Automated detection of eye and cornea on infrared thermogram using snake and target tracing function coupled withgenetic algorithm

Infrared imaging is widely used nowadays in health care field with more focus on detection of breast cancer and eye abnormalities. An algorithm is presented to correctly localize the eye and cornea in Infrared (IR) thermogram using gradient vector flow (GVF) snake coupled with target tracing function. The tat-get tracing function is used to overcome any human intervention on initialization, and helps to localize the eye accurately. The global minimums on that function are searched by genetic algorithm. From the final resultant snake points, corneal centre and radius are estimated. In this work, we have used 125 IR normal images and the proposed algorithm can localize the eye and cornea correctly in 90% of the IR thermograms.