Robust Multiple Car Tracking with Occlusion Reasoning
Robust Multiple Car Tracking with Occlusion Reasoning
Insight into Images: Principles and Practice for Segmentation, Registration, and Image Analysis
Insight into Images: Principles and Practice for Segmentation, Registration, and Image Analysis
Robust background subtraction with foreground validation for urban traffic video
EURASIP Journal on Applied Signal Processing
| Hi-index | 0.00 |
Recent work on wavelets applied to images or a video sequence has been exploited for extracting robust illumination invariant features. The paper presents robust background subtraction algorithm to segment motion based video scene in embedded platforms. Every machine or computer vision algorithm to be useful should be able to separate the different background and foreground information (e.g. objects) in the given scene. Therefore, it is essential to the success of any real time algorithm, the scene segmentation invariant to lighting conditions. We designed two main algorithms; Six frames (6-Frames) and Time Interval with Memory (TIME) to segment the video scene robustly based on motion detection in embedded platforms. The former uses the first six frames and the latter samples the frames at regular intervals of time with memory to generate a background reference frame. Our algorithms used bandpass video scene filtering with wavelets for extracting illumination invariant scene features and then combine them efficiently into the background reference frame. Hardware efficient image stabilization capability was added to remove the unwanted motion due to camera movement. The algorithms were tested using three moving bee videos sequences; static background, moving shadow and destabilized. Performance of algorithms was evaluated on the basis of number of frames in which the moving target was detected for each video sequence.