抄録
In this paper, the authors propose a threshold setting method for binary image. This method consists of two steps which are different threshold calculating algorithm. First step is an improved model of Convergent Floating Thresholding Method, by switching between two control values for updating a tentative threshold. By the step, thresholding of a boundary area of gray level change is performed. Second step is composed of Discriminant Analysis Method and is applied to area of flat on gray level. Feature of the proposed method is to compare gray level with two final tentative thresholds which are obtained using two coefficients as control value. According to the method, it is possible to decrease a thresholding error which occurs in case of low contrast and brightness change. Also a pattern in likeness gray level area can be clarified. The proposed method and the conventional methods were tested, using general scene images of outdoor. The result showed that the proposed method can output good quality binary images relatively and is useful for thresholding of the scene image.
