This paper presents a novel MRI super-resolution analysis model, MRISR. Through the utilization of generative adversarial networks for the estimation of degradation kernels and the injection of noise, we have constructed a comprehensive dataset of high-quality paired high- and low-resolution MRI images. The MRISR model seamlessly integrates VMamba and Transformer technologies, demonstrating superior performance across various no-reference image quality assessment metrics compared with existing methodologies. It effectively reconstructs high-resolution MRI images while meticulously preserving intricate texture details, achieving a fourfold enhancement in resolution. This research endeavor represents a significant advancement in the field of MRI super-resolution analysis, contributing a cost-effective solution for rapid MRI technology that holds immense promise for widespread adoption in clinical diagnostic applications.
Low-resolution magnetic tile defect recognition suffers from the loss of texture, edge and fine-grained defect information, which weakens discriminative features and reduces recognition accuracy in industrial inspection. To improve defect recognition under low-resolution imaging conditions, this paper proposes a task-aware low-resolution magnetic tile defect image recognition method based on AMCANet. AMCANet is used as the basic super-resolution reconstruction network, and a fixed high-resolution classifier is introduced during training to construct feature consistency loss and classification task loss. Thus, the reconstructed images are constrained to approach high-resolution images at the pixel level while preserving defect-discriminative features. Experiments on the Magnetic Tile Surface Defect Dataset show that the proposed TA-AMCANet achieves an Accuracy of 0.9065 and a Macro-F1 of 0.8204 under the fixed-classifier evaluation setting, improving on the original AMCANet by 0.3238 and 0.3726, respectively. Further ablation, loss-weight sensitivity, AMCA Block configuration, cross-classifier, degradation robustness, external-dataset generalization, Grad-CAM visualization, and computational complexity experiments verify the effectiveness of the proposed task-aware strategy. The results further indicate that higher pixel-level reconstruction metrics do not necessarily correspond to better downstream defect recognition performance, while TA-AMCANet improves recognition adaptability without introducing additional parameters or inference overhead. The method provides an effective solution for low-resolution industrial defect recognition.
Computed Tomography is a crucial diagnostic tool in medicine but exposes patients to harmful radiation. Low-dose CT reduces radiation but leads to noisy images. Super-resolution technology can reconstruct high-resolution images from low-resolution scans, enhancing image quality while reducing radiation. This study proposes a novel deep learning model, CACTSR, which integrates VMamba and Transformer technologies with Mixed Attention Blocks and Cross Attention Blocks to enhance feature utilization and facilitate cross-window information interaction. Experimental results on the QING LUNG dataset demonstrate that CACTSR surpasses existing methods in terms of image quality metrics, generating images with crisp edges and abundant details. This innovative approach effectively mitigates block artifacts and enhances image quality, providing a powerful solution for reducing radiation doses in clinical CT imaging while maintaining diagnostic accuracy.