Detection of Human Skin in Near Infrared Hyperspectral Imagery

One of the difficulties in search and rescue missions is finding a small target, such as a person, in a large cluttered area. Airborne hyperspectral cameras are now being deployed to aid in this SAR mission. Motivated by the successes of such systems, we define a hyperspectral model of human skin in the visible and near infrared regions of the spectra so we can exploit knowledge gained during the modeling process to aid in human skin detection. Based on observations of the skin model results, an efficient and robust skin detection algorithm using channels in the near infrared region of the spectra is developed. Our algorithm is denoted the Normalized Difference Skin Index, motivated by the Normalized Difference Vegetation Index used in the literature for detecting vegetation in hyperspectral imagery. We demonstrate the capabilities of our skin detection methodology to detect skin amongst objects known to cause false detections for methodologies using three channel color data.

1. INTRODUCTION

Detecting small targets in geographically large areas with a significant amount of background clutter is a difficult task. This is exactly the task of Search and Rescue (SAR) missions, where the targets of interest are people. Detection methods that exploit color imagery of the search area provide one way for locating lost persons [1], but are limited in capability. Hyperspectral data provides a diverse set of features that can possibly improve the successes in SAR, and are currently being exploited by the Civil Air Patrol (CAP) Airborne Real-Time Cueing Hyperspectral Enhanced Reconnaissance (ARCHER) system [2]. Data cubes collected by airborne hyperspectral cameras can be analyzed to determine if there is something in the image worthy of further investigation. It can be difficult, however, to quickly determine which pixels of the hyperspectral image are “background” and which ones are areas of interest to the SAR team. Using spectral signatures, one can locate missing persons or vehicles among the imagery, and dispatch rescue teams to those areas of interest. A simpler method for analyzing the hyperspectral images on the aircraft versus a ground station (as used in CAP ARCHER) could not only speed up the SAR process, but make aspects of this process easier.