A Computer Assisted Recognition System for Cotton

Joe Ellington and Tracey Carrillo


A computer assisted recognition system designed to assist an operator in classifying and identifying arthropods is currently being developed. The complete system runs in a windows environment on a personal computer with a VGA monitor supporting 24 bit color. Images are input to the system using tagged image file format (TIFF) files. The system segments the files into individual objects and attempts to identify each one automatically. If unable to identify the objects, the system will present the image to allow the operator to identify the object. The system is trained by loading TIFF images of multiple objects of a known sample. The data from the samples are used to generate a statistical data base which is used as a reference for future identification. This system is currently being designed to identify insects in the agricultural environment in order to maximize sampling efficiency.

Reprinted from Proceedings of the 1994 Beltwide Cotton Conferences pp. 1026 - 1028
©National Cotton Council, Memphis TN

[Main TOC] | [TOC] | [TOC by Section] | [Search] | [Help]
Previous Page [Previous] [Next] Next Page
Document last modified Sunday, Dec 6 1998