CCM-Net: Color compensation and coordinate attention guided underwater image enhancement with multi-scale feature aggregation

•We design an asymmetric encoder-decoder network, called CCM-Net, which employs the adaptive color compensation module, the global-local coordinate attention module, and the multi-scale feature aggregation module to solve the problem of inconsistent attenuation in different channels and spatial regions in underwater images.•We propose a color compensation (CC) module, which is used to solve the problem of inconsistent channel attenuation in underwater images by leveraging the green channel to adaptively compensate for heavily attenuated blue and red channels.•We introduce global-local coordinate attention (GLCA), which not only considers horizontal and vertical attention but also focuses on global and local information, aiming to extract features flexibly while considering the position and spatial information.•We present an effective multi-scale feature aggregation (MFA) module, which fuses color features in shallow layers, texture features in middle layers, and high-level semantic features in deep layers. The MFA module dynamically adjusts the weight of features based on the correlation between different layers to improve the enhancement effect of contrast, clarity, and color reproduction of underwater images.•We conduct extensive experiments on real-world underwater images and the results show the outstanding performance of our CCM-Net in terms of visual quality and objective metrics. Due to the light scattering and wavelength absorption in water, underwater images exhibit blurred details, low contrast, and color deviation. Existing underwater image enhancement methods are divided into traditional methods and deep learning-based methods. Traditional methods either rely on scene prior and lack robustness, or are not flexible enough resulting in poor enhancement effects. Deep learning methods have achieved good results in the field of underwater image enhancement due to their powerful feature representation ability. However, these methods cannot enhance underwater images with various degradations because they do not consider the inconsistent attenuation of different color channels and spatial regions. In this paper, we propose a novel asymmetric encoder-decoder network for underwater image enhancement, called CCM-Net. Concretely, we first introduce the prior knowledge-based encoder, which includes color compensation (CC) modules and feature extraction modules that consist of depth-wise separable convolution and global-local coordinate attention (GLCA). The