An important research area in computer vision is object detection. Object detection means detecting objects belonging to a specific justify; MARGIN: 0cm 0cm 0pt; unicode-bidi: embed; DIRECTION: ltr" dir=ltr The second part of the thesis introduces a method for finding corresponding features between model and hypothesis, resulted from Hough transform. Hough transform models the object's structure by considering each feature location with respect to its center. The shortcoming of this method is that it considers the location of each feature independent of others and ignores relative location of features. To overcome this problem, we formulate the feature correspondence problem between model and hypothesis as a Longest Common Subsequence (LCS) problem. Model and hypothesis images are represented as feature strings and longest common substring between them is calculated. So, by considering features order, corresponding features between model and hypothesis are obtained. In the final step, similarity of model and the corresponding features is calculated using shape context. We applied our method on two subsets of standard ETHZ shape dataset. The achieved results show that the proposed method improves Hough transform performance considerably and has comparable or better results in comparison with previous methods. Keywords: Contour-based object detection, In-plane rotation invariance, 4D hough space, Hypothesis verification, Longest Common Subsequence
