Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 3, 2026Scientific ReportsOpen Access

Small target detection of floating objects in river channels based on improved YOLOv7

View Full Paper
Ask AI
Bookmark
Share

Authors

BWBing WangBZBing ZhangSGSu Guo

Discussion

Loading...

Member takes

Overview

Demonstrates enhanced small target detection in river channels, indicating improved precision in challenging environments.

Key Points

  • The aim is to improve the detection accuracy of small floating objects in dynamic river environments using advanced computer vision techniques.
  • Introduced Region-Overlap Detection (ROD) method using Minimum Convoluted YOLOv7 (MCY) architecture.
  • Utilized YOLO classifier to identify the largest overlap area in multiple overlapping regions.
  • Extracted bounding boxes with minimal convolution from the final training layer of the neural network.
  • Achieved a mean Average Precision (mAP) of 73.1% for small floating object detection.
  • Obtained a recall rate of 70.2% in detecting small targets in dynamic river settings.

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/69a67eb2f353c071a6f0a098https://doi.org/10.1038/s41598-026-40688-z
View Full Paper
Ask AI
Bookmark
Share