Hough transform (HT) is a well-established method for curve detection and recognition due to its robustness and processing capability. Being the core principle, voting principle needs to find the max voting rate, which makes it impossible to detect many targets from single image synchronously. In this paper, an improved Hough transform algorithm which combines with clustering is presented. The algorithm can search many targets in the image once when the number of targets is known. The experiments on iris location show that the proposed approach can detect the needed targets in the image in reality.