An experimental analysis of the use of digital crime scene footwear capture methods compared to traditional manual alternatives

This PhD will examine innovative technology designed to digitise the capture of crime scene footwear evidence in comparison with the current manual handling process.

The technology has been developed to enable digital photographic capture of crime scene footwear evidence and will compare this digital evidence with innovative technology being used across London to capture detained person’s footwear samples through the use of a walk-on digital scanner known as Tread Finder.

The ultimate aim is to establish whether or not it is possible to digitally compare the two samples and provide evidence sufficient for prosecution in a more efficient, effective way.

The project will answer the research questions through a series of experiments in order to test new and emerging technology designed to provide a dark room environment out in the field.

The use of two different prototypes are planned. The first has been specifically designed to capture footwear evidence from crime scenes in situ from a variety of 2D surfaces and the second is designed to complement the existing 2D manual capture method of gel or electro static lifts.

In addition to this, and in response to Home Office Joint Review of the provision of Forensic Science 2018 – Forensics Implementation plan: Recommendation four: Ensure practitioners and policy makers have stronger evidence and data to support decision making and facilitate more effective working with partners.

This is needed to maximise the opportunity for forensic science to fulfil its role as an enabler of robust outcomes and strengthen investment cases.

Actions undertaken in response to this recommendation

Survey participants

The researcher will work with the Home Office and the National Police Chiefs' Council (NPCC) Transforming Forensics programme through the National Forensics Performance and Standards to engage with stakeholders across the Criminal Justice System (CJS) and obtain voluntary participants to undertake the survey.

Semi-structured interviews

The researcher will work with the Home Office and NPCC Transforming Forensics programme through the National Forensics Performance and Standards Group to engage with stakeholders across the CJS and obtain voluntary participants to undertake semi-structured interviews

Case file randomised sampling

The researcher will work with the Home Office and NPCC Transforming Forensics programme through the National Forensics Performance and Standards Group to engage with stakeholders across the CJS to identify a suitable sample size of cases to randomly examine and ascertain relevance of forensic evidence in each.

A series of experiments were planned between July 2018 – July 2019. Initial experiments took place in the controlled environment of a university and a police forensic science laboratory.

The plan was to test the hypothesis that the new digital technology was at least as good in terms of accuracy and superior in terms of efficiency. If this was found to be the case, then plans were in place to move experiments into the field and at live crime scene events.

The initial phase of experimentation in regard to testing crime scene equipment was planned to start in April 2018 as per the project timeline, however, prior to experimentation, it was necessary to broker agreement with the international hardware and software suppliers and import the equipment.

This stage took longer than expected and therefore testing could not commence until July 2018. The initial experiments were planned as a comparison using a sample pair of footwear to provide consistency throughout the testing. The testing was aligned with the NPIA Footwear Manual which set out a range of conditions and recommended traditional forensic techniques for recovering footwear marks (NPIA 2007).

Each of the conditions were replicated as far as was possible in the mock crime scene rooms based within the University of Bournemouth.

With regard to the 600dpi scanner experiments, these were intended for April 2018 to Sep 2018 and were planned to take place at the controlled police scientific laboratory situated at West Yorkshire Police Scientific Services headquarters. Research assistants (forensic science examiners) were recruited to undertake the sampling process so as to ensure no bias and samples were taken from real-life evidential comparison cases.

A sample size of 37 cases was calculated based on an overall population of 250 annual cases, a 95% confidence level and a confidence interval of 15. The method was to replicate a scan of each examined items of footwear using the traditional acetate method, the 300dpi scanner, the 600dpi scanner and a flat-bed scanner.

Treatments / independent variables were as follows.

1. The traditional manual method (gel lift, electro static, photograph or acetate) along with the laboratory based gel scanner.
2. The new, commercially available digital capture method using a Canon 5DSR and with EF 40mm f2.8 STM lens.
3. 300dpi scanner (Tread Finder design), 600dpi scanner – newly developed scanner and Konica scanner

Dependent variables were:

• time take to process sample
• cost associated with each treatment
• accuracy of footwear sample compared to the baseline original measurements measured using a calibrated tool

Upon review of the outcomes from the first phase of experimentation, consideration was given to extending the experiments to live crime scene environments. Upon reaching this stage, consideration to running a Randomised Control Trial (RCT) was also made. The same treatment and dependent variables would be used.

The first conclusion drawn was that the crime scene device was not sufficiently mature in its development to progress to field-based experiments. Errors were too frequent and the device was not capable of coping with numerous environmental challenges it faced. For example, the bright summer sunshine rendered the device worthless when out in the open and would only operate after draping a makeshift cloak over the device.

The decision was therefore taken to suspend the field-based experiments, bringing to an end research question three – how does the new crime scene digital technology compare to existing methods in terms of cost, accuracy of original mark preservation and benefit in a field environment?

Findings in the second branch of experiments conducted at West Yorkshire Police are yet to be examined in great detail as it was necessary to negotiate access to commercial image comparison software, which has now been secured free of charge on a six month basis.