Macroalgae recognition based on histogram oriented gradient

Tan, C.S. and Lau, P.Y. and Low, T.J. (2017) Macroalgae recognition based on histogram oriented gradient. Lecture Notes in Electrical Engineering, 398. pp. 257-266. ISSN 18761100

Full text not available from this repository.
Official URL: https://www.scopus.com/inward/record.uri?eid=2-s2....

Abstract

The distribution of the marine algae is an important indication of the biodiversity changes in the aquatic ecosystem which algae biologist normally monitors. One of the most prominent instances is the monitoring of the invasive alga, e.g. Caulerpa taxifolia, through conducting regular surveys. It usually involves highly trained algae biologist to annotate the obtained video in order to detect the location where the alga would be likely present within survey area. This may constitute to a lengthy and demanding task which could be prone to observer-induced error. Hence, a framework is proposed herein to automate the analysis of underwater image to deduce if it contains the targeted alga, which is Caulerpa taxifolia. The framework employed HOG feature descriptor for object detection. Its efficiency and reliable was verified by the experiments using our consolidated database. © Springer Science+Business Media Singapore 2017.

Item Type: Article
Additional Information: cited By 0; Conference of 9th International Conference on Robotic, Vision, Signal Processing and Power Applications, RoViSP 2016 ; Conference Date: 2 February 2016 Through 3 February 2016; Conference Code:184869
Uncontrolled Keywords: Algae; Aquatic ecosystems; Biodiversity; Computer vision; Gallium alloys; Graphic methods; Image analysis; Object detection; Object recognition; Robotics; Surveys, Caulerpa taxifolia; Feature descriptors; Its efficiencies; Macro-algae; Marine algae; Object classification; Object location; Oriented gradients, Signal processing
Depositing User: Mr Ahmad Suhairi UTP
Date Deposited: 09 Nov 2023 16:21
Last Modified: 09 Nov 2023 16:21
URI: https://khub.utp.edu.my/scholars/id/eprint/9421

Actions (login required)

View Item
View Item