Multilevel image segmentation with adaptive image context based thresholding
Applied Soft Computing
Hi-index | 0.00 |
A novel approach for urinary image segmentation based on cellular neural network (CNN) was presented in this paper. Before the image segmentation, a preprocessing by stretching the difference between every pixel and the local gray mean value for eliminating the disequilibrium of illumination and enhancing the edges of objects is considered here. The experiment results with more than 100 clinical urinary images show that this approach provides more accurate objects detection compared with conventional threshold based ones.