visual contents of the BAYER image corresponding to the water surface capture image as the input content of the AI ​​judgment model item by item, so as to run the AI ​​judgment model and obtain the BAYER image corresponding to the water surface capture image output by the AI ​​judgment model to the multiple image processing types that the water surface capture image has undergone; Among them, the water source area, average depth of the water source and wind force value at the water source site, distribution area of ​​photoelectric sensor components, number of pixels, exposure time and shutter speed, multiple visual contents of water surface snapshot images and multiple visual contents of BAYER images corresponding to the water surface snapshot images of the water source of the agricultural irrigation system are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model, and the multiple image processing types from the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model to the water surface snapshot image include: the numerical representation of the multiple image processing types from the BAYER image corresponding to the water surface snapshot image to the water surface snapshot image is a numerical combination of the ASCII code names corresponding to the multiple image processing types from the BAYER image corresponding to the water surface snapshot image to the water surface snapshot image; Among them, the water source area of ​​the water source of the agricultural irrigation system, the average depth of the water source and the wind speed value at the water source, the distribution area of ​​the photoelectric sensor components, the number of pixels, the exposure time and the shutter speed, the multiple visual contents of the water surface snapshot image and the multiple visual contents of the BAYER image corresponding to the water surface snapshot image are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model to obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model. The multiple image processing types that the water surface snapshot image undergoes also include: each image processing type is one of distortion processing, curve processing, color level adjustment, point image restoration, filtering processing, enhancement processing and exponential image enhancement processing.

2. The image content identification system for water source site as claimed in claim 1, characterized in that: The water source area, average water source depth and wind force value at the water source site, distribution area, number of pixels, exposure time and shutter speed of the photoelectric sensor components, multiple visual contents of the water surface snapshot image and multiple visual contents of the BAYER image corresponding to the water surface snapshot image of the water source of the agricultural irrigation system are used as input contents of the AI ​​judgment model item by item to run the AI ​​judgment model to obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model. The types of image processing to which the water surface snapshot image undergoes further include: the multiple visual contents of the water surface snapshot image are the values ​​of each color component of each pixel point of the water surface snapshot image in the CMYK color space; Among them, the water source area of ​​the water source of the agricultural irrigation system, the average depth of the water source and the wind force value at the water source site, the distribution area of ​​the photoelectric sensor component, the number of pixels, the exposure time and the shutter speed, the multiple visual contents of the water surface snapshot image and the multiple visual contents of the BAYER image corresponding to the water surface snapshot image are used as the input content of the AI ​​judgment model item by item to run the AI ​​judgment model to obtain the BAYER image corresponding to the water surface snapshot image output by the AI ​​judgment model. The types of image processing that the water surface snapshot image undergoes also include: the BAYER image corresponding to the water surface snapshot image is the image directly output by the photoelectric sensor component of the day and night capture mechanism.

3. The image content identification system for water source site according to claim 2, characterized in that: The system further comprises: A power detection mechanism, comprising a plurality of power detection units, for respectively detecting the current input power of the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device; Among them, the power detection mechanism includes multiple power detection units, which are used to respectively detect the current input power of the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, including: by respectively measuring the real-time power supply voltage and the real-time power supply current for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device to complete the calculation of their respective current input powers.

4. The image content identification system for water source site as claimed in claim 3, characterized in that: The calculation of the respective current input powers is completed by measuring the real-time power supply voltage and the real-time power supply current for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device respectively, including: for any device of the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, its current input power is the product of its real-time power supply voltage and its real-time power supply current.

5. The image content identification system for water source site as claimed in claim 3, characterized in that: The power detection mechanism includes multiple power detection units, which are used to respectively detect the current input power of the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device. It also includes: the multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device have the same power measurement upper limit threshold and power measurement lower limit threshold.

6. The image content identification system for water source site as claimed in claim 5, characterized in that: The power detection mechanism includes multiple power detection units, which are used to respectively detect the current input power of the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device. It also includes: the internal structures of the multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device are the same.

7. The image content identification system for water source site according to claim 2, characterized in that: The system further comprises: The on-site notification mechanism is respectively connected to multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to perform corresponding power notification operations based on the power measurement results of the multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device.

8. The image content identification system for water source site as claimed in claim 7, characterized in that: The on-site notification mechanism is respectively connected to multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to perform corresponding power notification operations based on the power measurement results of the multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, including: the on-site notification mechanism includes a static storage unit for storing power notification thresholds.

9. The image content identification system for water source site according to claim 8, characterized in that: The on-site notification mechanism is respectively connected to multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device, and is used to perform corresponding power notification operations based on the power measurement results of the multiple power detection units respectively used for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device. It also includes: the on-site notification mechanism performs corresponding power notification operations for the processing and analysis device, the on-site detection mechanism, the day and night capture mechanism, the parameter extraction mechanism and the target training device based on the voice playback mode.