Scale-invariant object identification for jointly addressing issues in moving oject and background

摘要

The area of video process pertaining to analysis of the real-time object behaviour is increasing in its pace. However, such process is also accompanied by various critical challenges e.g. variation in illumination condition, complete or partial occlusion, and uncertain mobility of the object. Although, in the last decades there has been archives of research papers discussing about the object detection, but there are only few literatures pertaining to the precise moving object detection with moving background. Therefore, the core objective of this paper is to present a technique that efficiently performs moving object detection for the moving background using enhanced version of Scale-Invariant Feature Transform. The outcome of the study was compared with the most significant work to find it outperforms the existing system with respect to processing time and thereby it is quite found suitable for real-time processing of object detection.