http://docplayer.net/16497156-Moving-object-tracking-of-vehicle-detection-a-concise-review.html
영역을 기반으로 물체를 추적한다. In these methods, the region of the moving objects are tracked and used for tracking the vehicles.
영역은 배경제거 기법등을 통해 추출 된다. These regions are segmented using the subtracting process between the input frame image and prior stored background image.
This model worked on series of traffic scenes recorded by a stable camera for automobiles monocular images and provided position and speed knowledge for each vehicle as long as it is visible.
The processing algorithms of this model represented by three levels:
차량의 윤곽선 정보를 이용한다. These methods depend on contours (the boundaries of vehicle)which are updated dynamically in successive images of vehicle in Tracking Vehicle Process [36].
영역 기반 방식보다 성능이 좋다. These methods provide more efficient descriptions of objects than Region-Based Methods and have been successfully applied to practice.
But objects occlusion and automatic initialization of tracking are difficult to handle and tracking precision is limited by a lack of precision in the location of the contour.
A vehicle anisotropic distance measurement achieved through the 3D geometric shape of vehicles.
A new 3D model-based vehicle detection and depiction framework is based on a probabilistic boundary feature grouping,which is used for vehicle detection and tracking process [37].
In this paper, the occlusion of vehicles detection process uses a 3D solid cuboid form with up to six vertices, and this cuboid is used to fit any different types and sizes of vehicle images by changing the vertices for a best fit.
Therefore, vehicle detection, segmentation and tracking can be achieved efficiently due to changes in the region proportion, prototype width and height with consideration to previous images.
The particular vehicles are detected, segmented and tracked in image sequence by assembling, bunching and approximating the 3D world coordinates of vehicle's feature points.
An iterative and distinguishable framework based on edge points as features is used in similarity process, these features represents a large region of set of features forms a strong depiction for object classes.
This proposed framework showed a good performance for vehicle classification in surveillance videos[38].
A linearity feature technique is a proposed line-based shade method which uses line groups to remove all undesirable shades and properly under takes the occlusion resulting from shades.
This technique is used to analyze color of image series of traffic supervision views [39].
Through the practical experiments, this system proven to work well under several weather situations,and it is insensitive to light variations.
This model-based system is used for real-time traffic supervision for continuous visual tracing and classification of vehicles for busy multi-lane highway scene[40].