Locating victims in hot environments using combined thermal and optical imaging. (July 2020)

• Record Type: Journal Article
• Authors: Dawdi, Takwa M.; Abdalla, Norhan; Elkalyoubi, Yusra M.; Soudan, Bassel
• Is Part Of: Computers & electrical engineering. Volume 85(2020)
• Publication Date: 2020-07
• Subjects: Thermography -- Image blending -- Victim detection -- Template matching -- Unmanned Aerial Vehicle -- Image Processing; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2020.106697
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Optical images are blended into thermographs to alleviate thermal noise effects of hot environment. An algorithm is proposed for performing image processing on blended images in real-time on-board a UAV. Human victims are detected in processed blended images using outline-based template search. Algorithm executes in less than 2 s on a raspberry PI SBC enabling survey of a significant area. Algorithm succeeds in detecting all victims in elevated ambient temperatures with few false positives. Abstract: This work presents an autonomous system based on an Unmanned Aerial Vehicle (UAV) for victim detection using shape template matching. The system blends an optical image into a thermograph to overcome limitations inherent in the visible and infrared spectra. The system also implements image enhancements to counter thermal image degradation in warm climates. All processing is performed in real-time on-board the UAV to maximize autonomy. The outcome is a set of coordinates allowing direct access to the detected victims by rescue crews. An evaluation vehicle was constructed using a readily available drone with standard off-the-shelf components. The prototype accurately detected all victims in the test area, with rare false positives. The proposed image analysis combination was effective even under elevated ambient temperatures. The prototype was able to capture and analyze an image every three seconds allowing travel at a speed close to 2 m/sec. An area of 10, 000 m 2 can be scanned from an altitude of 10 m without the need for recharging.