基于声呐图像的水下目标检测算法OA

Underwater object detection based on sonar images is one of the key tasks of underwater intelligent sensing systems.However,inherent defects in the sonar imaging mechanism,such as missing target details and blurred edges,hinder the model from learning stable discriminative features.To address this issue,a high-precision underwater object detection algorithm based on sonar images is proposed in this study.Based on a single-stage target detection framework,it introduces a space mapping depth convolution module,a feature enhancement module,and a spatial and channel collaborative attention mechanism to achieve lossless downsampling,enhance shallow feature representation,and improve feature spatial sensitivity and channel selectivity.Experimental results on the sonar common target detection dataset demonstrate that this method effectively improves detection accuracy while reducing the number of model parameters.The experimental results verify the effectiveness and practicality of the proposed method,achieving a good balance between lightweight parameters and high detection accuracy,providing reliable technical support for optimizing the object detection performance of underwater intelligent sensing systems.