Automated sizing,counting and identification of zooplankton by pattern recognition |
| |
| Authors: | H P Jeffries M S Berman A D Poularikas C Katsinis I Melas K Sherman L Bivins |
| |
| Institution: | (1) Graduate School of Oceanography, University of Rhode Island, 02882-1197 Narragansett, Rhode Island, USA;(2) Department of Electrical Engineering, University of Rhode Island, 02881 Kingston, Rhode Island, USA;(3) National Marine Fisheries Service Laboratory, 02882 Narragansett, Rhode Island, USA;(4) National Oceanic and Atmospheric Administration, 20852 Rockville, Maryland, USA |
| |
| Abstract: | A computerized system was developed to automate the analysis of zooplankton samples. Classification to major taxonomic group was based on discriminant analysis of morphological features. Images were generated either from preserved organisms or from silhouette photographs. The latter technique simplified large-scale sample storage. Accuracy of correct classification, among organisms regularly occurring in New England coastal waters, exceeded 90%. Critical problems were due to limitations inherent to the imaging of low contrast, randomly oriented objects by a vidicon camera. One solution would utilize an incoherent-to-coherent transducer in a binocular field of observation through which plankton entrained in a flowing medium passed.This work was supported in part by NOAA grant NA80AA-4 00023 |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
