Harmful Algal Bloom recognition based on microscopic images

Author

Zheng, Haiyong ; Ji, Guangrong ; Zheng, Bing ; Zhang, Hao ; Gulliver, T. Aaron

Author_Institution

Dept. of Electron. Eng., Ocean Univ. of China, Qingdao, China

fYear

2011

fDate

23-26 Aug. 2011

Firstpage

852

Lastpage

855

Abstract

Using a microscope to recognize a Harmful Algal Bloom (HAB) requires specialized knowledge and is time-consuming. This method is, however, widely employed by HAB taxonomists. To provide a better solution, a HAB recognition system has been developed which is based on image analysis. This paper presents the architecture of this system, as well as sone performance results. The key component is the libiplhab library based on computer vision libraries (such as OpenCV and LIBSVM), which contain image processing and pattern recognition algorithms suitable for HAB microscope images.

Keywords

computer vision; image recognition; HAB taxonomists; LIBSVM; OpenCV; computer vision libraries; harmful algal bloom recognition; image analysis; image processing; libiplhab library; microscopic images; pattern recognition; time-consuming; Feature extraction; Graphical user interfaces; Image recognition; Libraries; Microscopy; Pattern recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications, Computers and Signal Processing (PacRim), 2011 IEEE Pacific Rim Conference on

Conference_Location

Victoria, BC

ISSN

1555-5798

Print_ISBN

978-1-4577-0252-5

Electronic_ISBN

1555-5798

Type

conf

DOI

10.1109/PACRIM.2011.6033006

Filename

6033006