Visual chasing with a novel data association with a track survival prospect by associating online recognitions with the corresponding tracks beneath partial occlusions. Experimental results using interesting public data sets show the obvious performance enhancement of the proposed system, equated with other state-of-the-art tracking systems. A hierarchical association outline is developed to produce longer tracklets at each level regularly. We propose an connected tracking system, which can forcefully track many objects even in difficult scenes but also be appropriate for online tracking solicitations. The track administration part terminates tracks with low survival probabilities and links the terminated tracks with other tracks or findings fit in to the same objects so as to link them. It denotes that characteristics of the tracklets can be transformed by linking results at each repetition. The online schemes are likely to produce split trajectories under occlusions when detections of occluded objects are not accessible or inaccurate. Object detection and tracking are important and challenging tasks in many computer vision applications such as surveillance, vehicle navigation and autonomous robot navigation. Object detection involves locating objects in the frame of a video sequence. Every tracking method requires an object detection mechanism either in every frame or when the object first appears in the video.