Low-light image enhancement aims to improve the perception of images collected in dim environments and provide high-quality data support for image recognition tasks. When dealing with photographs captured under non-uniform illumination, existing methods cannot adaptively extract the differentiated luminance information, which will easily cause overexposure and underexposure. From the perspective of unsupervised learning, we propose a multi-scale attention Transformer named MSATr, which sufficiently extracts local and global features for light balance to improve the visual quality. Specifically, we present a multi-scale window division scheme, which uses exponential sequences to adjust the window size of each layer. Within different-sized windows, the self-attention computation can be refined, ensuring the pixel-level feature processing capability of the model. For feature interaction across windows, a global transformer branch is constructed to provide comprehensive brightness perception and alleviate exposure problems. Furthermore, we propose a loop training strategy, in which diverse images generated by weighted mixing and a luminance consistency loss are used to effectively improve the model’s generalization ability. Extensive experiments on several benchmark datasets quantitatively and qualitatively prove that our MSATr is superior to state-of-the-art low-light image enhancement methods. The enhanced images have more natural brightness and outstanding details. The code is released at https://github.com/fang001021/MSATr.