|
|
IMPRESS – IMPRession Evidence Serial-crime System
Universities of Surrey, Sheffield and Strathclyde
Introduction
Property related offences, especially theft offences, criminal damage, and burglary, accounted for over 75% of all the 5.5 million offences reported in the UK in 2001/02; many of these crimes are committed by habitual criminals. Edmond Locard's Principle of Exchange states that when any two objects come into contact, there is always transference of material from each object onto the other. Material transferred in this way, and subsequently used as evidence, is referred to as impression evidence. Criminals leave similar impression evidence, including fingermarks, footwear impressions, ear prints and clothes prints, at subsequent crime scenes: A system to identify commonalities in impression evidence would assist in detecting and apprehending serial criminals. It is important to systematically store and cross-reference different kinds of impression evidence. It is equally important to retrieve the evidence and identify similarities in this evidence in a timely manner. The correlations of impression evidence found at different crime scenes related to the same offender will yield robust evidence. The awareness amongst criminals that tiny amounts of impression evidence left at the scene of crime can be traced back to them will act as a deterrent. The objective of the proposed two-year IMPRESS project is to design, build, test and exploit an information system that can help store, retrieve and link a range of impression evidence data. This data will be used primarily to relate the impressions at a scene to previous crimes committed by the same perpetrator. The interdependent modalities of impression evidence images and the textual descriptions provided by forensic scientists may be used to store, retrieve and link impression evidence.
Aims and objectives of the technology research
The main aim of this project is to specify, design and prototype an investigation management system that will allow the interrogation of a number of different impression evidence databases, possibly from remote locations, for identifying (habitual) perpetrators of crime. Impression evidence typically comprises an image together with its description. Transmission and reception of privileged digital images is resource intensive and requires a secure channel of communication. These images are described in a specialist language by trained scene of crime officers and forensic scientists. The research challenge to be addressed in this project is to use the description of an individual image to generate a keyword index of the image, which may include keywords specific to the crime scene and to the perpetrator. The keywords together with the image-specific features, shape and texture, from different types of impression evidence will help in building an impression identikit of the criminal. The IMPRESS project is being proposed by three universities – Sheffield, Surrey and Strathclyde – for the intelligent management of impression evidence. Five Police Forces and two software houses currently support this proposal. Further support is being solicited from the UK Forensic Science Society. The combination of two computer science departments, Surrey and Sheffield, and a Forensic Science academic unit, Strathclyde, together with the other partners, will introduce a new approach to evidence gathering and subsequently pooling information from different types of impression evidence to build up an offender profile. Crimes related to property will be our main focus. Our proposal is to demonstrate that developing and delivering collaborative Information Extraction technologies for processing texts and images in combination over the new UK Grid infrastructures can provide these facilities. Such a system should raise awareness of the problems inherent in crime detection by a novel collaboration between computer scientists and forensic scientists and may lead to a safer living environment for the general public.
The excitement and novelty of the research challenges to be addressed
The challenges to be addressed relate to the descriptive interpretation of information contained in images. Descriptions applied to images of impressions need to be rigorous and unambiguous so that the evidence based on impression data can be presented in a court of law; indeed, with the accepted accuracy of biological analysis, it is easier to attack the credibility of the expert rather than the credibility of the evidence. Forensic scientists perform a crucially important real-world task where heuristics abound and much knowledge is personal knowledge; in some respects, the forensic scientists behave like other diagnosticians that have been investigated in the AI literature and in other respects they behave like hypotheses-makers as in law and cognate subjects. Literature on contents management, knowledge management and multimedia systems constantly refer to the intellectual challenge of dealing with a mixture of perceptual and cognitive modalities. Information fusion, where it helps in the enterprise of forensic intelligence, will throw light on intelligent information processing in human beings on the one hand and will help in the development of robust systems on the other. We believe our approach to building repositories for Information Extraction, especially in ontology systems, is different from that of conventional prescriptive or highly theoretical approaches in the literature for building knowledge bases; we have used use pre-existing semi-structured repositories of knowledge including free text, lexical and terminological resources, to maximally exploit the content in them with our integrated extraction techniques. The proposed research will lead to a unique combination of advanced and developing technologies enabling the fusion of information and extraction of key facts from this fusion.
The significance of selected research area, how this meets the vision of TCPD
The need for intelligence led policing has been stressed by ACPO and it has been noted by its chairperson that "police activity has shifted its centre of balance away from reactive investigation after events towards targeting active criminals on the basis of intelligence.” (David Phillips, ACPO Chair). The impression evidence is collected, stored and retrieved in an idiosyncratic manner by each Police Force. This idiosyncracy can hamper subsequent retrieval of the data by other forces. There are serious training implications. The three Universities in the project will help undertake an investigation of the variation in the way in which impression evidence is described, stored and retrieved. Strathclyde will use the results of this investigation to inform their teaching and learning programmes in Forensic Science. We will conform to the TCPD vision by realising the potential of Information Extraction, data mining and Grid technologies, for underpinning R&D efforts in forensic science for detecting crime, especially property crimes. We will raise awareness of research in our specialised areas within the forensic science and the police community through novel manners of interaction and dissemination.
Nature of the research team and its ability to deliver the research project aims
The specific research advance from the existing EPSRC SoCIS cooperation (Universities of Surrey and Sheffield, EPSRC Grant No. GR/M89041/01) has been an integrated working prototype that links the searching of structured images to the searching of structured titles; this bringing together of meaningful structures in both language and vision has been a holy grail of artificial intelligence for over thirty years and we believe we have made real progress as partners in SOCIS, rather than just using one modality (e.g. language) as an index for searching another (e.g. pictures or video).
Given the partial success of this collaboration so far, we believe this can now be deepened and strengthened by the addition of Strathclyde's experience of research in forensic science in general and in DNA impression evidence in particular. Sheffield and Surrey have brought extensive and evaluated experience in Information Extraction, a widely distributed architecture for modular Natural Language Processing applications of this sort, and techniques for building and integrating ontologies with semantic structural representations and Information Extraction. Work has also been successfully carried out in neural network based learning of descriptions of images. The University of Strathclyde is a long established academic institute that has a track record of being practitioners in addition to delivering training in the forensic sciences. Strathclyde will act in the first instance as a user and co-ordinate the development of the next generation of SOCIS. This will be the first interface between the design of SOCIS and the use of the programme in a forensic context. Immediate feedback by forensic practitioners to the designer is possible allowing deliverable goals within the life of the project.
Outline of proposal methodology
Our proposed methodology will be:
1. To develop a method for fusing information from different types of impression evidence and relate it to an offender.
2. To investigate the similarities and differences in which the images of different impressions are described and to investigate indexer variability within the same types of images.
3. To explore the fastest and most secure computing systems for fusing images from different sources by exploring the proposed UK Grid (or ‘Internet II’).
4. To incorporate computer-based evidence methods and techniques within forensic teaching and training curriculum.
With regard to point 3 above it should be noted that Surrey have submitted a proposal (GR/S26828/01) to become an e-Science centre of excellence, for exploring the use of the Grid for language based information and knowledge analysis.
Evidence of connectivity and explanation of the value of the proposed Collaboration
We have formed a consortium that includes police forces (Hampshire, Kent, South Yorkshire, Surrey and the Metropolitan Police Training College) and software systems houses (Solcara and Motion Touch) specializing in crime prevention and detection technologies. To ensure that this goal is achieved we propose to appoint a knowledge dissemination officer who will work with the Forensic Science Society and provide a conduit by which research results can be communicated to the Society’s members. We are also approaching Police Forces in Scotland, Wales and Northern Ireland. Solcara and Motion Touch are evaluating the current SOCIS system for commercial exploitation.
Outline of management structure
The proposed consortium includes five UK police forces, three Universities, and software systems developers specializing in crime prevention and detection technologies. The project will be co-coordinated by a Round Table chaired by an experienced forensic science practitioner from one of the five Police forces. The Universities will draw up an EPSRC-conformant work plan of research and development in consultation with the other partners. The Round Table will be involved in chasing progress as per the work plan and will suggest changes or modifications accordingly. The universities together will produce quarterly progress reports for the duration of the project. The software houses will explore possible exploitation routes with the support of the Universities. The proposed consortium strongly reflects the need for user-led development of research projects.
Arrangements for take up (e.g. IPR ownership) and dissemination of results
During the life of the project IPR ownership will rest jointly with the Universities of Surrey, Sheffield and Strathclyde. This could be transferred, in whole or in part subject to discussion, to any interested parties on completion of the project. We propose three routes for dissemination: Firstly, the conventional route of research dissemination, journal publications and peer-reviewed conference presentations, will be followed in IMPRESS. Secondly, the Round Table for continually disseminating results to the larger forensic science and police community. The Round Table has been a source of invaluable experiential knowledge and for on-going evaluation and feedback on system development. Typically, the potential user community does not know the results of research projects. The Round Table for our current SOCIS project has led to the introduction of a new Police Force (Metropolitan Police) to the project and we hope to replicate this effort by including other institutions in the Round Table. The Forensic Science Society and other related professional bodies may be persuaded to join the Round Table. Thirdly, the University of Strathclyde in collaboration with the Universities of Surrey and Sheffield will produce a training programme that can be delivered either at universities teaching forensic science or at recognised training establishments. Such a training programme will standardise and enhance teaching programmes within the UK.
|
