Integrating spatio-temporal information in image sequence analysis to enforce consistency of interpretation

We present a technique of image-sequence analysis which ensures consistency of interpretation in terms of the classification of segmented regions, bot h spatially within an image and temporally through a sequence of forward-lo oking infrared images taken from a low-flying aircraft. The technique is ba sed on classical relaxation labeling techniques but is novel in that it tre ats segmented regions as single entities that have both spatial and tempora l neighbors. This enables classification probabilities both to be propagate d through the sequence and to be used in the minimization of the Kullback e ntropy of information. Sample results are shown from the application of the method to a 12-s sequence of infrared images that have been segmented usin g co-occurrence-based techniques. For completeness, the paper also describe s the whole analysis algorithm including segmentation, texture analysis usi ng discrete Hermite functions, and classification using neural network tech niques, (C) British Crown copyright 1998.