SECURITY MONITORING SYSTEMS: SENSORS, COMMUNICATIONS AND COMPUTING

ACCURATE AND AUTOMATIC DETECTION OF HUMAN PRESENCE IN ORDER TO PREVENT UNAUTHORIZED USE OF HIGH VALUE ASSETS IS AN IMPORTANT ISSUE FOR BOTH DOE AND SOCIETY IN GENERAL. TO DATE, SECURITY SYSTEMS HAVE BEEN VERY EXPENSIVE IN TERMS OF CAPITAL AND VERY LABOR INTENSIVE TO OPERATE. IN ADDITION, THE SYSTEMS HAVE BEEN PRONE TO FALSE ALARMS, REDUCING CONFIDENCE IN THE ALARMS GENERATED. IN THIS PROJECT A SECURITY MONITORING SYSTEM BASED ON A UNIQUE COMBINATION OF VIDEO ACQUISITION, WIRELESS SPREAD SPECTRUM COMMUNICATIONS, AND MORPHOLOGICAL IMAGE PROCESSING TO DETECT THE PRESENCE OF HUMAN INTRUDERS IS PLANNED. THE ULTIMATE GOAL IS TO REDUCE BY A FACTOR OF 10 THE CAPITAL COSTS AND INSTALLATION LABOR (COMPARED TO CONVENTIONAL SECURITY SYSTEMS), REDUCE FALSE ALARMS TO NEAR ZERO, PRODUCE THE POSITIVE DETECTION OF HUMANS WITH NEAR 100% ACCURACY, AND PROVIDE FULLY AUTOMATED OPERATION. PHASE I INVESTIGATES MORPHOLOGICAL IMAGE PROCESSING FOR HUMAN DETECTION, IMPLEMENTS THE TEST BED DETECTION SYSTEM, AND VALIDATES THE SYSTEM CONCEPT BASED ON ACTUAL VIDEO. THE SOFTWARE FOR IMAGE PROCESSING RUNS ON A WORKSTATION OR PERSONAL COMPUTER FOR THE TEST BED. THE SOFTWARE WILL BE ANALYZED TO DETERMINE THE REQUIREMENTS FOR PORTING TO LOW COST DEDICATED DIGITAL SIGNAL PROCESSING. COST ANALYSIS WILL DETERMINE IF THE SIGNAL PROCESSING SHOULD BE DONE IN HARDWARE, IN SOFTWARE, AT REMOTE NODES, OR AT CENTRAL HUBS. DESIGN ALTERNATIVES WILL BE SUGGESTED FOR CHANGES IN THE IMAGE PROCESSING SOFTWARE OR IMAGE FILES TRANSFERRED TO OPTIMIZE COST AND PERFORMANCE. THE FINAL REPORT WILL PROVIDE TEST RESULTS, SUGGEST DESIGN ALTERNATIVES BASED ON TEST RESULTS, AND SUMMARIZE THE ANALYSIS OF SOFTWARE AND HARDWARE REQUIREMENTS TO DEVELOP A PROTOTYPE FOR FIELD TESTING.