수중 해양환경 조건에 따른 물체의 형상 측정에 관한 연구

Abstract

This study is a basic study on the method of measuring the figuration of underwater objects installed in the coast and sea areas. The final purpose of this study is to measure the amount of change in the shape of an underwater object. In this study, the development and use cases of ROVs and underwater drones were investigated and analyzed through literature review. In addition, the amount of change in the shape of an underwater object according to turbidity and light angle was derived through the white ratio through the water tank experiment. The results obtained in this study are as follows. 1) The classification method of ROV and underwater drone is divided into three keywords, and the keywords can be divided into Portability, Figuration, and Place. The size change according to the distance between the underwater drone's camera and the underwater object can be estimated by using a formula(P_R = 348.31 exp ^-0.0634D) that expresses the amount of change in the number of pixels according to the distance of the subject. 3) According to the change of the turbidity of the underwater object, the distortion rate of the object shape was changed according to the light angle. The maximum value increased by 34.02% under the 90° light angle condition, and the minimum value increased by 22.97% under the 45°+180° light angle condition. Under the fixed turbidity condition, the distortion rate increased by 16.01% at maximum when the turbidity was 22.70 NTU, and by 1.04% at the maximum when the turbidity was 20.30 NTU. 4) When the turbidity is less than 20 NTU, the captured image can be identified more clearly through binarization using Otsu's method. 5) If the turbidity exceeds 25.30 NTU, an object with a size of 30×30 cm at a distance of 1 m becomes invisible. Finally, through the results of this study, it was confirmed that if the distance to the subject or the size of the object was known in advance in a marine environment with low turbidity, it was possible to quantify the shape of an object through shooting and image analysis of a small underwater drone.