Light-weight small target detection method for coal mine complex scene

By using an improved ML-YOLO model and employing C3k2_MDConv, LAPConv, and MBPA detection heads, the problems of lightweight models and insufficient accuracy in small target detection in underground coal mines were solved, achieving efficient and accurate detection in complex scenarios.

CN122289653APending Publication Date: 2026-06-26XIAN UNIV OF SCI & TECH
0 Cites 0 Cited by

Patent Information

Application Number
CN202610400598.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-30
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing technologies for detecting small targets in underground coal mines suffer from increased model parameters and computational overhead, making it difficult to meet the lightweight requirements of edge computing scenarios. At the same time, the detection accuracy is insufficient, especially in complex scenarios where small targets are easily lost or falsely detected.

Method used

An improved ML-YOLO model is adopted. By replacing the convolutional modules of the YOLOv11s model with C3k2_MDConv and LAPConv modules in the backbone network, and using the MBPA detection head in the detection head, the feature extraction and detection process is optimized, the number of model parameters and computational overhead are reduced, and the detection accuracy of small objects is improved.

Benefits of technology

It significantly reduces the number of model parameters and computational overhead, improves the detection accuracy and robustness of small targets in complex downhole scenarios, and achieves reliable detection in resource-constrained environments.

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

Abstract

The invention belongs to the technical field of target detection, and discloses a coal mine complex scene-oriented lightweight small target detection method, which comprises the following steps of: inputting an image to be detected into a trained ML-YOLO model; obtaining and outputting a target detection result of the to-be-detected image by the ML-YOLO model; wherein the ML-YOLO model is obtained through improvement based on a YOLOv11s model, the ML-YOLO model comprises a backbone network, a neck network and a detection head, and the improvement comprises the steps that in the backbone network, a C3k2MDConv module is used for replacing a C3k2 module in the YOLOv11s model, and an LAPCConv module is used for replacing a convolution module in the deep extraction stage of the YOLOv11s model; in the neck network, a C3k2MDConv module is used for replacing a C3k2 module in the YOLOv11s model; in the detection head, replacing an original detection head of the YOLOv11s model with an MBPA detection head; according to the method, on the premise of keeping the detection capability, the parameter quantity and the calculation overhead can be remarkably reduced, and compared with the prior art, the method has better lightweight performance and can better meet the requirement of a coal mine underground edge calculation scene for model deployment.
Need to check novelty before this filing date? Find Prior Art