Non-Anchor-Based Vehicle Detection for Traffic Surveillance Using Bounding Ellipses
نویسندگان
چکیده
Cameras for traffic surveillance are usually pole-mounted and produce images that reflect a birds-eye view. Vehicles in such images, general, assume an ellipse form. A bounding box the vehicles includes large empty space when vehicle orientation is not parallel to edges of box. To circumvent this problem, present study applied ellipses non-anchor-based, single-shot detection model (CenterNet). Since does depend on anchor boxes, non-max suppression (NMS) requires computing intersection over union (IOU) between predicted boxes unnecessary inference. The SpotNet extends CenterNet by adding segmentation head was also tested with ellipses. Two other anchor-based, models (YOLO4 SSD) were chosen as references comparison. performance compared based local dataset doubly annotated As result, two exceeded reference boxes. When backbone pretrained open (UA-DETRAC), further enhanced. Several data augmentation schemes improved proposed models. best mAP score exceeds 0.95 augmenting heatmaps
منابع مشابه
BoxCars: Improving Vehicle Fine-Grained Recognition using 3D Bounding Boxes in Traffic Surveillance
In this paper, we focus on fine-grained recognition of vehicles mainly in traffic surveillance applications. We propose an approach orthogonal to recent advancement in fine-grained recognition (automatic part discovery, bilinear pooling). Also, in contrast to other methods focused on fine-grained recognition of vehicles, we do not limit ourselves to frontal/rear viewpoint but allow the vehicles...
متن کاملTraffic surveillance camera calibration by 3D model bounding box alignment for accurate vehicle speed measurement
In this paper, we focus on fully automatic traffic surveillance camera calibration which we use for speed measurement of passing vehicles. We improve over a recent state-ofthe-art camera calibration method for traffic surveillance based on two detected vanishing points. More importantly, we propose a novel automatic scene scale inference based on matching bounding boxes of rendered 3D models of...
متن کاملVehicle Velocity Estimation for Traffic Surveillance System
This paper describes an algorithm to estimate realtime vehicle velocity using image processing technique from the known camera calibration parameters. The presented algorithm involves several main steps. First, the moving object is extracted by utilizing frame differencing technique. Second, the object tracking method is applied and the speed is estimated based on the displacement of the object...
متن کاملShadow-Based Vehicle Detection in Urban Traffic
Vehicle detection is a fundamental task in Forward Collision Avoiding Systems (FACS). Generally, vision-based vehicle detection methods consist of two stages: hypotheses generation and hypotheses verification. In this paper, we focus on the former, presenting a feature-based method for on-road vehicle detection in urban traffic. Hypotheses for vehicle candidates are generated according to the s...
متن کاملTraffic Surveillance: A Review of Vision Based Vehicle Detection, Recognition and Tracking
Video-based analysis of traffic surveillance is an active area of research, which has a wide variety of applications in intelligent transport systems (ITSs). In particular, urban environments are more challenging than highways due to camera placement, background clutter, and vehicle pose or orientation variations. This paper provide a comprehensive review of the state-of-the-art video processin...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Access
سال: 2021
ISSN: ['2169-3536']
DOI: https://doi.org/10.1109/access.2021.3109258