抄録
The ocean that covers 71% of the Earth's surface is the source of origin of the life and biological evolution. Furthermore, ocean plays major role in regulating the weather and the climate of the planet. The features of the deep ocean also effects on controlling the climate of the earth. In addition, lots of discoveries, abundant mineral resources, energy, marine lives, and hydrothermal vent in the deep ocean have attracted research scientists of various fields. In spite of attractive discoveries, we are still unable to get clear information about the mechanism of the Deep Ocean and ecological system, because Deep Ocean is an extreme environment that has high pressure, darkness and radio attenuation. Recently, Autonomous Underwater Vehicles (AUVs) are used in Deep Ocean research activities as a powerful tool. AUVs are enrich with various facilities that can take photographs include lifestyle of marine lives, etc. In this paper, we have proposed a method to recognize the Benthos on the deep seabed automatically for various research activities. As a result, the proposed method has shown that it can detect about 84.7% of targets in the deep ocean images and capability of recognition is about 74% of the detected targets.
