No abstract

A definite mention to all who have influenced, guided and supported me throughout this PhD. On all issues I gratefully thank my supervisors Dr Mark 'M otivator' Hodgetts (SIRA) and Dr Alistair Greig (UCL) for continued effort in working as a team to get the job done. Particular thanks to Mark for saving my life and taking me onboard If it wasn't for him this Thesis simply would not exist I must also thank Professor David Broome (UCL / GRL) for his cheery interest, Dr John Gilby (SIRA) for his strong visions, and special thanks to Jason Tisdall (UCL) and Duncan Wilson (UCL/SIRA) for continuous brainstorming, support, lively discussions and hard times-may we all be successful in This Game. A strong Thank-you to all in the Technology Centre, SIRA, and the Department of Mechanical Engineering, UCL, who have been a pleasure to work with and paid keen interest throughout. At SIRA I thank Bill Simmonds, Linda Huntingford, Bob Smith, Diane & Delia and, in the PR engine room, Ann Spencer for their continued support and efforts-it has not gone unnoticed At UCL I thank Charlotte Reisch, Margaret Harrison and in particular Linda Luck for making me so welcome.

Since the 1970s, there has been increasing interest in the use of Markov Random Fields (MRFs) as models to aid in the segmentation of noisy or degraded digital images. MRFs can make up for deficiencies in observed information by adding a-priori knowledge to the image interpretation process in the form of models of spatial interaction between neighbouring pixels. In data fusion problems, interaction might also be assumed between corresponding pixels in two different kinds of image of the same scene. Alternatively, temporal interaction might be assumed between corresponding pixels in consecutive frames of a video sequence. In object tracking or robotic navigation problems, a similar relationship may exist between pixels of an observed image and those of a predicted image, derived from models of the motion and scene. In all of these cases the MRF model can be extended to incorporate this additional knowledge. This paper explains the theory of Extended-Markov Random Field (E-MRF) segmentation techniques, surveys the research which has been crucial to their development and presents results from new work in this area with an application to robotic vision in conditions of extremely poor visibility.

No abstract

No abstract