Unsupervised Texture Image Segmentation Based on Gabor Wavelet and Multi-PCNN

This paper is a research on unsupervised texture segmentation technique using Gabor filters and multi-PCNN. The partitioning method based on Gabor filters gets the good segmentation result, but causes the huge data. PCNN is a parallel way and can be easily realized with hardware, especially VLSI, which makes PCNN process image fast. We present an algorithm which uses Gabor filters to extract image texture character which has been inputted into PCNNs to segment the image. This method can get good segmentation result and improve the algorithm´s processing speed.