Sep 18, 2020 · When used to recognize side-scan sonar images of shipwreck targets, the Faster R-CNN model is time-consuming and has low efficiency and a ...
Based on the Darknet-53 network, we froze part of the convolutional layer of the YOLOv3 model trained on COCO dataset images, and conducted transfer learning.
A transfer learning YOLOv3 model that can automatically recognize side-scan sonar images of shipwreck targets based on transfer learning improves detection ...
Oct 1, 2020 · Obviously, the YOLOv3 model based on transfer learning has better recognition precision, positioning precision, and overall performance than ...
Shipwreck Target Recognition in Side-Scan Sonar Images by Improved YOLOv3 Model Based on Transfer Learning Available Online ...
The You Only Look Once version 3 (YOLO V3) model is used to perform one-time neural network processing on side scan sonar images applied to sonar image ...
Jul 9, 2024 · The research results can effectively improve the detection accuracy and efficiency of multi-scale targets in side-scan sonar images.
Based on the yolov8 network, we proposed a new underwater target detection model based on upsampling and downsampling.
Tang et al. (2020b) proposed an improved YOLOv3 network based on transfer learning for the detection of underwater wreck targets, and the results showed that ...
Missing: Shipwreck | Show results with:Shipwreck
Nov 5, 2024 · Shipwreck Target Recognition in Side-Scan Sonar Images by Improved YOLOv3 Model Based on Transfer Learning. IEEE Access, 2020. Underwater ...