An expert system for the computer-aided identification of pollen grains is described. The system consists of a computer program writeen in Pascal and a knowledge base containing the description of a large number of pollen grains. The operator is guided by the progam to perform a whole series of microscope observations of the grain which lead to a final decision concerning the identity of the pollen: family, and eventually genus and species if clearly recognizable. The structure of the logical process is particulary simple so that no expertise in computer programming is required for routine use. The system is designed not only for research objectives but mainly for use in operational environments such as, for example, allergological monitoring. The knowledge base has initially been constructed to include about 100 individuals of European, mostly Mediterranean, flora with special emphasis on aerobiological factors. The system is, however, entirely flexible, being conceived to recognize pollen of different types together with other biological objects such as spores and bacteria. The software has been designed as a self-contained package to be installed and run on IBM-PC compatible computers with MS DOS operative system.
POLKEY: an expert system for the identification of pollen grains
1989
Abstract
An expert system for the computer-aided identification of pollen grains is described. The system consists of a computer program writeen in Pascal and a knowledge base containing the description of a large number of pollen grains. The operator is guided by the progam to perform a whole series of microscope observations of the grain which lead to a final decision concerning the identity of the pollen: family, and eventually genus and species if clearly recognizable. The structure of the logical process is particulary simple so that no expertise in computer programming is required for routine use. The system is designed not only for research objectives but mainly for use in operational environments such as, for example, allergological monitoring. The knowledge base has initially been constructed to include about 100 individuals of European, mostly Mediterranean, flora with special emphasis on aerobiological factors. The system is, however, entirely flexible, being conceived to recognize pollen of different types together with other biological objects such as spores and bacteria. The software has been designed as a self-contained package to be installed and run on IBM-PC compatible computers with MS DOS operative system.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.