Special Issue on Visual Information Processing for Underwater Images and Videos: Theories, Algorithms, and Applications
摘要截稿:
全文截稿: 2020-05-01
影响因子: 2.779
期刊难度:
CCF分类: C类
中科院JCR分区:
• 大类 : 工程技术 - 3区
• 小类 : 工程:电子与电气 - 3区
Overview
Underwater images and videos play significant roles in developing, exploring, and protecting the underwater world. However, there are many challenges that need to be addressed due to the complex and uncontrollable underwater imaging conditions. Underwater images and videos taken by sensors always suffer from the effects of quality degradation due to light selective absorption and scattering as well as the use of artificial light. The degraded underwater data have low contrast and brightness, color deviations, blurry details, nonuniform bright speck, etc., which not only affects the experience of human perception but also challenges the computer vision algorithms. Despite the prolific work in underwater visual information processing, the related theories are antique and current algorithms remain largely unsatisfactory. Further, it is difficult to achieve decent performance by directly transplanting the computer vision applications (e.g., object detection, recognition, segmentation, etc.) for conventional images and videos into underwater ones. Besides, large-scale real-world underwater benchmark datasets and specialized underwater image and video quality assessment metrics are lacking, which keep this research area at a standstill. Thus, there is a pressing demand for novel theories and algorithms that can effectively deal with the problems of underwater image and video quality degradation, accurately evaluate underwater image and video quality and efficiently compress underwater images and videos, and also call for the various applications of computer vision algorithms in underwater images and videos.
This special issue will feature original research papers related to the theories and algorithms for underwater visual information processing together with widespread applications. The main topics of interest are, but are not limited to:
Underwater optical imaging physical model and quality degradation theory.
Underwater image and video enhancement and restoration algorithms, including traditional methods, physical model-based methods, and deep learning-based methods.
Underwater image and video quality assessment methods, including full-reference assessment metrics, non-reference assessment metrics, etc.
Underwater image and video compression, coding, representation, transformation, etc.
Underwater image and video applications: aquatic robotics visions, underwater machine visions, object detection, object recognition, segmentation, tracking, 3D modeling, etc.
New benchmark datasets related to the aforementioned topics.