A generalized method of region of interest evaluation and its application in cognitive sharing through co-operative action

Region of Interest (ROI) selection remains very crucial problem for its notion of subjectivity. Many methods are available for its selection. However, there is no generalized method of selection and evaluation. Evaluation is particularly useful when different ROIs need to be shared between agents such as robots for co-operative task performance. This type of sharing is called cognitive sharing. This paper is proposing a generalized method of ROI Evaluation. Moreover, ROI evaluation by individual robots is ranked to select Best ROI for solving a particular task. In order to select Best ROI, ‘Semantic stability’ of object relations in image regions is defined. This paper presents cooperative actions by robots to evaluate the proposed algorithms.