The paper aims to develop an object detection method combining classification and cluster. Object detection is a challenging, yet important vision task. It is a critical part in many applications such as image search, image auto-annotation and scene understanding. Our proposed method work is dividing in two part.1) image classification and 2) clustering method. In this classification part LTP (local ternary pattern) features used to match the dataset image. This LTB features extracted from our input de-blurred image. Finally object is classified and then object is detected using clustering methods. As far as the robustness and effectiveness are concerned, our method is better than the existing image segmentation algorithms.