Aiming at the problems that most of the current cloud and cloud shadow detection methods are prone to misdetection, serious edge detail loss and insufficient detection accuracy, a remote sensing image cloud and cloud shadow detection method based on the dual attention convolutional neural network model (RDA-Net) is proposed. In the model, the dual attention module is introduced to can effectively capture the dependence of global features, the recursive residual module is used to avoid degradation of the deep network, and the improved atrous spatial pyramid pooling module can extract multi-scale features without changing the size of the feature map. First, the remote sensing image dataset is preprocessed and made the corresponding labels, and then the Gaofen-1 WFV remote sensing image dataset is used for training and testing. Experimental results show that the proposed method can effectively improve the detection accuracy of cloud and cloud shadow, and can still obtain better edge details of cloud and cloud shadow under complex conditions.