Underwater Image Enhancement Based on Dense Residual Blocks and Attention Mechanism

Underwater images usually suffer from blurring, contrast degradation, color deviation, and loss of local details, which hinder the visual quality and subsequent processing of underwater images. In order to address these challenges, this study proposes an underwater image enhancement method based on dense residual blocks and attention mechanisms. The encoder introduces a contextual feature aggregation module to extract multi-scale features from the input image for global feature aggregation. In the decoder, the dense residual block enriches the detailed features of the image. At the same time, the bi-level routing attention mechanism captures critical local detail features, thus effectively reducing local detail loss. Experimental results on standard datasets show that the proposed method outperforms previous deep learning models in recovering blur, color bias, and local details in underwater images.