Object Detection and Counting Using Unsupervised Method

Differentiating and finding similar patterns of images from a video frame source emerges as the basic elemental task in image processing. In this paper, various algorithms on object identification, shape identification, color identification, object counting are analyzed and compared to obtain new unsupervised method. These methods find applications in the field of defense, security check, healthcare and Traffic monitoring. Various challenging features and applications of object Detection, identification and counting methods are elaborated with new algorithm with least set of dataset. In addition, the different steps involved in object detection and results for several steps are discussed. Moreover, the proposed algorithmic method is able to process the unstructured and structured images in multiple visual concepts. Further, these algorithms can be applied in wide area and crowded scenes with high precision identification.