Feature and color analysis of images of the cervix

Cervical cancer is the second most common type of cancer that prevails among women. The death rate due to cervical cancer in developing countries is much more than developed countries due the lack of trained personnel. This thesis discusses techniques investigated to create a fully automated system that aids in detecting cancer. Image matching techniques helps in pairing similar images in the database. Computer-based tools are developed that process the images for enhancement due to conditions such as lighting, and which extract and analyse [sic] key features of the cervix. A very efficient algorithm for segmentation of acetowhite regions is developed --Abstract, page iii