NPT-UL: An Underwater Image Enhancement Framework Based on Nonphysical Transformation and Unsupervised Learning

Underwater images often face visual degradation problems such as color deviations, low contrast, and blurred details. Previous studies suffer from poor visual perception performance and overenhancement, which pose a challenge to improve the visibility of underwater images comprehensively and consistently. This article proposes a framework for underwater image enhancement based on nonphysical transformation and unsupervised learning (NPT-UL). The framework includes an image data augmentation strategy and an UL model. First, a data augmentation strategy based on NPT is proposed, which consists of three types of transformation processes to optimize the color deviation, contrast, and detail blur in the initial image dataset. Second, an UL model based on contrast learning and generative adversarial network (GAN) is proposed to enhance the input underwater image. Three loss functions including adversarial loss, PatchNCE loss, and identity loss are consolidated into the network to ensure structural similarity and color authenticity. Finally, qualitative, quantitative, and ablative experiments are presented to evaluate the performance using different underwater image datasets. Compared with the state-of-the-art methods, NPT-UL has better performance which can effectively improve the image visual quality. Tested in the UIEB dataset, the values of underwater color image quality evaluation (UCIQE), colorfulness, contrast, and fog density (CCF), underwater image quality measurement (UIQM), BC(e), and BC(r) metrics obtained by NPT-UL reach 0.644, 42.311, 4.325, 4.761, and 2.719, separately. The proposed framework can help to solve the problems of color distortion and feature loss, showing significant application potential in the field of underwater image restoration and target detection.