Omnidirectional Thermal Imaging Surveillance System Featuring Trespasser and Faint Detection

This paper proposed an efficient omnidirectional thermal imaging surveillancesystem featuring trespasser and faint detection. In this thermal imaging system,the omnidirectional scenes in a monitored site such as old folks home, nursinghome, hospital etc. are first captured using a thermal camera attached to acustom made hyperbolic IR (infrared radiation) reflected mirror. The capturedscenes to be monitored with trespasser or faint detection are then fed into alaptop computer for image processing and alarm purposes. Log-polar mapping isproposed to map the captured omnidirectional thermal image into panoramicimage, hence providing the observer or image processing tools a complete wideangle of view. Two effective human behavioral detection algorithms namely:Human head detection algorithm and home alone faint detection algorithm arealso designed for monitored the trespasser or fainted people detection. Theobserved significances of this new proposed omnidirectional thermal imagingsystem include: it can cover a wide angle of view (360o omnidirectional), usingminimum hardware, low cost and the output thermal images are with higher datacompression. Experimental results show that the proposed thermal imaging surveillance system achieves high accuracy in detecting trespasser andmonitoring faint detection for health care purpose.