Saliency target detection algorithm of hierarchical multi-receptive field network

The invention belongs to the field of computer vision, and provides a saliency target detection algorithm of a hierarchical multi-receptive field network, which comprises the following steps: 1) taking ResNet-50 as a backbone framework to extract multi-scale feature information from an RGB image and then encoding; 2) utilizing a hierarchical multi-receptive-field convolution mechanism to optimize the multi-level features to generate high-quality features; and 3) performing complementarity fusion on the optimized features by using an invisible relation feature fusion mechanism, and then generating a final saliency map. Compared with the prior art, according to the saliency target detection algorithm of the hierarchical multi-receptive-field network, convolution operation of different hierarchical receptive fields is utilized to optimize multi-level features and carry out stealth relation feature fusion, and the multi-level features are optimized layer by layer to generate a high-performance saliency map. 本发明属于