A method for extraction of arbitrary figure using one-dimensional histogram

An extraction of a specific figure in image has basic problems in intelligent image sensing. The generalized Hough transform (GHT) is the representative method to extract arbitrary figures which are rotated and enlarged or reduced. Many the improvement models were also proposed. However, for extraction of arbitrary figures, it takes much processing time and needs much memory space. In addition, it is impossible to apply the GHT to figures including branches. For an improvement of the problems, a new method to extract arbitrary figure using one-dimensional histogram is proposed in this study. The method utilizes the polytope method which is one of minimization algorithms. For the extraction of figures, one-dimensional histogram is used. The histogram has two characteristics. (1) The distribution of histogram changes if the parameters representing figure changes. (2) The best parameters are gotten, if the value of most frequency of histogram becomes maximum. Therefore, by using the polytope method, the best parameters are searched so that the maximum value of most frequency can be maximum. In comparison with conventional method, it is understood that memory space is very small, processing time is very short and figures including branches can be extracted. In addition, this method is effective for an extraction of arbitrary figure with different aspect ratio.