Related Experiment Video
Updated: May 5, 2026

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging
Published on: May 5, 2011
Screening for clandestine graves via the analysis of infrared images and temperature metrics
Paige Riley-Carrier1, Mindy Greco1, Aldo D'Ottavio2
1Visiting Scientist Program, Research and Support Unit, Federal Bureau of Investigation Laboratory Division, Quantico, Virginia, USA.
Abstract:
Locating a clandestine grave is a challenging yet crucial step in a criminal investigation and eventual prosecution. Many techniques have been used to locate clandestine graves, including light detection and ranging and ground penetrating radar. Recently, thermal imaging has also been evaluated as a method for this task. However, questions such as the best time of day for use and the technique's longevity still exist. This study served as a means to address these topics. First, six simulated clandestine graves were created in a forested environment. Then, the graves were monitored for two years with long-wave infrared cameras. Temperature probes were also employed to collect surface and subsurface level data. After collection, a visual evaluation of the thermal images was conducted, and then an image processing code was employed to obtain image temperature data. From there, the data were evaluated for possible statistically significant temperature differences. Temperature probe data and camera data were also compared. In addition to the visual differences noted in the images, statistically significant differences were measured between the regions of interest and the controls throughout the experimental time period. Specifically, the location of the grave and time of day influenced the temperature response. For this specific study location, it was determined that image collection between 1700 and 0800 h provided the greatest ability to discern the regions of interest from the surrounding undisturbed areas. The temperature probes presented differences when inter- and intra-comparisons were performed. Differences also existed between the temperature probe and camera data.

