Integrated color and texture tools for colposcopic image segmentation

Colposcopic images are characterized by color, texture and relief information. Thus, their automatic analysis is difficult. However, the diagnoses of experts on much debated images are often different, because of the very high specialization required. There is also a real need for a computer-aided diagnosis tool. We present an integrated analysis tool for helping gynecologists to build their colposcopic diagnosis. Based on interactive conception, it allows one to choose the different color components (from 7 color representation spaces) and texture attributes (cooccurrence and run length parameters, fractal dimension, Laws filtering, autoregressive parameters) which are to be taken into account in order to perform segmentation. Moreover, specific preprocessing methods and different segmentation methods are available such as principal component analysis and multidimensional histogram analysis. Results show that texture indices improve the segmentation process