In the manufacturing process of industrial robots, the defect detection of raw materials includes two types of tasks, which makes the defect detection guarantee its accuracy. It also makes the defect detection task challenging in practical work. In analyzing the disadvantages of the existing defect detection task methods, such as low precision and low generalization ability, a detection method on the basis of attention mechanism and dilated convolution module is proposed. In order to effectively extract features, a two-stage detection framework is chosen by applying Resnet50 as the pre-training network of our model. With this foundation, the attention mechanism and dilated convolution are utilized. With the attention mechanism, the network can focus on the features of effective regions and suppress the invalid regions during detection. With dilated convolution, the receptive field of the model can be increased without increasing the calculation amount of the model. As a result, it can achieve a larger receptive field, which will obtain more dense data and improve the detection effect of small target defects. Finally, great experiments are conducted on the NEU-DET dataset. Compared with the baseline network, the proposed method in this paper achieves 81.79% mAP, which are better results.

Multi-Camera Multi-Vehicle Tracking (MCMVT) is a critical task in Intelligent Transportation Systems (ITS). Differently to in urban environments, challenges in highway tunnel MCMVT arise from the changing target scales as vehicles traverse the narrow tunnels, intense light exposure within the tunnels, high similarity in vehicle appearances, and overlapping camera fields of view, making highway MCMVT more challenging. This paper presents an MCMVT system tailored for highway tunnel roads incorporating road topology structures and the overlapping camera fields of view. The system integrates a Cascade Multi-Level Multi-Target Tracking strategy (CMLM), a trajectory refinement method (HTCF) based on road topology structures, and a spatio-temporal constraint module (HSTC) considering highway entry–exit flow in overlapping fields of view. The CMLM strategy exploits phased vehicle movements within the camera’s fields of view, addressing such challenges as those presented by fast-moving vehicles and appearance variations in long tunnels. The HTCF method filters static traffic signs in the tunnel, compensating for detector imperfections and mitigating the strong lighting effects caused by the tunnel lighting. The HSTC module incorporates spatio-temporal constraints designed for accurate inter-camera trajectory matching within overlapping fields of view. Experiments on the proposed Highway Surveillance Traffic (HST) dataset and CityFlow dataset validate the system’s effectiveness and robustness, achieving an IDF1 score of 81.20% for the HST dataset.

Scour hole that occurs downstream of the hydraulic structures threatens the safety and stability of the hydraulic structures. The scour around the structures is a complex and important hydraulic phenomenon; hence, it requires a data extensive research for the accurate estimation of scour depth. Although many analytical models are available for scour depth estimation, they suffer from huge limitations. In this research, the support vector regression (SVR) model and SVR ensemble with the metaheuristic algorithm of innovative gunner (SVR-AIG) models have been developed for accurate prediction of scour depth downstream of the ski-jump spillways. Field measurements including head and discharge intensity are used for developing the models. The performances of the models are compared using root mean square error (RMSE), mean average error (MAE), and correlation coefficient (CC) criteria and some statistical plots. The results showed that the hybrid SVR-AIG-based estimations (with CC = 0.987, 0.991, RMSE = 2.839, 1.987, and MAE = 2.247, 1.201) are more accurate than the SVR standalone model estimations (with CC = 0.942, 0.975, RMSE = 5.686, 4.040, and MAE = 4.114, 3.201) at the training and testing phases. This study is an important reference for analyzing the high capability of the AIG as an optimization tool in improving scour estimations of a standalone model. Also, this algorithm eliminates the trial-and-error procedure to optimize the internal parameters during the model development. Graphical abstract

With the increasing urbanization and the popularity of traffic video surveillance, and the rapid development of object detection algorithms, object detection of traffic events through video analysis has become possible. This paper proposes the design and implementation of a traffic event detection system based on video analysis. Firstly, the video stream is processed, mainly for video access and display, and the compression of neural networks achieves the network acceleration. Secondly, the structured information of vehicles is obtained by nighttime vehicle detection and joint detection and tracking. Finally, the car's driving behavior is analyzed through video calibration and video analysis. The system has been used online in many places in China and has achieved remarkable results.

The surface quality of steel directly affects its performance and durability in industrial production. However, traditional deep learning technology struggles to balance performance and efficiency in detecting intricate steel surface flaws. This paper presents an enhanced defect detection network (EDDN) for real-time steel surface defect detection. First, we design a large kernel feature enrichment module (LFEM) to fuse the feature layers with rich detail information, while using continuous large kernel convolution to capture multi-scale features. Notably, a dynamic selection mechanism that adaptively captures significant spatial features on the basis of global information. This design addresses the problem of tiny and complicated target detection failures and significantly increases detection accuracy. In addition, we propose a dynamic task interaction detection head (DT-Head) that balances the detection by enabling the dynamic interaction between the localization and classification tasks. This interaction mechanism allows the two tasks to collaborate better for a more accurate prediction. Finally, extensive experiments on NEU-DET and GC10-DET show that EDDN improves mAP@50 accuracy by 2.9% and 4.6%, achieving detection speeds of 79 and 72 frames per second (FPS), respectively, outperforming current mainstream algorithms in terms of accuracy and efficiency