AUTOMATED CLASSIFICATION OF ASTRONOMICAL IMAGES

Current efforts to perform automatic image measurement and classification are reviewed. As an example, we discuss the acquisition, calibration, and star-galaxy classification of O- and E-band imagery from POSS I obtained with the Minnesota Automated Plate Scanner (APS). For galaxies with isophotal diameters (mu(B) = 24.5 mss) larger than 25'', it is shown that a variety of two-dimensional photometric a parameter spaces provide a crude segregation of Hubble types. Initial results are presented on the training and testing of two artificial neural networks developed to map input image parameter vectors to an eight-step morphological-type scale.