A glance report on segmentation methods for implementing thermal image

Thermography is a method of recognizing invisible heat radiation reflected by objects. Thermography produces thermal images that helps to identify intensity of color. The intensity of color helps to recognize meaningful object in the image, image segmentation plays a significant role to discover the region of interest in the image. The goal of segmentation is to decompose an image into different areas for further analysis and another is to perform a change of the representation of an image for faster analysis. On the basis of application, a single or a combination of segmentation techniques can be applied to solve the problem effectively. Segmentation is performed by mark off an object on an image using pixel - level or object - level properties of the object. These properties can be edges, texture, pixel intensity distinction inside the object, shape, size and orientation. There are many segmentation techniques are available that segments the thermal image. The segmentation techniques in particular are watershed segmentation, thresholding segmentation, clustering based segmentation and artifitial neural network. These segmentation technique uses algorithm namely global thresholding, watershed transform, k-medoids, k-means clustering, otsu thresholding, kapur thresholding. This paper involves the literature review on pioneering segmentation techniques and applicable algorithms used for segmentation of thermal images.