Underwater static small target detection method based on feature fusion

The invention relates to the field of underwater static target detection, in particular to an underwater static small target detection method based on feature fusion. The method comprises the following steps: preprocessing, segmenting and post-processing echo data of a sonar image in sequence to obtain a potential target area; extracting Hu moment features of the potential target area, and extracting CNN depth features of the potential target area through a convolutional neural network; and splicing and fusing the extracted Hu moment features and CNN depth features to obtain a target classification result. According to the method, most of targets in the sound image can be extracted, and meanwhile, the integrity of the targets is reserved to the maximum extent; the low-level shape feature Hu moment of the target and the CNN depth feature are fused to form a feature fusion network with higher shape feature expression ability for classification, so that the precision of target detection is effectively improved;