Thermal image acquisition and segmentation of human foot

Abstract

Digital Infrared Thermal Imaging (DITI) or Infrared Thermography is a non-invasive, non-contact, harmless technique which measures and records the surface temperature of the skin under study as thermogram or thermal images. Segmentation of the region of interest from thermal images is a very difficult task especially with the foot because of the presentation of the ankle bones in the images and the temperature difference between the foot and background being small which interfere with the segmentation task. Our segmentation procedure is an attempt to work on these challenges and has shown promising results. In this paper, a semi-automatic segmentation algorithm is proposed and implemented in MATLAB to segment the foot from thermal images taken using Fluke TiX560 thermal imaging camera. Image Acquisition protocol plays a very important role in successful segmentation of thermal images. Images acquired using different setups at three different occasions are considered for the current study. Segmentation outcome of our method is analyzed using the three measures such as Jaccard index, False Positive Rate and False Negative Rate in comparison with the result of manual segmentation.