Ground target detection distributed system based on artificial intelligence

Abstract

The application discloses a ground target detection distributed system based on artificial intelligence and relates to the technical field of ground target detection.The system comprises the following steps: S1, a data acquisition module, which is used for collecting original graphic data, digital information and language data; S2, an artificial intelligence analysis module, which is used for analyzing and classifying image data collected by the data acquisition module in a system coverage area; S3, a ground target detection module; and S4, a central control module.The ground target detection distributed system based on artificial intelligence can quickly explore a target area without relying on satellites, has two different modes, namely, a passive mode and an active mode, and adopts a distributed cooperative search mode, so that the operation speed of the system is greatly increased, and the ground target can be quickly searched and found.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ground target detection, in particular to a ground target detection distributed system based on artificial intelligence. BACKGROUND

[0002] Nowadays, the demand for ground target recognition technology is increasing, and ground target recognition plays an important role in the fields of automobile auxiliary driving, intelligent traffic monitoring, military, etc. Target recognition is a basic topic in the fields of image processing and machine vision. In fact, people often use their eyes and brains to recognize objects in learning, work and life. With the invention of computers in the 1940s and the continuous in-depth study of artificial intelligence technology by human beings, people are eager to play the role of eyes through image sensors and use computers to replace or assist human beings in mental work.

[0003] The existing ground target recognition system is not intelligent enough and relies heavily on satellites, which inevitably leads to an increase in cost or causes the system to be unable to function properly when competitors are unwilling to provide satellite services. At the same time, it is not intelligent enough, resulting in poor system operation efficiency. Therefore, we propose such a ground target detection distributed system based on artificial intelligence. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a ground target detection distributed system based on artificial intelligence, which solves the problems raised in the background art.

[0005] To achieve the above purpose, the present application realizes the following technical scheme, a ground target detection distributed system based on artificial intelligence, comprising the following steps:

[0006] S1, a data acquisition module, the data acquisition module is used for collecting original graphic data, digital information and language data, and the data acquisition module comprises: a fixed data collection unit, a mobile target collection unit and a data information summary unit, the data information summary unit is used for summarizing and classifying the image data collected in the fixed data collection unit and the mobile target collection unit, the data acquisition module is provided with two or more than two, and each data acquisition module corresponds to a calibration area;

[0007] S2, an artificial intelligence analysis module, the artificial intelligence analysis module is used for analyzing and classifying the image data collected in the data acquisition module in the system coverage area, and the artificial intelligence analysis module comprises an image data classification unit, an image data storage unit, an image data comparison unit and an encrypted data sharing unit, the artificial intelligence analysis module is provided with two or more than two;

[0008] S3, a ground target detection module, the ground target detection module is used for analyzing and classifying the data transmitted by the artificial intelligence analysis module and confirming the ground target, and the ground target detection module includes a ground target model establishing unit, a passive ground target searching unit and an active ground target searching unit;

[0009] S4, a central control module, the central control module is used for controlling the data acquisition module, the artificial intelligence analysis module and the ground target detection module and transmitting data, and the central control module includes a data transmission unit, a central cloud database unit and an image display unit, the data transmission unit is used for data transmission between the data acquisition module, the artificial intelligence analysis module and the ground target detection module, the central cloud database unit is used for storing the data and image data model transmitted by the data acquisition module, the artificial intelligence analysis module and the ground target detection module, and the image display unit is used for displaying the image analysis results of the data acquisition module, the artificial intelligence analysis module and the ground target detection module.

[0010] Further, the fixed data collection unit of the data acquisition module is provided with at least four or more acoustic imaging detectors and video sensors, and the fixed sensors are distributed in a grid shape, the fixed data collection unit collects image data in the calibration area through the fixed detectors, the fixed data collection unit is also provided with at least four or more fixed infrared sensors, and the fixed infrared sensors are distributed in a grid shape, the fixed data collection unit collects image data in the calibration area at night through the fixed infrared sensors.

[0011] Further, the mobile target collection unit in the data acquisition module is provided with at least four or more unmanned aerial vehicle mobile platforms, and the unmanned aerial vehicles are each installed with visible light and infrared video sensors, the unmanned aerial vehicles cruise in the calibration area, and the mobile target collection unit collects image data in the calibration area through the video sensors on the unmanned aerial vehicles, the mobile target collection unit is also provided with a detection vehicle mobile platform, and the detection vehicle is installed with a vehicle-mounted video sensor and a vehicle-mounted laser imaging radar, and the mobile target collection unit collects image data in the calibration area through the video sensor and the laser imaging radar.

[0012] Further, the image data classification unit in the artificial intelligence analysis module is used for analyzing the image data collected in the data acquisition module, and the image classification model adopted by the image data classification unit includes any two or more combinations of AlexNet, VGG, GoogLeNet and ResNet.

[0013] Furthermore, the image data storage unit in the artificial intelligence analysis module is used to store image data classified by the image data classification unit and image data models.

[0014] Furthermore, the image data comparison unit in the artificial intelligence analysis module is used to compare the new image model classified in the image data classification unit with the image model in the image data storage unit. If the difference between the two is greater than 60%, the data is encrypted through the data chain to form a key.

[0015] Furthermore, the encrypted data sharing unit in the artificial intelligence analysis module is used to encrypt and share the data in the image data storage unit with the data in the adjacent image data storage unit of the artificial intelligence analysis module. The encrypted data sharing unit is also used to transmit the key formed by comparing the new image model classified in the image data classification unit with the image model in the image data storage unit with a difference of more than 60% to the passive ground target search unit.

[0016] Furthermore, the ground target model building unit in the ground target detection module builds a model of the ground target using any two combinations of AlexNet, VGG, GoogLeNet, and ResNet.

[0017] Furthermore, the passive ground target search unit in the ground target detection module is used to input the target established in the ground target model building unit into the central control module, and search the database through the central cloud database unit to find data that matches the ground target model.

[0018] Furthermore, the active ground target search unit in the ground target detection module is used to search for ground targets in the calibration area corresponding to each data acquisition module through the fixed data collection unit and the mobile target collection unit in the data acquisition module.

[0019] This invention provides a distributed ground target detection system based on artificial intelligence, which has the following beneficial effects:

[0020] The ground target detection distributed system based on artificial intelligence does not need to rely on satellites, can quickly explore the target area, has two different modes of mobile and active, and greatly speeds up the operation speed of the system in a distributed cooperative search manner, can quickly search and find the ground target, and the system inputs the target in the central control module through the ground target model establishment unit, and searches the database through the central cloud database unit to find data conforming to the ground target model; The active ground target search unit is used for searching the ground target in the calibration area corresponding to each data acquisition module through the fixed data collection unit and the mobile target collection unit in the data acquisition module; The fixed data collection unit collects image data in the calibration area through a fixed camera, and at least four or more fixed infrared sensors are arranged in the fixed data collection unit, and the fixed infrared sensors are distributed in a grid shape, and the fixed data collection unit collects image data in the calibration area at night through the fixed infrared sensors. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 It is a flowchart of the ground target detection distributed system based on artificial intelligence of the present application;

[0022] Figure 2 It is a flowchart of the data acquisition module of the ground target detection distributed system based on artificial intelligence of the present application;

[0023] Figure 3 It is a flowchart of the artificial intelligence analysis module of the ground target detection distributed system based on artificial intelligence of the present application;

[0024] Figure 4 It is a flowchart of the ground target detection module of the ground target detection distributed system based on artificial intelligence of the present application;

[0025] Figure 5 It is a flowchart of the central control module of the ground target detection distributed system based on artificial intelligence of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below with specific embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all.

[0027] Please refer to Figure 1 and Figure 3 The ground target detection distributed system based on artificial intelligence comprises the following steps:

[0028] S1, a data acquisition module, the data acquisition module is used for the collection of original graphic data, digital information and language data, and the data acquisition module comprises: a fixed data collection unit, a mobile target collection unit and a data information summary unit, the data information summary unit is used for summarizing and classifying the image data collected in the fixed data collection unit and the mobile target collection unit, and the data acquisition module is provided with two or more than two, and each data acquisition module corresponds to a calibration area;

[0029] S2, an artificial intelligence analysis module, the artificial intelligence analysis module is used for analyzing and classifying the image data collected in the data acquisition module in the system coverage area, and the artificial intelligence analysis module comprises an image data classification unit, an image data storage unit, an image data comparison unit and an encrypted data sharing unit, and the artificial intelligence analysis module is provided with two or more than two;

[0030] S3, a ground target detection module, the ground target detection module is used for analyzing and classifying the data transmitted by the artificial intelligence analysis module and confirming the ground target, and the ground target detection module comprises a ground target model establishing unit, a passive ground target searching unit and an active ground target searching unit;

[0031] S4, a central control module, the central control module is used for controlling the data acquisition module, the artificial intelligence analysis module and the ground target detection module and transmitting data, and the central control module comprises a data transmission unit, a central cloud database unit and an image display unit, the data transmission unit is used for data transmission between the data acquisition module, the artificial intelligence analysis module and the ground target detection module, the central cloud database unit is used for storing the data and image data model transmitted by the data acquisition module, the artificial intelligence analysis module and the ground target detection module, and the image display unit is used for displaying the image analysis results of the data acquisition module, the artificial intelligence analysis module and the ground target detection module.

[0032] Please refer to Figure 1 and Figure 2 , specifically, the fixed data collection unit of the data acquisition module is provided with four or more than four acoustic imaging detectors and video sensors, and the fixed sensors are distributed in a grid shape, the fixed data collection unit collects image data in the calibration area through the fixed detectors, the fixed data collection unit is also provided with four or more than four fixed infrared sensors, and the fixed infrared sensors are distributed in a grid shape, and the fixed data collection unit collects image data in the calibration area at night through the fixed infrared sensors.

[0033] The mobile target collecting unit in the data acquisition module is provided with at least four or more unmanned aerial vehicle mobile platforms, and the unmanned aerial vehicles are each provided with a visible light and infrared video sensor, the unmanned aerial vehicles cruise in the calibration area, and the mobile target collecting unit collects image data in the calibration area through the video sensor on the unmanned aerial vehicle, and the mobile target collecting unit is further provided with a detection vehicle mobile platform, and the detection vehicle is provided with a vehicle-mounted video sensor and a vehicle-mounted laser imaging radar, and the mobile target collecting unit collects image data in the calibration area through the video sensor and the laser imaging radar.

[0034] Referring to Figure 1 and Figure 4 , specifically, the image data classification unit in the artificial intelligence analysis module is used for analyzing the image data collected in the data acquisition module, and the image classification model adopted by the image data classification unit includes any two or more combinations of AlexNet, VGG, GoogLeNet and ResNet.

[0035] The image data storage unit in the artificial intelligence analysis module is used for storing the image data classified by the image data classification unit and the image data model.

[0036] The image data comparison unit in the artificial intelligence analysis module is used for comparing the new image model classified in the image data classification unit with the image model in the image data storage unit, and if the difference between the two is higher than 60%, the data is encrypted to form a key through the data chain.

[0037] The encrypted data sharing unit in the artificial intelligence analysis module is used for encrypting and sharing the data in the image data storage unit with the data in the image data storage unit of the adjacent artificial intelligence analysis module, and the encrypted data sharing unit is used for transmitting the key formed by the data with a difference higher than 60% between the new image model classified in the image data classification unit and the image model in the image data storage unit to the passive ground target searching unit.

[0038] Referring to Figure 1 and Figure 5 , specifically, the ground target model establishing unit in the ground target detection module establishes a model of the ground target through any two combinations of AlexNet, VGG, GoogLeNet and ResNet.

[0039] The passive ground target searching unit in the ground target detection module is used for inputting the target established by the ground target model establishing unit into the central control module, and searching for data conforming to the ground target model through the central cloud database unit.

[0040] The active ground target searching unit in the ground target detection module is used for searching the ground target in the calibration area corresponding to each data acquisition module through the fixed data collection unit and the mobile target collection unit in the data acquisition module.

[0041] In summary, the ground target detection distributed system based on artificial intelligence firstly inputs the target established in the ground target model establishment unit into the central control module, and searches the database through the central cloud database unit to find data conforming to the ground target model; the active ground target searching unit is used for searching the ground target in the calibration area corresponding to each data acquisition module through the fixed data collection unit and the mobile target collection unit in the data acquisition module.

[0042] The fixed data collection unit collects image data in the calibration area through the fixed camera, and at least four or more fixed infrared sensors are arranged in the fixed data collection unit and distributed in a grid shape, so that the fixed data collection unit collects image data in the calibration area at night through the fixed infrared sensors.

[0043] Then, at least four or more unmanned aerial vehicles are arranged in the mobile target collection unit in the data acquisition module, and a gimbal camera, an airborne laser imaging radar and an infrared sensor are installed on each unmanned aerial vehicle, the unmanned aerial vehicles cruise in the calibration area, and the mobile target collection unit collects image data in the calibration area through the gimbal camera, the airborne laser imaging radar and the infrared sensor on the unmanned aerial vehicles, a mobile mapping vehicle is arranged in the mobile target collection unit, and a vehicle-mounted camera and a vehicle-mounted laser imaging radar are installed on the mobile mapping vehicle, and the mobile target collection unit collects image data in the calibration area through the vehicle-mounted camera and the vehicle-mounted laser imaging radar.

[0044] The image data classification unit in the artificial intelligence analysis module is used for analyzing the image data collected in the data acquisition module, and the image data storage unit in the artificial intelligence analysis module is used for storing the image data classified by the image data classification unit and the image data model.

[0045] Finally, the data transmitted by the artificial intelligence analysis module is analyzed and classified by the ground target detection module, and the ground target is confirmed.

Claims

1. A distributed ground target detection system based on artificial intelligence, characterized in that, Includes the following steps: S1. Data acquisition module, which is used to collect raw graphic data, digital information and language data, and includes: a fixed data collection unit, a moving target collection unit and a data information aggregation unit. The data information aggregation unit is used to summarize and classify the image data collected in the fixed data collection unit and the moving target collection unit. There are at least two data acquisition modules, and each data acquisition module corresponds to a calibration area. S2. Artificial intelligence analysis module, which is used to analyze and classify the image data collected by the data acquisition module in the system coverage area. The artificial intelligence analysis module includes an image data classification unit, an image data storage unit, an image data comparison unit, and an encrypted data sharing unit. The artificial intelligence analysis module is provided with at least two units. S3, Ground Target Detection Module, which is used to analyze and classify the data transmitted by the artificial intelligence analysis module and to confirm the ground targets. The ground target detection module includes a ground target model building unit, a passive ground target search unit and an active ground target search unit. S4. Central Control Module: The central control module is used to control the data acquisition module, the artificial intelligence analysis module, and the ground target detection module, as well as to transmit data. The central control module includes a data transmission unit, a central cloud database unit, and an image display unit. The data transmission unit is used for data transmission between the data acquisition module, the artificial intelligence analysis module, and the ground target detection module. The central cloud database unit is used to store the data and image data models transmitted by the data acquisition module, the artificial intelligence analysis module, and the ground target detection module. The image display unit is used to display the image analysis results stored in the data acquisition module, the artificial intelligence analysis module, and the ground target detection module. The fixed data collection unit of the data acquisition module is equipped with at least four acoustic imaging detectors and video sensors, and the fixed sensors are arranged in a grid pattern. The fixed data collection unit collects image data in the calibration area through the fixed detectors. The fixed data collection unit is also equipped with at least four fixed infrared sensors, and the fixed infrared sensors are arranged in a grid pattern. The fixed data collection unit collects image data in the calibration area at night through the fixed infrared sensors. The mobile target collection unit in the data acquisition module is equipped with at least four UAV mobile platforms, each equipped with visible light and infrared video sensors. The UAVs patrol the designated area, and the mobile target collection unit collects image data of the designated area through the video sensors on the UAVs. The mobile target collection unit is also equipped with a detection vehicle mobile platform, which is equipped with vehicle-mounted video sensors and vehicle-mounted laser imaging radar. The mobile target collection unit collects image data of the designated area through the video sensors and laser imaging radar. The passive ground target search unit in the ground target detection module is used to input the target established in the ground target model building unit into the central control module, and search the database through the central cloud database unit to find data that matches the ground target model; The active ground target search unit in the ground target detection module is used to search for ground targets in the calibration area corresponding to each data acquisition module through the fixed data collection unit and the mobile target collection unit in the data acquisition module.

2. The distributed ground target detection system based on artificial intelligence according to claim 1, characterized in that: The image data classification unit in the artificial intelligence analysis module is used to analyze the image data collected in the data acquisition module, and the image classification model used by the image data classification unit includes any two or more combinations of AlexNet, VGG, GoogLeNet and ResNet.

3. The distributed ground target detection system based on artificial intelligence according to claim 1, characterized in that: The image data storage unit in the artificial intelligence analysis module is used to store image data classified by the image data classification unit and image data models.

4. The distributed ground target detection system based on artificial intelligence according to claim 1, characterized in that: The image data comparison unit in the artificial intelligence analysis module is used to compare the new image model classified in the image data classification unit with the image model in the image data storage unit. If the difference between the two is greater than 60%, the data is encrypted through the data chain to form a key.

5. The distributed ground target detection system based on artificial intelligence according to claim 1, characterized in that: The encrypted data sharing unit in the artificial intelligence analysis module is used to encrypt and share the data in the image data storage unit with the data in the adjacent image data storage unit of the artificial intelligence analysis module. The encrypted data sharing unit is also used to transmit the key formed by the data of the new image model classified in the image data classification unit and the image model in the image data storage unit with a difference of more than 60% to the passive ground target search unit.

6. The distributed ground target detection system based on artificial intelligence according to claim 1, characterized in that: The ground target detection module's ground target model building unit builds a model of the ground target using any two combinations of AlexNet, VGG, GoogLeNet, and ResNet.

Images