The maintenance level of rural roads is relatively low, and the automated detection of pavement distress is easily affected by the shadows of rows of trees, weeds, soil, and distress object scale disparities; this makes it difficult to accurately evaluate the distress conditions of the pavement. To solve the above problems, this study specifically designed a target detection network called Crack Convolution (CrackYOLO) for pavement crack extraction on rural roads. CrackYOLO is based on an improved YOLOv5. The shadow created by rows of trees leads to the loss of crack features in the feature extraction and downsampling stages of the network; therefore, CrackConv and Adapt-weight Down Sample (ADSample) were introduced to strengthen the ability to locate and identify cracks. Due to disturbances such as soil and weeds, which cause the extraction of more redundant features, the Channel And Spatial mixed attention mechanism (CAS) was introduced to enhance crack weight. To address the issue of missed detections of fine cracks due to significant scale variations in crack objects in the same image, Multi Scale Convolution (MSConv) and Multi Scale Head (MSHead) were incorporated during the feature fusion and prediction inference stages of the network, thereby improving the multi-scale detection performance. In order to verify the effectiveness of the proposed method, the detection accuracy of CrackYOLO when used on the LNTU_RDD_NC dataset was determined to be 9.99%, 12.79%, and 4.61% higher than that of the current pavement crack detection models YOLO-LWNet, Faster R-CNN, and YOLOv7. At the same time, we compare the above model on public datasets of different scenarios, and the experimental results show that CrackYOLO has the same strong performance in urban roads and other scenarios.

Pavement distress detection is crucial in road health assessment and monitoring. However, there are still some challenges in extracting pavement cracks based on deep learning: such as insufficient segmentation, extraction errors and discontinuities. In this paper, we propose DARNet, a network for pavement distress extraction. A Distress Aware Attention Module (DAAM) is proposed to solve the problem of discontinuity in distress extraction due to inaccurate recovery of distress pixels during upsampling. Based on the characteristics of distress morphology, a Refinement Extraction Module (REM) is designed to effectively capture horizontal and vertical pavement damage features by fusing high-level and low-level features, which improves the accuracy of the model in extracting details of pavement damage information. Finally, a Weighted Cross-Entropy Loss function (WCEL) is introduced to assign weights according to the distance of the pixel point to the boundary of the damage, which solves the problem that the traditional cross entropy function treats each pixel point equally. We also propose a set of pavement distress datasets LNTU_RDD_GS, and the experimental results show that DARNet can reach 81.77% mIoU and 89.49% F score in the datasets in this paper, 80.63% mIoU and 88.35% F score in the four public datasets.

Traffic sign detection under adverse weather conditions remains challenging due to severe feature degradation caused by rain, fog, and snow, which significantly impairs the performance of existing detection systems. This study presents the CW-DETR (Complex Weather Detection Transformer), an end-to-end detection framework designed to address weather-induced feature deterioration in real-time applications. Building upon the RT-DETR, our approach integrates four key innovations: a multipath feature enhancement network (FPFENet) for preserving fine-grained textures, a Multiscale Edge Enhancement Module (MEEM) for combating boundary degradation, an adaptive dual-stream bidirectional feature pyramid network (ADBF-FPN) for cross-scale feature compensation, and a multiscale convolutional gating module (MCGM) for suppressing semantic–spatial confusion. Extensive experiments on the CCTSDB2021 dataset demonstrate that the CW-DETR achieves 69.0% $AP$ and 94.4% $A{P}_{50}$, outperforming state-of-the-art real-time detectors by 2.3–5.7 percentage points while maintaining computational efficiency (56.8 GFLOPs). A cross-dataset evaluation on TT100K, the TSRD, CNTSSS, and real-world snow conditions (LNTU-TSD) confirms the robust generalization capabilities of the proposed model. These results establish CW-DETR as an effective solution for all-weather traffic sign detection in intelligent transportation systems.

Crowd flow prediction plays a vital role in modern city management and public safety prewarning. However, the existing approaches related to this topic mostly focus on single sites or road segments, and indoor regional crowd flow prediction has yet to receive sufficient academic attention. Therefore, this paper proposes a novel prediction model, named the spatial–temporal attention-based crowd flow prediction network (STA-CFPNet), to forecast the indoor regional crowd flow volume. The model has four branches of temporal closeness, periodicity, tendency and external factors. Each branch of this model takes a convolutional neural network (CNN) as its principal component, which computes spatial correlations from near to distant areas by stacking multiple CNN layers. By incorporating the output of the four branches into the model’s fusion layer, it is possible to utilize ensemble learning to mine the temporal dependence implicit within the data. In order to improve both the convergence speed and prediction performance of the model, a building block based on spatial–temporal attention mechanisms was designed. Furthermore, a fully convolutional structure was applied to the external factors branch to provide globally shared external factors contexts for the research area. The empirical study demonstrates that STA-CFPNet outperforms other well-known crowd flow prediction methods in processing the experimental datasets.

Cracks are among the most common pavement distresses, and their timely detection is crucial for road maintenance. Existing methods struggle to completely capture elongated and irregular cracks, often resulting in fragmented detection outputs, which leads to the inaccurate assessment of crack length and affects the reliability of pavement condition evaluation. To address this issue, this paper proposes a dual-path crack segmentation network that integrates CNN and Transformers. The CNN branch incorporates a dynamic multi-branch convolution module to enhance the directional perception and structural modeling of elongated cracks. The Transformer branch employs a lightweight DCNv4 module to replace traditional self-attention mechanisms, effectively capturing long-range dependencies while reducing computational complexity. A multi-path fusion module is designed to achieve the collaborative enhancement of dual-path features, improving the semantic representation of continuous crack regions. Additionally, a combined loss function of BCE and Dice is adopted to alleviate the severe class imbalance between crack and background pixels, further improving the completeness of crack segmentation. Experiments on four datasets, including CFD, DeepCrack537, Gaps384, and Crack500, demonstrate that the proposed model outperforms all compared methods in terms of F-score and mIoU. Ablation studies further validate the effectiveness of the dual-path architecture and its key modules in improving performance. Furthermore, in field validation on real road scenarios, the pavement condition index (PCI) calculated based on the proposed method shows an average deviation of only 0.81 compared to manually interpreted ground truth, demonstrating the practical value of continuous crack detection for pavement maintenance assessment.