Detection and Identification of Natural and Man-Made Objects in Thermal Images

Abstract

This project is about a system that uses a thermal image as an input in order to detect and distinguish man-made objects, humans, animals, military targets or any other object of prominent heat profile. The object's classification should be implemented based on the shape analysis, and how this shape is evolving over time. Many techniques were provided in the literature for the detection, description and identification of shapes. Therefore, it is very important to verify those techniques and select the applicable one for thermal shape processing. Moreover, there is a need to study pre-processing schemes of thermal images to identify the most useful algorithm for shape detection. The future work can be done on some specific applications, where the images of, for example, animals will be captured and the remaining images will be ignored (such as in visual surveillance of wild life at night and recognition of different types of animals). This thesis presents the suggested solution and the way it is implemented. Indexing Terms: Thermal images, object detection, shape descriptor, object classification, image binarization.

Date of Award	Jun 2018
Original language	American English
Supervisor	Andrzej Sluzek (Supervisor)