KAPPA:: A fully automated counter of pollen grains.

With the advent of high-speed general purpose digital computers it is becoming possible to perform mathematical or algorithmic processes on pictorial data from image of very high quality. One of the rapidly emerging areas of pattern recognition has been medical image processing, mainly in radioisotope scanning[1], chromosome analysis [2], classification of blood cells [3] and clinic diagnostics areas. The fully automated classification and count of pollen grains using pattern recognition techniques, is the subject of our study. Image analysis using the eyes has always played an important role in the medical and biological science, so in diagnostic subjects as in others fields like blood cells or bacteria computation. A Fully Automated Pollen Counter was conduced using a standard analysis system of several textural features obtained from co-occurrence level gray matrix. In order to determine if these features could be used to differentiate between those types a PCA analysis was performed. The goal of present work was to design and implementate of a low cost image (KAPPA) analysis and recognize fully automated system of pollen grains in the air and other media, e.g., honey, based in pattern recognition techniques applied to digital images obtained through optician microscope. This system presents two different applications: a) Biomedical: An easy and low cost way to determine the amount of pollen grains in the air and other media. b) Investigation. A powerful tool to search possibilities in the fields of palaeography and palinology.