Due to the interference of thermal noise of the instrument, sample surface characteristics, and sample staining preparation conditions during microscopic imaging with Cell microdissection, the imaging quality is easily impaired. Aiming at this problem, a denoising network model based on multi-module combination (DNMMC) is proposed. The model processes the input images using channel correlation module, multi-scale denoising module and fusion compression module, then outputs the denoised image. Furthermore, by comparing the proposed method with the current mainstream denoising methods in terms of quantitative measurement and visual quality perception, the experimental results show that DNMMC has higher robustness in denoising biological microscopic images. It can be seen that the method will significantly improve the imaging quality of biological microscopes.