Abstract Image restoration is a vital research area in computer vision, focusing on reconstructing high-quality clear images from degraded observations. Common types of degradation include noise and blur, which may stem from imaging device limitations, environmental interference, and other factors. This paper centers on the design and optimization of multi-stage image restoration networks, conducting in-depth exploration of feature extraction, feature fusion, attention mechanisms, and their practical applications. A multi-stage hybrid attention mechanism-based image restoration network is proposed. Initially, each stage progressively extracts and restores image features. Then, an adaptive feature fusion block enables effective cross-stage information transfer. Finally, by calculating losses at each stage and assigning different weights, the network achieves stable convergence during training. The hybrid attention mechanism enhances the model’s focus on critical features and improves its understanding of the overall image structure. Outstanding performance has been achieved in both image deblurring and denoising tasks. On the GoPro dataset, the restored results achieved a PSNR of 33.26 and an SSIM of 0.963. On the SIDD dataset, the restored results reached a PSNR of 40.23 and an SSIM of 0.963. Furthermore, ablation experiments demonstrated the effectiveness of the multi-stage model, hybrid attention mechanism, and adaptive feature fusion block. Similar content being viewed by others Data availability The datasets generated and/or analysed during the current study are available in the public repositories listed below. For image de-blurring: the GoPro blur dataset (3214 images, 1280 \(\times\) 720 px) was downloaded from https://github.com/SeungjunNah/DeepDeblur_release. The originally provided train/test split (2103/1111 images) was adopted. High-resolution images were cropped into 512 \(\times\) 512 px patches to accelerate training and inference. For image de-noising: the Smartphone Image Denoising Dataset (SIDD) was obtained from https://www.eecs.yorku.ca/ kamel/sidd/. It contains 31888 noisy/clean image pairs; we used the standard split (30 608 training and 1 280 validation images) after per-image