The brain tumor detection is a very important vision application in the medical field. In this paper, an efficient brain tumor detection using the object detection and modified hough metric has been proposed. To enhance the tumor detection rate further we have integrated the proposed object based tumor detection with the Decision based alpha trimmed global mean. The proposed technique has the ability to produce effective results even in case of high density of the noise.   This approach has been tested on various images thus defining an efficient and robust technique for automated detection and segmentation of brain tumors.