Urinary Sediment Images Segmentation Based on Efficient Gabor flters

Urinary sediments are very important to help diagnose diseases such as kidney inflammation, urethra inflammation, bladder inflammation and so on. Gabor filter is a widely used feature extraction method, especially in image texture analysis. The selection of optimal filter parameters is usually problematic and unclear. This paper present a improved and robust method, algorithm based on efficient Gabor filters in combination with simulated annealing and K-means clustering, for urinary sediment segmentation. And the method presented are consist of two steps: first, using multi-channel Gabor-filters to extract the features of the urinary sediment from the microimages; second, segment the urinary sediment image by simulated annealing and K-means clustering. The experiment results show that the method can segment the urinary sediment images effectively and precisely.