Static rupture disk detection method based on YOLO-FD network

By improving the structure and loss function of the YOLOv8 network, the YOLO-FD network can effectively detect static explosion fragments in complex backgrounds, solving the problems of false detection and missed detection from the perspective of UAVs, and achieving high-precision and efficient fragment detection.

CN122336591APending Publication Date: 2026-07-03XIAN TECH UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202610309368.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-13
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

Existing technologies for detecting static explosive fragments, especially from the perspective of drones, suffer from false detections and missed detections. In particular, they lack detection accuracy in complex background environments and are difficult to effectively extract features of small targets.

Method used

The YOLO-FD network is adopted, and the CMUNEXt module is introduced into the backbone part of the YOLOv8 network, and the DyHeadV4 detection head is introduced into the head part. The Inner-DIoU loss function is used for training to improve the network's global information extraction capability and detection performance.

Benefits of technology

It improves the accuracy and speed of static fragment detection, reduces the probability of false detection and missed detection, and enhances detection efficiency and robustness.

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

Abstract

The invention belongs to the technical field of target range tests, relates to a target range ammunition static explosion damage test, and provides a static rupture disk detection method based on a YOLO-FD network in order to solve the problem of data analysis of static rupture disks. The method comprises the following steps: 1, constructing a static rupture disk image data set, and obtaining a training set, a verification set and a test set for subsequent training; 2, constructing and improving a YOLO-FD neural network model for training; and 3, inputting the acquired fragment image into the trained YOLO-FD neural network model, and outputting to obtain a fragment image containing a fragment detection frame. According to the method, the target feature processing speed during detection is accelerated, and the small target detection performance of the network is enhanced; light weight is kept through reasonable design, and the extraction capability of global information of the convolutional network is enhanced; during network training, the stability is good, and the robustness is strong; the method is high in precision and speed, can effectively improve the measurement efficiency, reduces the danger coefficient, and specifically solves the problems of missing detection and false detection of the static rupture disk under the view angle of the unmanned aerial vehicle.
Need to check novelty before this filing date? Find Prior Art