Adaptive Object Analysis for Online System Using Illumination Classification

This research proposes dynamic changing environments, and context identification and categorization models. In this research, we have employed classifier combination to promote the classification efficiency and online evolutionary approach for creating intelligence. Our proposed system uses K-means to carry out context awareness, modeling, and identification for the context environment and online evolutionary algorithm stores its experiences in terms of artificial chromosomes to make the context knowledge for real-time adaptation. The novelty of this proposal is that the proposed system can be able of operating in dynamically changing environment with varying illumination, its classifier system can be able to downgrade the search space, and the evolutionary computing guides the system to be adaptive to varying environments.