基于多尺度融合卷积神经网络的微光图像增强算法

Abstract

To overcome the problems of poor generalizability and an inability to adapt to complex real scenes of traditional lowlight image enhancement algorithms, a new method based on multiscale concat convolutional neural network is proposed here. This method achieves lowlight image enhancement by learning the mapping relationship between lowlight and normal images. Taking the lowlight image as input, the shallow layer information of the image is extracted through the preprocessing module. Then, Selective Kernel Network (SKNet) is fused to the local path to form a feature extraction network. Finally, the global feature is fused with the local feature, which is obtained by weight learning of the feature map with a channel attention module. Bilateral guided upsampling is used to restore the image size and obtain the mapping function of the lowlight image, after which the image enhancement is completed. Based on the MIT -Adobe 5K dataset, a comparative experiment with nine other advanced methods showed that the proposed method can effectively improve the brightness and details of lowlight images. Hence, the proposed method is superior to other contrast algorithms in terms of visual effects and quantitative evaluation.