A semantic understanding of road traffic can help people understand road traffic flow situations and emergencies more accurately and provide a more accurate basis for anomaly detection and traffic prediction. At present, the overview of computer vision in traffic mainly focuses on the static detection of vehicles and pedestrians. There are few in-depth studies on the semantic understanding of road traffic using visual methods. This paper aims to review recent approaches to the semantic understanding of road traffic using vision sensors to bridge this gap. First, this paper classifies all kinds of traffic monitoring analysis methods from the two perspectives of macro traffic flow and micro road behavior. Next, the techniques for each class of methods are reviewed and discussed in detail. Finally, we analyze the existing traffic monitoring challenges and corresponding solutions.

The transportation detection of long-distance small objects has low accuracy. In this work, we propose DMF, which is based on disparity depths. We map different disparity regions to 2D candidate regions according to the distance to solve the small-object detection problem. This method clusters disparity maps of different depths. The projected image is extracted with image features in the mapping region. On the one hand, it uses a multicluster method to unsample 2D mapping regions. On the other hand, the feature fusion of different scales is performed on each cluster region. The experimental results on two datasets show that DMF can improve the detection accuracy of small objects.

The existence of specular highlights hinders high-level computer algorithms. In this paper, we propose a novel approach to remove specular highlights from a single grayscale image by regarding the problem as an image-to-image translation task between the highlight domain and the diffuse domain. We solve this problem by using the generative adversarial network framework, where a generator removes highlights and discriminator judges whether outputs of the generator are clear and highlight-free. Specular highlight removal is intractable as we should remove specular highlights while keeping as many details as possible. Considering the similarity between the highlight image and diffuse image, we adopt an attention-based submodule that generates a mask image, which we call the highlight intensity mask, to locate pixels that contain specular highlights and help the skip-connected autoencoder to remove highlights. A pixel discriminator and Structural Similarity loss are utilized to ensure that more details can be retained in the output images. For training and testing models, we build a grayscale highlight images dataset. It consists of more than a thousand sets of grayscale highlight images with ground truth. Finally, quantitative and qualitative evaluations demonstrate the effectiveness of our method than other contrast generative adversarial network methods.

Traffic flow prediction methods commonly rely on historical traffic data, such as traffic volume and speed, but may not be suitable for high-capacity expressways or during peak traffic hours. Furthermore, downstream flow can have significant impacts on traffic flow. To address these challenges, our study proposes a novel traffic flow prediction model, V-STF, which integrates visual methods to quantify macroscopic traffic flow indicators, as well as density features in temporal and flow feedback in spatio features. The contribution of our proposed model lies in its ability to improve prediction accuracy during non-periodic peak hours, by taking into account the impact of congested road conditions on traffic flow. Our experiments using the STREETS dataset demonstrate that V-STF outperforms state-of-the-art methods, especially in predicting sudden changes in traffic flow, resulting in more accurate predictions.

In an industrial environment, measuring and reconstructing metal objects using computer vision methods can be affected by surface highlight reflections, leading to inaccurate results. In this article, we propose a novel network with broad applicability for removing highlight reflections based on dynamic highlight masks, which is suitable for industrial metal highlight images where a baseline image with no highlight reflections is not available. First, we use a pretrained model to learn the highlight features of the metal surface, and then construct an adaptive dynamic highlight soft mask that allows texture features in highlight regions to be preserved while removing highlights. Second, we adjust the shape and shift of convolution operations using an adaptive highlight mask to better match the metal surface structure and avoid unnatural transitions at the edges of highlights in partially convolutional networks. We conducted quantitative evaluations on synthetic and real-world highlight datasets to demonstrate the effectiveness of our method. Compared with the state-of-the-art methods, our method shows improvements of over 0.4 in terms of correlation and Bhattacharyya distance of the histogram curves for synthesized highlight datasets. In terms of nonhighlight region invariance of real industrial highlight datasets, our method shows improvements of over 8, 0.1, and 23 points in terms of peak signal-to-noise ratio, structural similarity, and mean squared error, respectively.