This research work presents an efficient approach of detecting unattended or stolen objects in live videos based on background subtraction and foreground analysis. The most common algorithm for performing background subtraction is the Gaussian Mixture model (GMM). An improved Multi-Gaussian Adaptive background model is employed for background subtraction to determine the static region. A simple split and merge method is used to detect the static region from which the static objects are identified. The time and presence of static objects, which may be either unattended or stolen, are informed by sending a mail and SMS to the security officials. Also, Haralick's texture operators are employed for images to identify objects under low contrast situations. The system is efficient to run in real time and produce good results
展开▼
