Transformer-based vehicle detection for surveillance images

Abstract

. Dense vehicle detection in rush hours is important for intelligent transportation systems. Most existing object detection methods can work well in off-peak vehicle detection for surveillance images. However, they may fail in dense vehicle detection in rush hours due to severe overlapping. To address this problem, a dense vehicle detection network is proposed by embedding the deformable channel-wise column transformer (DCCT) into the current you only look once (YOLO)-v5l network with a novel asymmetric focal loss (AF loss). The proposed DCCT fully extracts the column-wise occlusion information of vehicles in the images and guides the network to pay more attention to the visible area of partially occluded vehicles to improve the detection and positioning accuracy of weak feature targets. The proposed AF loss is used to balance the performance between easy and hard targets and address class imbalance. Extensive results demonstrate that the proposed network can accurately detect on-road densely located vehicles, even the minority classes in real time. Compared with the baseline YOLO-v5l, the mean average precision is improved by 3.93%, and it achieves comparable results with the existing state-of-the-art methods on the UA_Detrac dataset.

Lane Detection Methods Survey for Automatic Driving

“…As for the model-based detection method, the models' assumptions are mainly linear models and curve models. Hough transformation is generally used for straight lanes to establish a linear model for detection and the least square method for fitting [18][19]. For curving lanes, the curve model provides a broader range of feature extraction results in the feature extraction stage, which contains more noise, so the RANSAC and luminance curve (such as three cubed Bezier) fitting construction methods are used to solve the problem.…”

“…Hough transformation based on lane edge pixel is one of the most widely used lane detection algorithms in the past, it cannot detect curved lanes [9][10]. The curve model fitting method is based on the least square method, and RANSAC can detect straight lanes and curved lanes [18][19]. RANSAC fits the lane model through a recursive test model fitting score to seek the optimal model parameters, so RANSAC has a solid ability to deal with abnormal features.…”

A Lightweight Vehicle Detection Method Fusing GSConv and Coordinate Attention Mechanism

“…Machine vision technology has a powerful video data processing capability, from which it can extract key information, such as vehicle color, model, brand, and license plate number [ 4 , 5 , 6 ]. This information enables the transportation department to grasp the road conditions in real time, e.g., the supervisory department can use this information to accurately identify various sorts of motor vehicles on the road, thus enhancing the monitoring of dangerous vehicles.…”

Learned image compression via multiscale prior for machine recognition

“…Many existing image compression methods 1 – 6 pursue pixel fidelity at low bit rates to produce pleasing visual images. With the development of emerging scene applications (such as video processing and analysis 7 – 9 ), machines become end consumers like humans 10 . However, low bit-rate image compression based on maintaining pixel fidelity can negatively impact machine recognition accuracy 11 .…”