Detection and Classification of Moving Object for Smart Vision Sensor
Mohd Razali, B.J. Adznan · 2006
Conventional surveillance system requires human power to monitor and thus not applicable for a long hour monitoring. An automated method is proposed here, an integration of a moving object detection and recognition. First, the moving object is detected and segmented by using selectiveness adaptive background subtraction technique followed by noise and shadow removal algorithm for removing disturbances. Then, standardize moment invariant is employed to extract the features for each moving blobs. To recognize these blobs, the calculated moment values are fed to a neural network module that is equipped with trained extracted moment values for human and vehicle silhouettes. The results of the experiments showed a satisfied performance with the proposed approach