Structured channel pruning target recognition lightweight method and device and medium

Through the structured channel pruning method, the batch normalized layer scaling factor is used to measure the importance of the convolution channel, the redundant connections of the bomb-load platform model are cut off, and lightweight object detection is realized, which solves the model adaptability problem under the limitation of hardware resources and improves detection efficiency and accuracy.

CN120298674APending Publication Date: 2025-07-11CHINESE PEOPLES LIBERATION ARMY ARMY ARTILLERY & AIR DEFENSE ACAD
0 Cites 2 Cited by

Patent Information

Application Number
CN202510464245.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-14
Publication Date
2025-07-11

AI Technical Summary

Technical Problem

The hardware resources of the bomb-loaded platform are limited, and the traditional target detection algorithm model is huge and time-consuming to calculate, making it difficult to directly apply to the bomb-loaded platform.

Method used

The structured channel pruning method is adopted, and the scaling factor of the batch normalization layer is used as the standard to measure the importance of convolutional channels. The model is lightweighted through pruning redundant weight connection, combined with sparse training and cropping threshold calculation.

Benefits of technology

Significantly reduce the amount of model parameters and calculations, maintain detection accuracy, adapt to the hardware resource limitations of the bomb-mounted platform, and improve the deployment efficiency and reliability of the model in resource-constrained environments.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a structured channel pruning target recognition lightweight method, which is used for lightweight reconstruction of a target detection algorithm model, and comprises the following steps executed by computer equipment: S1, in a network model training process, taking a scaling factor of a batch normalization layer as an importance degree standard for measuring a convolution channel, performing linear transformation on each output feature image pixel of the convolutional layer to accelerate model network convergence; s2, calculating a cutting threshold according to the maximum value of the scale factor of the batch normalization layer, and deleting redundant weight connections of which the model performance influence is lower than the cutting threshold through a channel structure pruning compression operation so as to reduce the model parameter quantity and the calculation quantity; and S3, through network re-training fine tuning, model lightweight is realized. According to the method, coding and other operations are not needed, software and hardware implementation and deployment are facilitated, the detection precision can be almost not affected, the comprehensive performance of the model is improved, and the effect of model lightweight is achieved.
Need to check novelty before this filing date? Find Prior Art