Soldier behavior analysis and deep learning optimization method and system

By establishing ResNet and TensorFlow models, and using particle swarm optimization algorithm to optimize parameters, combining AR skeleton model and battlefield image recognition, the problems of insmart and inaccuracy of traditional soldier behavior analysis methods are solved, and fast and accurate soldiers' behavior recognition and depth are achieved Learning optimization reduces costs.

CN119066965AInactive Publication Date: 2024-12-03CHINESE PEOPLES LIBERATION ARMY UNIT 61932
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411136211.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional soldier behavior analysis methods rely on manual discrimination and traditional image recognition, lacking AR interaction assistance functions and deep learning capabilities, making it difficult to quickly and accurately identify soldier behavior in complex battlefield environments, which is expensive and not intelligent enough.

Method used

Establish a ResNet network model and TensorFlow framework model for soldier behavior analysis, build an AR framework model, and use particle swarm optimization algorithm to optimize model parameters, combine battlefield image recognition and disability probability judgment model to improve the interactive ability and adaptive learning ability of soldier behavior recognition .

Benefits of technology

It realizes fast and accurate soldier behavior recognition, reduces human and resource costs, and improves the interaction and optimization capabilities of AR soldier behavior analysis and deep learning systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066965A_ABST
    Figure CN119066965A_ABST
Patent Text Reader

Abstract

The invention provides a soldier behavior analysis and deep learning optimization method and system, relates to the field of intelligent behavior analysis, solves the problem that a traditional soldier behavior analysis method is not intelligent enough and not accurate enough, and adopts the scheme that a ResNet network model for soldier behavior analysis is established; constructing an AR skeleton model of soldier behavior analysis; a deep learning TensorFlow framework model for soldier behavior analysis is constructed; taking the weight parameter and the soldier behavior deep learning parameter as an operation object, and performing iterative optimization operation on the operation object by using a particle swarm optimization algorithm to obtain an optimal or suboptimal parameter sequence; and regulating and controlling a network model and a framework model of soldier behavior analysis according to the optimal or suboptimal parameter sequence. According to the scheme of the invention, the deep learning network oriented to the AR soldier behavior skeleton model can be constructed, the parameters of the soldier behavior deep learning network are optimized by using the particle swarm optimization algorithm, and the interaction capability, the adaptive learning capability and the optimization capability of soldier behavior recognition are improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Pedestrian detection method based on Mask RCNN

    CN109284669A

  • Intelligent real-time supervision and error correction system and method for individual soldier queue actions

    CN112257568A

  • Human body behavior recognition method based on deep learning

    CN114220176A

  • Individual soldier tactical training and assessment intelligent system and method

    CN114399707A

  • Individual soldier tactical action judgment method based on human body posture estimation

    CN114399836A