Abstract
Fingermarks play important role in forensic science. Based on the ridge detail information present in a fingermark, individualization or exclusion of a donor is possible by comparing a fingermark obtained from a crime scene to a reference fingerprint. In this process, the intrinsic features of a fingermark are used to determine the source of the fingermark. However, in some cases, the source of a fingermark is not argued but the activity that led to the deposition of the fingermark. The question changes from ‘Who left the fingermark?’ to ‘How did the fingermark end up on the surface?’ which requires a different assessment of the findings. The aim of this dissertation is to determine how fingermarks could provide information about activities in a reliable way, in order to be used in the forensic evidence process. To answer this main research question, several studies were conducted which are described in Chapters 2 to 5 of this dissertation.
Chapter 2 describes the development of a general framework to evaluate fingermarks given activity level propositions. Relevant variables that function as sources of information when evaluating fingermarks given activity level proposition were identified. Based on these variables, three Bayesian networks were presented for different evaluations of the fingermarks given activity level propositions in a case example. The presented networks function as a general framework for the evaluation of fingermarks given activity level propositions, which can be adapted to specific case circumstances.
Chapter 3 shows how the proposed framework in Chapter 2 can be used in casework by showing a case example. In order to use a Bayesian network, probabilities need to be assigned to the Bayesian network. In this study, a case specific experiment with the use of knives was conducted and the resulting data was used to assign probabilities to two Bayesian networks, both focusing on a different use of the experimental data. This study has shown how different uses of the data resulting from a case specific experiment on fingermarks can be used to assign probabilities to Bayesian networks for the evaluation of fingermarks given activity level propositions.
In Chapter 4, we focus on the location of fingermarks on an item. In this study, we developed a classification model to evaluate the location of fingermarks given activity level propositions based on an experiment with pillowcases. The results showed that fingermark patterns left on a pillowcase by smothering with a pillow can be well distinguished from fingermark patterns left by changing a pillowcase of a pillow. The result of this study is a model that can be used to study the location of fingermarks on two-dimensional items in general, for which is expected that different activities will lead to different trace locations.
Chapter 5 investigates the application of the location model presented in Chapter 4 to a dataset of letters, to study whether the model could also be used to distinguish between fingermark patterns left when writing a letter and fingermark patterns left when reading a letter. Based on the results of this study we conclude that the model proposed in Chapter 4 is indeed applicable to other objects for which it is expected that different activities lead to different fingermark locations, given the condition that the training set is representative for the object to be tested with regards to the size of the object and the activity that was carried out with the object.
This dissertation supports the view that fingermarks contain valuable information about the activity that caused the deposition of the fingermarks and provides the forensic community with reliable methods that can be used when evaluating fingermarks given activity level propositions.
Original language | English |
---|---|
Qualification | Dr. |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 28 Oct 2021 |
Place of Publication | s.l. |
Publisher | |
Print ISBNs | 9789090351582 |
Publication status | Published - 28 Oct 2021 |
Keywords
- forensic science
- fingermarks
- fingerprints
- activities
- activity level
- fingermark location
- classification
- bayesian network
- evidence
- evidence interpretation