IEEE Transactions on Pattern Analysis and Machine Intelligence
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Image sequence processing using spatiotemporal segmentation
IEEE Transactions on Circuits and Systems for Video Technology
| Hi-index | 0.00 |
To support the content-based functionalities in the new video coding standard MPEG-4 and MPEG-7, each frame of a video sequence must first be segmented into video object planes (VOPs), each of which represents a meaningful moving object. However, segmenting a video sequence into VOPs remains a difficult and unresolved problem. Accordingly, this paper presents a genetic algorithm (GA) for unsupervised video segmentation. The method is specifically designed to enhance the computational efficiency and the quality of segmentation results than the standard genetic algorithms. In the proposed method, the segmentation is performed by chromosomes, each of which is allocated to a pixel and independently evolved using a distributed genetic algorithm (DGA). For effective search space exploration, except the first frame in the sequence, the chromosomes are started with the segmentation results of the previous frame. Then, only unstable chromosomes, corresponding to the moving objects parts, are evolved by crossover and mutation. The advantages of the proposed method include the fast convergence speed by eliminating the redundant computations between the successive frames. The advantages have been confirmed with experiments where the proposed method was successfully applied to the synthetic and natural video sequences.