The use of gradient information in visual and thermal infrared images for scene analysis

The research reported in this thesis deals with the analysis of visual and thermal infrared (TIR) images for deriving properties of objects appearing in an industrial scene. A physical interpretation and justification for the combining of information from visual and thermal infrared sensors are given. A method is proposed for the integration of information which can be derived from the two sensing modes based on the use of edge (dissimilarity) properties of the two types of images. The method involves the use of the Sobel operator to extract and classify edge points according to their strengths. A point-by-point comparison of edges from the visual and TIR images is made. Heuristics are employed to give a probabilistic interpretation to each point in the images based on the strengths of edge points detected in the two images at the point. This is shown to work on several types of scenes, as long as scene objects can be readily characterized by their thermal/visual signatures.