a model for diagnosing breast cancerous tissue from thermal images using active contour and lyapunov exponent

background: the segmentation of cancerous areas in breast images is important for the early detection of disease. thermal imaging has advantages, such as being non-invasive, non-radiation, passive, quick, painless, inexpensive, and non-contact. imaging technique is the focus of this research. methods: the proposed model in this paper is a combination of surf and corners that are very resistant. obtained features are resistant to changes in rotation and revolution then with the help of active contours, this feature has been used for segmenting cancerous areas. results: comparing the obtained results from the proposed method and mammogram show that proposed method is accurate and appropriate. benign and malignance of segmented areas are detected by lyapunov exponent. values obtained include tp=91.31%, fn=8.69%, fp=7.26%. conclusion: the proposed method can classify those abnormally segmented areas of the breast, to the benign and malignant cancer.   keyword: breast cancer, surf and sift algorithm, active contour, thermography