Image Processing and Analysis

Visual Information Processing studies elements of optics/imaging, signal processing, artificial intelligence, machine learning, and mathematics and is intricately weaved into other fields such as robotics, visual computing, information retrieval and human-computer-interaction. We are researching and developing methods and tools for a wide range of real world applications, for example: 

• Next generation digital imaging technologies - e.g. high dynamic range imaging
• Managing large digital image and video repositories - content-based image indexing and retrieval

Research is also underway into a range of problems in the automatic analysis of individual and sequences of images. Attention is focussed on identifying and recovering the motion of objects of interest. Recent real life applications include:

• Detecting and tracking the projections of torch and spotlight beams - this lead to the development of the Enlighten interactive flashlights technology
• Detecting, tracking and analysing the motion of multiple sperm in semen samples - the resulting computer-aided sperm analysis (CASA) system has been in daily use at QMC, Nottingham, since spring 2008 and is currently being commercialised
• Recovering descriptions of the growth of plant roots from sequences of CCD and confocal microscope images of live seedlings - the resulting rootrace software is freely available - to date some 750 downloads have been recorded
• Tracking groups of individuals in CCTV footage - this work has been selected for inclusion in the 2009 INSTINCT Technology Demonstrator