Research Grants

National Institute of Justice, Award No. 2018-MU-MU-0003 ($178,080.00). Poisson Processes and Randomly Acquired Characteristics: Are Wear Features on Footwear Randomly Distributed? January 2019 – March 2022.

Women of West Virginia University Enrichment Fund ($5,000.00). Funding to support course development. June 2019 – June 2020.

National Institute of Justice, Award No. 2016-DN-BX-0152 ($213,735.00). Understanding the Expert Decision Making Process in Forensic Footwear Examinations: Accuracy, Decision Rules, Predictive Value and the Conditional Probability of an Outcome. October 2016 – March 2020.

American Academy of Forensic Sciences and Forensic Science Foundation, Lucas Research Grant. Recipient: Speir research group graduate student Madonna A. Nobel, ($4,874.00), Understanding the Expert Decision Making Process in Forensic Footwear Examinations: Accuracy, Decision Rules, Predictive Value and the Conditional Probability of an Outcome. 2016-2017.

WVU Advance/WVU Sponsorship Program. ($3,000.00). Hyperspectral Imaging to Support Forensic Evidence Recognition, Collection and Interpretation. October 2014 – April 2015.

National Institute of Justice, Award No. 2013-DN-BX-K043 ($389,321.00). A Quantitative Assessment of Shoeprint Accidental Patterns with Implications Regarding Similarity, Frequency and Chance Association of Features. January 2014 – May 2017.

 

Publications

Richetelli, N., Bodziak, W., Speir, J. Empirically Observed and Predicted Estimates of Chance Association. Part 1: Estimating the Chance Association of Randomly Acquired Characteristics in Footwear Comparisons. Forensic Science International. September 2018. Accepted.

Petraco, N.D.K., Berry, R., Del Valle, A., Gambino, C., Kammrath, B., Speir, J., Shenkin, P. Specific pattern identification via computer matching systems. Circa 2017. Under Revision

Watalingam, R.D., Richetelli, N., Pelz, J., Speir, J. Eye tracking to evaluate evidence recognition in crime scene investigations. Forensic Science International. Vol. 280, 2017, pp. 64-80.

Richetelli, N., Lee, M., Lasky, C., Gump, M., Speir, J. Classification of footwear outsole patterns using Fourier transform and local interest points. Forensic Science International. Vol. 275, 2017, pp. 102-109.

Richetelli, N., Nobel, M., Bodziak, W., Speir, J. Quantitative assessment of similarity between randomly acquired characteristics on high quality exemplars and crime scene impressions via analysis of feature size and shape. Forensic Science International. Vol. 270, 2017, pp. 211-222.

Speir, J., Richetelli, N., Fagert, M., Hite, M., Bodziak, W. Technical Note: Quantifying randomly acquired characteristics on outsoles in terms of shape and position. Forensic Science International. Vol. 266, 2016, pp. 399-411.

Pieszala, J., Diaz, G., Pelz, J., Speir, J., Bailey, R. 3D Gaze Point Localization and Visualization Using LiDAR-based 3D Reconstructions. In Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications (ETRA `16). ACM, New York, NY, USA, 201-204.

Speir, J., Hietpas, J. Frequency Filtering to Suppress Background Noise in Fingerprint Evidence: Quantifying the Fidelity of Digitally Enhanced Fingerprint Images. Forensic Science International. Vol. 242, 2014, pp. 94-102.

Hietpas, J., Samson, S., Speir, J., Moecher, D. Assessing detrital garnet chemical composition as a quantitative provenance tool: A multivariate statistical approach. Journal of Sedimentary Geology. Vol. 83, No. 12, 2013, pp.181-1197.

Petraco, N.D.K., Shenkin, P., Speir, J., Diaczuk, P., Pizzola, P., Gambino, C., Petraco, N. Addressing the National Academy of Sciences’ Challenge: A Method for Statistical Pattern Comparison of Striated Tool Marks, Journal of Forensic Sciences, Vol. 57, No. 4, 2012, pp. 900-911.

Speir, J., Goodenough, A., Schott, J., Brown, S. Validation of In-Water 3D Radiative Transfer using DIRSIG. IEEE Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), 2010, 2nd Workshop, Reykjavik, Iceland 2010, pp. 1-4.

Speir, J. A Case Study: The Observation of Dispersed Extinction in Fibrous Richterite/Winchite. The Microscope. Vol. 54, 1st Quarter, 2006, pp 25-29.

Speir, J. Cleaning, Maintenance and Adjustment of the Light Microscope. The Microscope. Vol. 54, 2nd Quarter, 2006, pp 61-70.

 

Funding for Undergraduate Research in Microscopy

The Microscopy Society of America Undergraduate Research Scholarship Program will support your research!  If you are interested in submitting an application and would like to work in my laboratory,  please read the announcement available via the following weblink, and feel free to contact me to discuss project options.

Click to access ResearchScholarshipApplication.pdf

 

NEAFS

 

Tara Fikes was recently honored at the Northeastern Association of Forensic Scientists Anuual 2011 meeting, receiving the “Best Poster” award for research entitled “Fast Fourier Transformation and Frequency Filtering to Suppress Background Noise in Fingerprint Evidence”.   Congratulations, Tara!

 

Shoeprint Registration

Comparing shoeprints using semi-automated methods requires that all images be registered to a common 2D “space”.  For this research, a simple translation and rotation is considered sufficient to accomplish full image registration.  The series of images on the right illustrate the result of registration for two shoes of the same size. The original image shows a 600-PPI digital scan of a shoeprint, including contrast reversal (Ref. Manal Khalil, 2011).  The rightmost image in the series depicts the same image, but after registration.  The original image has been padded on all sides, translated to ensure that a single feature in the image resides at a common (x,y) coordinate, and then rotated about the translation point to ensure that the details of the tread pattern distribute along orthogonal N-S and E-W axes.  The central image overlays the registered scan with a baseline print used to represent the common 2D “space”.  This overlay illustrates the agreement between the registered image and a “master print”.

The second series of images illustrate the same registration process, but for a shoe several sizes smaller than the master print.  The overlay illustrates that the smaller shoeprint is coincident with the master print at the translation point and that all features are distributed along N-S and E-W orthogonal axes, but naturally all features will not agree since the overlaid shoes are of different sizes.

To see details and a magnified version of either illustration, please click on the image which will open up a new page with zoom capabilities.

 

Gunshot Reside (GSR) Distributions

The distribution of shotgun pellets on a target varies depending on several parameters, including the shotgun choke and firing distance.  The resulting distribution can be characterized by the radius of a “best-fit circle” that describes the spread of the pellets.  This theory was applied to hand-loaded ammunition (with variations in both powder load and bullet weight) discharged from  9mm handguns in an attempt to evaluate the manner in which these variables impact the distribution of unburned and partially burned gunshot residue (GSR).

 

Ref. This research was conducted in partial fulfillment of the Master of Science in Forensic Science at John Jay College of Criminal Justice (New York, New York), in collaboration with Jack Hietpas, Peter Diaczuk and Dr. Peter De Forest.  Work was carried out at the Pedico Research Institute (PRI) in Pennsylvania.

 

Forensic Microscopy of Explosives

The polarizing light microscope (PLM) can be used to characterize the optical properties of numerous materials.  The image at right depicts conoscopic (top and leftmost image) and orthorscopic (bottom most images) observations associated with a single crystal of the highly explosive material known as TATP, or triacetone triperoxide.  In the last several years, this explosive has been implicated in terrorist activities, making its characterization and identification of paramount importance to first responders.  (Note: this figure illustrates how immersion methods and polarized light microscopy can be used to determine the refractive indices associated with a biaxial material).

Ref:  This research was carried out while working at the McCrone Research Institute (Chicago, Ilinois), and in collaboration with Jack Hietpas, formerly of Microtrace (Elgin, Illinois) and currently at Syracuse University.

 

Synthetic Image Generation – DIRSIG

The radiative transfer equation (RTE) is a mathematical description of radiative gains and losses experienced by a propagating electromagnetic wave in a participating medium. Except for an isotropic lossless vacuum, all other volumes have the potential to scatter, absorb and emit radiant energy. Of these possible events, the global scattering term is the greatest obstacle between a radiative transfer problem and its solution. Historically, the RTE has been solved using a host of analytical approximations and numerical methods. Typical solution models exploit plane-parallel assumptions where it is assumed that optical properties may vary vertically with depth, but have an infinite horizontal extent. For more complicated scenarios that include pronounced 3D variability, a photon mapping Monte Carlo statistical approach to the radiative transfer solution is often utilized.  The synthetic image shown here depicts the photon mapped solution of a step function submerged in a scattering volume at three different depths (increasing from top to bottom) and with variable photon mapping parameters that impact the resolution and SNR associated with the step function.

Ref. This research was carried out in partial fulfilment of a Doctorate of Philosopy in Imaging Science at the Rochester Institute of Technology (Rochester, New York), in collaboration with Dr. John Schott, Dr. Adam Goodenough and Dr. Scott Brown.

 

Fingerprints & Fourier Transformations

Frequency Filtering of Fingerprints to Remove Noisy Backgrounds

The Fourier transformation (FT) of an image allows selective filtering of frequencies associated with noisy backgrounds; the end result is a cleaner filtered image.

The image shown here illustrates the utility of the filtering process, beginning with a photograph of a fingerprint (topmost  image, Ref. Fikes, T. (2011)), the FT power spectrum and four filters that remove frequencies associated with the interfering vertical lines (center image), followed by the inverse transformation (bottom image), or a cleaned-up version of the original fingerprint.  Elliptical regions highlight frequences that have been removed by the filtering process.

 

Correlation as a Metric for Similarity in Shoeprints

Assessing the Degree of Similarity between Accidental Patterns on Shoeprints Associated with Wearers that Participate in Shared and Independent Activities. 

 

We are currently working on a project that attempts to describe the nature of accidental patterns as a function of wearer-context. The goal of this research is to address the similarity and rate at which wear-derived random accidental characteristics develop on the outsoles of individuals that participate in shared versus independent activities.  To carry out this research, two groups of volunteers were solicited and provided with new, approved footwear. The first group was asked to wear the footwear while repeatedly participating in shared group activities over a three-month period of time, while the second group was permitted to wear the approved footwear while carrying out daily independent activities.  At predetermined step-intervals participants submitted their footwear for analysis, which consisted of data (a.) acquisition, (b.) registration, (c.) segmentation, (d.) processing, and finally, (e.) comparison.