The image of expressway asphalt pavement crack disease obtained by a three-dimensional line scan laser is easily affected by external factors such as uneven illumination distribution, environmental noise, occlusion shadow, and foreign bodies on the pavement. To locate and extract cracks accurately and efficiently, this article proposes a three-stage asphalt pavement crack location and segmentation method based on traditional digital image processing technology and deep learning methods. In the first stage of this method, the guided filtering and Retinex methods are used to preprocess the asphalt pavement crack image. The processed image removes redundant noise information and improves the brightness. At the information entropy level, it is 63% higher than the unpreprocessed image. In the second stage, the newly proposed YOLO-SAMT target detection model is used to locate the crack diseases in asphalt pavement. The model is 5.42 percentage points higher than the original YOLOv7 model on mAP@0.5, which enhances the recognition and location ability of crack diseases and reduces the calculation amount for the extraction of crack contour in the next stage. In the third stage, the improved k-means clustering algorithm is used to extract cracks. Compared with the traditional k-means clustering algorithm, this method improves the accuracy by 7.34 percentage points, the true rate by 6.57 percentage points, and the false positive rate by 18.32 percentage points to better extract the crack contour. To sum up, the method proposed in this article improves the quality of the pavement disease image, enhances the ability to identify and locate cracks, reduces the amount of calculation, improves the accuracy of crack contour extraction, and provides a new solution for highway crack inspection.

In the daily inspection task of the expressway, accuracy and speed are the two most important indexes to reflect the detection efficiency of nondeformation diseases of asphalt pavement. To achieve model compression, accelerated detection, and accurate identification under multiscale conditions, a lightweight algorithm (PDNet) based on improved YOLOv5 is proposed. The algorithm is improved based on the network structure of YOLOv5, and the improved network structure is called YOLO-W. Firstly, a novel cross-layer weighted cascade aggregation network (W-PAN) is proposed to replace the original YOLOv5 network. Secondly, more economical GhostC3 and ShuffleConv modules are designed to replace C3 and Conv modules in the original network model. In terms of parameter setting, CIoU is selected as the loss function of the model, and the K-Means ++ algorithm is used for anchor box clustering. Before the model training, the confrontation generation network (GAN) and Poisson migration fusion algorithm (Poisson) are used for data enhancement and the negative sample training (NST) method is used to improve the robustness of the model. Finally, Softer-NMS is used to remove the prediction box in the prediction stage. Seven common asphalt pavement disease data sets (FAFU-PD) are constructed at the same time. Compared with the original YOLOv5 algorithm, PDNet improves the scores of FAFU-PD data sets on F1-score by 10 percentage points and FPS by 77.5%.

Remote sensing (RS) images often suffer from spatial resolution degradation and edge blurring due to sensor limitations and complex degradation factors such as atmospheric turbulence and motion. These issues negatively impact subsequent tasks like classification and detection. Most existing super-resolution (SR) methods rely on synthetic low-resolution(LR) data generated by bicubic interpolation, which fails to reflect real degradation processes, resulting in limited performance in practical applications. Furthermore, many methods insufficiently exploit high-frequency details and lack the ability to effectively model both global context and local structures. To address these challenges, this paper proposes an edge-enhanced SR framework for remote sensing images. A more realistic degradation model is constructed by estimating blur kernels from real LR images. The Edge-Enhanced and Structure-Aware Generative Adversarial Network (EESAGAN) is designed, incorporating the edge prior enhancement module (EPE), the edge-aware local attention module(ELAM), and the dynamic feature space attention module(DFSA). The edge-guided loss is also introduced to enhance image sharpness and perceptual consistency. Experimental results demonstrate that the proposed method achieves superior performance in PSNR, SSIM, and ERGAS compared to existing SR approaches.

The majority of existing single remote sensing image super‐resolution models rely on bicubic downsampling to generate low‐resolution (LR) and high‐resolution image pairs. This approach overlooks the error‐affecting factors in the space‐based imaging chain, thereby limiting the models’ applicability to real‐world remote sensing images (RSIs). This issue is further exacerbated in large‐scale super‐resolution tasks (e.g., ×8 or ×9), where increased scaling factors reduce the amount of information available in LR images, making accurate RSI reconstruction challenging. To address these challenges, we propose a novel method, termed SRDDSAGAN, which is based on a real‐world degradation model (RWDM) and a deep generative adversarial network (SRDSAGAN). The RWDM integrates blur kernels and imaging noise to simulate real‐world imaging conditions, thereby generating more realistic training and testing datasets. The SRDSAGAN incorporates dense sampling and residual learning, enhanced by local and global attention mechanisms, to capture salient features and dependencies across spatial and channel dimensions. This approach significantly improves the accuracy of super‐resolution reconstruction for large‐scale RSIs. Additionally, the SRDSAGAN employs a deep residual attention network to learn gradient priors, which helps to preserve image edge smoothness and suppress visual artefacts. Multiple validation experiments demonstrate that the proposed SRDDSAGAN model outperforms state‐of‐the‐art methods.