Landslide Recognition of ResNet Network Model with Fusion Attention Mechanism

A landslide target recognition algorithm based on a residual convolutional neural network with fusion attention mechanism is proposed in this paper to solve the problems of strong subjectivity and complex data processing in traditional landslide recognition methods. The algorithm is based on the residual convolutional neural network framework, and introduces the convolutional block attention mechanism to enhance the model's attention to local features of landslide images, extract the correlation information of landslide features from two dimensions of channel and space, and realize landslide target recognition. The experimental results show that the recognition accuracy of the improved network model can reach 97.2%, and the convergence speed is fast, which has higher efficiency and recognition accuracy.