Noting the “huge practical advantage” of using an imaging device to quickly examine a crime scene for forensic evidence — particularly bloodstains and fingerprints — researchers at the New Jersey Institute of Technology developed a prototype handheld multispectral camera for potential future use by investigators.
There have been a number of efforts in forensic science to develop a relatively inexpensive device that could provide real-time multispectral imaging of crime scenes that would allow investigators to see latent evidence such as blood stains, ink stains, latent fingerprints, and tire prints. Earlier efforts have shown promise, but, according to researchers at the New Jersey Institute of Technology, the technical challenges of weight, cost, complexity, ruggedness, portability, and user friendliness have held the technology back.
The goal of these NIJ-supported researchers, led by physicist John Federici, was to develop a compact, lightweight camera capable of simultaneous imaging at both visible and near infrared wavelengths. The researchers developed the device utilizing multiple cameras linked to mini-computers and combined into one unit.
Although the miniaturization of off-the-shelf cameras and mini-computers didn’t progress as rapidly as the researchers had hoped, they did create a Dichroic camera that can see 16 spectral colors in the 400-1000 nanometer range, and they demonstrated “the utility of the camera system on latent blood stains and fingerprints.” Multispectral images were acquired under various conditions for various surfaces, including dark cloth, leather, and paint.
A multispectral image captures data in the form of different wavelengths in the electromagnetic spectrum, including visible light, ultraviolet, and infrared. Multispectral imaging is designed to synthesize image data, allowing the user to view details that far exceed what can be seen by the human eye. This technology is commonly used in satellite imagery, health care, and military applications.
If bloodstains were deposited on lightly colored materials, such as a white or yellow T-shirt, the device could visualize the stain as blood through its “reflectivity spectra.” If the bloodstains were deposited on a dark material, such as a black T-shirt, the blood and background spectra would be too similar to allow identification as blood. The multispectral camera could, however, serve as a “stain detector,” the researchers said, alerting investigators to the presence of a stain.
In most cases, the camera was able to detect stains even if there had been attempts to remove them with cleaning agents. Although washing cloth in a household washing machine removed all evidence of stains (except on white fabrics), less aggressive forms of cleaning left sufficient residues of the stains that were detectable by the camera.
The attempts to locate and possibly identify latent fingerprints with the multispectral camera exhibited “some successes, some failures, and some inconclusive results,” the researchers said. Although the resolution of monochrome cameras were good enough to allow the fingerprint images to be compared to computerized database of prints, the multispectral approach did not offer any advantages in contrast in this instance.
“Given the wide variety of background/substrate materials and the need to adjust the illumination conditions to obtain the best contrast,” the researchers concluded, “multispectral imaging of fingerprints ... does not appear to be a promising technology for crime scene investigation.”
They also noted that by modifying the camera to work in the 1600-2500 nm range, the device could “spectroscopically identify blood stains even on dark clothing.” The issue for an infrared camera capable of that range, they noted, is one of cost and image format size. “The cost would be about 10 times as high as the current device,” they said, “and the pixel size (or resolution) would be significantly less.”
A next step, according to the researchers, would be to determine if a multispectral camera that can locate blood, semen, or other stains, would be useful enough to crime scene investigators to continue the efforts to create such a device.
About This Article
The research described in this article was funded by NIJ award 2014-DN-BX-K003 awarded to the New Jersey Institute of Technology. This article is based on the grantee report “Hand-Held Multispectral Camera for Crime Scene Investigation” (pdf, 13 pages) by John Federici.
This research is part of a broader portfolio of impression and pattern evidence projects managed by NIJ physical scientist Danielle McLeod-Henning, Ph.D. Find more information on NIJ's impression and pattern evidence portfolio.