Underwater Obstacle Detection Via Relative Total Variation and Joint Guided Filtering for Autonomous Underwater Vehicles
Abstract
Underwater Obstacle Detection is Essential for Safe Deployment of Autonomous Underwater Vehicles (Auvs). in This Paper, We Make an Attempt to Explore One Kind of Underwater Obstacle Detection Strategy with the Help of the Relative Total Variation(Rtv) and Joint Guided Filtering(Jgf). We First Introduce One Kind of Virtual Retinex Model. Then We Utilize the Relative Total Variation to Extract the Essential Visual Structures. We Further Take Possible Visual Differences between Reference and the Target Underwater Image into Account and Estimating the Mutual Structures with the Joint Guided Filtering. Then We Generate the Underwater Obstacle Edges with Gaussian Filter in a Lab Color Space. It is Shown in Our Simulation Experiments that the Developed Approach in This Paper Could Achieve Great Performances in Underwater Obstacle Detection for Auv.
Recommended Citation
Y. Chen et al., "Underwater Obstacle Detection Via Relative Total Variation and Joint Guided Filtering for Autonomous Underwater Vehicles," OCEANS 2017 - Anchorage, pp. 1 - 6, Institute of Electrical and Electronics Engineers, Dec 2017.
Department(s)
Engineering Management and Systems Engineering
Keywords and Phrases
Gaussian filter; Joint Guided Filtering; Relative Total Variation; The LAB color space; Underwater obstacle detection; virtual retinex model
International Standard Book Number (ISBN)
978-069294690-9
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2024 Institute of Electrical and Electronics Engineers, All rights reserved.
Publication Date
19 Dec 2017
