Criminals captured on film could be put in the frame thanks to an expert in graphic and image manipulation from Staffordshire University.
As part of a PhD investigation, James Robinson is analysing how an advanced software program he has developed can be used on poor quality security camera footage to generate enhanced 3 Dimensional mug shots of offenders.
‘Security camera footage is often poor in quality and while it records offences taking place it does not accurately reveal the identity of those who are caught on film,’ said James, 23, from the School of Computing.
‘The combination of poor lighting, small images with low resolution and footage recorded on video tape, which is subject to distortion and noise, means generating clear and identifiable profiles is often not possible.’
During the pioneering new software process, currently being interrogated by James as part of his doctorate, a selection of poor quality images of a suspect are imported into a computer as a series of still frames.
The operator then marks on-screen the key features and characteristics of the face by plotting a series of markers onto the images. An electronic ‘mesh’ frame is then generated by the software and moulded onto the images (example screenshot enclosed) – linking the highlighted key features and mapping the textures of the face.
The software then gathers common features such as such as the eyes, nose, mouth and hairline from the range of mesh models and stitches them together to generate a 3D model of the face and head.
Once the model is created it is fleshed out and can be increased in size and rotated, offering the police authorities an enhanced and more accurate image of the offender or suspect.
James, who has a degree in software engineering, outlined how this advanced computer modelling process when fully developed could remove the need for the human monitoring of security camera footage and ultimately make CCTV more effective.
‘At present operators are needed to scan multiple-screens of images generated by the eyes-in-the-sky. Operators have to manually override the system and zoom in on suspicious characters to reveal their identity, which focuses their attention on just one location.’
‘I am continuing to investigate how effective my software can be in producing enhanced images from unsupervised footage which will make the system more efficient and effective.’
James, who hopes to completed his thesis in Autumn of 2001, is also considering other potential uses for this image manipulation process – such as the generation of electronic version of your own image to put yourself into the latest computer games or within a virtual environment.