Remote Coating Defects Inspection by Drone Using Visual Imagery Deep Learning Method

Heidir, M. Rafif Fauzan (2022) Remote Coating Defects Inspection by Drone Using Visual Imagery Deep Learning Method. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 04211841000044-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Ship inspection is one usual activity to be done in order to ensure the seaworthiness of a vessel. In general, ship structural inspection with close-up survey requirements may cause problems. Close-up survey usually carried out to inspect detailed defects in the ship constructions. This activity poses a safety risk to the ship surveyor. method that are used in this research are deep learning method using convolutional neural network (CNN) to identify and classify paint coating defects on a steel structure. The program developed to be implemented on drone to substitute human surveyor. the concept of using drone to substitute human surveyor are also called as remote inspection technologies (RITs)The architecture of the CNN that is selected in this research is the MobileNet architecture the architecture has higher accuracy at lower processing or computing load at 2000% lower than other CNN architecture in general. This can be an advantage for coating defect detection for fully-remote on-board of the drone. Program training is done at 50.000 steps with 0,7945 loss value at final training step. The result of the program is a satisfactory as the program successfully detect and classify the paint coating defects on the ship super-structure with accuracy over than 60% and in a real-time processing on PC.

Item Type:	Thesis (Other)
Additional Information:	RSSP 623.746 9 Hei r-1 2022, - 3100022096515
Uncontrolled Keywords:	Coating Defects, RITs, CNN, Drone, MobileNet
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing.
Divisions:	Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	03 Mar 2026 06:03
Last Modified:	03 Mar 2026 06:03
URI:	http://repository.its.ac.id/id/eprint/132686

Actions (login required)

View Item