Coal Mine Image Dust and Fog Clearing Algorithm Based on Deep Learning Network

The purpose of coal mine dust fog image sharpening is to recover clear content from low visibility images. In response to the problems of over-enhancement and insufficient applicability of traditional image defogging methods based on prior knowledge, this paper proposed an end-to-end parallel high-resolution deep learning network to directly restore the final clear images. In this network, we used a parallel high-resolution deep network structure to deeply integrate multi-scale features between branches to help the network obtain contextual information. We also proposed a lightweight attention module, which reduced the parameter burden of the model and paid more attention to the target area when extracting feature information. Finally, in order to evaluate the effectiveness and versatility of the clarification algorithm proposed in this paper, experiments were conducted on the public data set RESIDE and the coal mine dust and fog image data set produced by myself, and compared with the existing classic clarification algorithm GCANet. The results show that the algorithm proposed in this paper can effectively solve the over-enhancement phenomenon, and improve the clarity and visualization of coal mine images.