Online Dead Body Detection Experiment with an Unmanned Underwater Vehicle
Merve Berik, Seda Karadeniz Kartal, None None
proceeding (Uluslararası, Hakemli) | Yazar(lar): Merve Berik, Seda Karadeniz Kartal, None None | Yıl: 2018 | Dergi/Kurum: 2018 26TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU) | İndeks: WOS.ISTP | Anahtar Kelimeler: underwater body detection, educated classifiers, unmanned underwater vehicle, haa
In this study, real-time online body detection under water was conducted using an unmanned underwater observation tool. According to the underwater position of the vehicle, data from the vehicle camera is provided to identify different body parts of a body in a real-time video stream. Here we present an approach for underwater human body detection based on the use of highly educated classifiers. The algorithm's performance in real time video shooting of the car is optimized to reduce the false positive rate by aiming to identify a corpse part of each picture frame. According to the results obtained, it was ensured that the corpse parts were successfully detected under changing conditions with incorrect positive perception. Algorithms were developed using the Pyhton program.