The automated detection of cyanobacteria using digital image processing techniques

The aim of this paper is to show how the process of detection of cyanobacteria can be automated by means of a digital image processing package. In Britain's lakes and reservoirs, the occurrence of nine different species of cyanobacteria is of interest. This paper determines a way to detect and...

Full description

Saved in:
Bibliographic Details
Published in:Environment international 1995, Vol.21 (2), p.233-236
Main Authors: THIEL, S, WILTSHIRE, R. J
Format: Article
Language:English
Subjects:
Algae
Applied sciences
Automation
Bacteria
Continental surface waters
Exact sciences and technology
Freshwater
Lakes
Natural water pollution
Pollution
Q1
Water pollution
Water treatment and pollution
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The aim of this paper is to show how the process of detection of cyanobacteria can be automated by means of a digital image processing package. In Britain's lakes and reservoirs, the occurrence of nine different species of cyanobacteria is of interest. This paper determines a way to detect and distinguish two of them, Anabaena and Oscillatoria, automatically. Water samples containing cyanobacteria were obtained and examined under a microscope. Through an attached camera, microscopic pictures of algae were recorded on a videotape and transferred to the computer, a SUN Workstation. Image processing techniques were applied to the image to improve the quality and enhance particular features. In particular, the Logarithm of the Gaussian (LoG) operator proved to be effective for the enhancement of biological organisms. The shape and texture were analyzed and differences in the algae were obtained using features given in the biological key. With the aid of the textural features, it was possible to distinguish between Anabaena and Oscillatoria with an accuracy of over 90%.
ISSN:0160-4120
1873-6750
DOI:10.1016/0160-4120(95)00013-5