Accession Number : ADA083094


Title :   On Two-Dimensional ARMA Models for Image Analysis.


Descriptive Note : Technical rept.,


Corporate Author : SOUTHEASTERN MASSACHUSETTS UNIV NORTH DARTMOUTH DEPT OF ELECTRICAL ENGINEERING


Personal Author(s) : Chen,C H


Full Text : http://www.dtic.mil/get-tr-doc/pdf?AD=ADA083094


Report Date : 24 Mar 1980


Pagination or Media Count : 15


Abstract : The two-dimensional autoregressive moving-average (2-D ARMA) models have been useful for image coding and compression, statistical image modeling, image feature extraction and segmentation, texture characterization, and image restoration and enhancement. This paper provides a critical review of the 2-D ARMA models for image analysis. Particular emphasis is placed on restoration of noisy images using 2-D ARMA models. Computer results are presented for comparative evaluation study. Problem areas such as order determination are examined. Although there are shortcomings and unsolved problems it is concluded that the models are very effective linear models for image analysis. (Author)


Descriptors :   *MATHEMATICAL MODELS , *IMAGES , *IMAGE RESTORATION , KALMAN FILTERING , REGRESSION ANALYSIS , CODING , COMPUTER APPLICATIONS , STATISTICAL PROCESSES , SEQUENCES(MATHEMATICS) , GAUSSIAN QUADRATURE


Subject Categories : Human Factors Engineering & Man Machine System


Distribution Statement : APPROVED FOR PUBLIC RELEASE