A fast multilevel thresholding method based on lowpass and highpass filtering

Abstract A new method of thresholding a gray-scale image into a desired number of classes is proposed in this paper. The proposed method first evaluates the number of peaks in the histogram, then finds a suitable lowpass or highpass filter such that the number of peaks in the filtered histogram is equal to the desired number of classes. Then, valleys in the filtered histogram are used as thresholds. The proposed method is computationally fast and corresponds to human intuition. In the experiments, we compare the performance of our proposed method with other thresholding methods and the method is shown effective and efficient in multilevel thresholding.