Vehicle detection in severe weather based on pseudo-visual search and HOG–LBP feature fusion

Abstract

Vehicle detection in severe weather has always been a difficult task in the environmental perception of intelligent vehicles. This paper proposes a vehicle detection method based on pseudo-visual search and the histogram of oriented gradients (HOG)–local binary pattern (LBP) feature fusion. Using radar detection information, this method can directly extract the region of interest (ROI) of vehicles from infrared images by imitating human vision. Unlike traditional methods, the pseudo-visual search mechanism is independent of complex image processing and environmental interferences, thereby significantly improving the speed and accuracy of ROI extraction. More notably, the ROI extraction process based on pseudo-visual search can reduce image processing by 40%–80%, with an ROI extraction time of only 4 ms, which is far lower than the traditional algorithms. In addition, we used the HOG–LBP fusion feature to train the vehicle classifier, which improves the extraction ability of local and global features of vehicles. The HOG–LBP fusion feature can improve vehicle detection accuracy by 6%–9%, compared to a single feature. Experimental results show that the accuracy of vehicle detection is 92.7%, and the detection speed is 31 fps, which validates the feasibility of the proposed method and effectively improve the vehicle detection performance in severe weather

Deep Learning Techniques for Vehicle Detection and Classification from Images/Videos: A Survey

“…The cascade classifier (Haar–LBP–HOG feature) [ 18 ] is detects vehicles with bounding boxes. In addition to the previously mentioned features and classifiers for vehicle detection and classification problems, statistical architectures, based on horizontal and vertical edge features, were proposed for vehicle detection [ 19 ], side-view car detection [ 20 ], online vehicle detection [ 21 ], and vehicle detection in severe weather using HOG–LBP fusion [ 22 ].…”

“…At the end of the MobileNet algorithm, the framework of depth-wise separable convolution is used to generate a classifier for each identified vehicle. Wang et al [ 22 ] proposed an R-FCN algorithm equipped with deformable convolution and RoI pooling for vehicle detection. It has a better detection time and more precision.…”

Real-time vehicle target detection in inclement weather conditions based on YOLOv4

“…Ogunrinde and Bernadin ( 2021 ) used CycleGAN combined with YOLOv3 for the KITTI dataset to improve the detection efficiency of moderate haze images. Wang et al ( 2022 ) proposed a vehicle detection method based on pseudo-visual search and the histogram of oriented gradients (HOG)-local binary pattern feature fusion, which achieved an accuracy of 92.7% and a detection speed of 31 fps. Guo et al ( 2022 ) proposed a domain-adaptive road vehicle target detection method based on an improved CycleGAN network and YOLOv4 to improve the vehicle detection performance and the generalization ability of the model under low-visibility weather conditions.…”

Local Texture Representation for Timber Defect Recognition based on Variation of LBP

“…The recent literature proposes several techniques related to LBP. Recent work used pseudo-visual search and the histogram of oriented gradients (HOG)-LBP feature fusion for intelligent vehicle detection in severe weather [21]. In addition, to completing local quartet patterns, a new version of LBP plays a crucial role in fabric textile quality-control systems [22].…”