Open AccessUnderwater Object Segmentation Algorithm Based on Depth Information
Author(s) -
Xiaowen Zhu,
Zheyu Hu,
Dawei Tu,
Xu Zhang
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/450/1/012078
Subject(s) - artificial intelligence , computer vision , underwater , segmentation , computer science , image segmentation , process (computing) , object (grammar) , key (lock) , segmentation based object categorization , image (mathematics) , scale space segmentation , geography , computer security , archaeology , operating system
In the process of developing the ocean, the underwater robot's accurate positioning of the object is the key to the success of the underwater mission. Due to the interference of environment and other factors, the traditional GrabCut algorithm cannot accurately locate the target from the image collected by the visual sensor. This paper proposes an image segmentation algorithm based on depth Information, which is improved on the GrabCut algorithm and fuses depth information. First, the foreground area of the original image is extracted by using the depth information of the image to become a new image to be segmented. Then, two interactive operations are conducted on the new image. Finally, GrabCut algorithm is used to obtain the segmentation result of the target. Compared with GrabCut algorithm, the algorithm in this paper is more effective, which can improve segmentation accuracy and target positioning effect.