A ship detection method and system based on compression pruning of complex models

By modifying the size of the convolution kernel, increasing the number of convolution channels and the number of convolution kernels in the ship detection model, and performing compression pruning and phased model training, the problems of slow ship detection speed and high hardware requirements in the existing technology are solved, and high precision and fast detection are achieved.

CN118982801BActive Publication Date: 2025-04-29ZHEJIANG WHYIS TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411458275.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-10-18
Publication Date
2025-04-29
Estimated Expiration
2044-10-18

AI Technical Summary

Technical Problem

In the prior art, ship inspection cannot speed up the inspection speed while ensuring accuracy, and has high hardware requirements.

Method used

The compression pruning method based on complex models is adopted to improve the model complexity and receptive field by modifying the convolution kernel size, number of convolution channels and the number of convolution kernels in each module of the backbone network; at the same time, the low-weight convolution layer, channel and kernel are reset to 0, the number of model parameters is reduced, and the model structure is optimized through phased model training and fine-tuning.

Benefits of technology

It improves the accuracy and speed of ship detection, reduces the demand for hardware computing power, and ensures that the model inference speed is accelerated while ensuring the detection rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982801B_ABST
    Figure CN118982801B_ABST
Patent Text Reader

Abstract

The present invention discloses a ship detection method and system based on compression pruning of complex models. The method effectively improves the complexity and receptive field of the original ship detection model by modifying the convolution kernel size, the number of convolution channels, and the number of convolution kernels of each module of the backbone network, thereby improving the accuracy of ship detection; by resetting the lower weights of layer convolution, convolution channels, and convolution kernels to 0, the number of model parameters is reduced, thereby reducing the model's demand for hardware computing power and accelerating the ship detection speed; adopting staged model training, fine-tuning the model when modifying it, thereby achieving the speed of inference of the ship model while ensuring the ship detection rate; deleting the convolution channels and convolution kernels with a weight of 0, thereby reducing the model's demand for hardware computing power and further accelerating the ship detection speed.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Lightweight SAR image ship detection model and method based on strip pruning

    CN114283331A

  • Infrared ship target detection method based on improved yolov7

    CN117152691A