Given training sample, the problem of classifying Gaussian spatial data into one of two populations specified by conditional autoregressive model (CAR) with different mean functions is considered. This paper concerns with classification procedures associated with Bayes Discriminant Function (BDF) under deterministic spatial sampling design. In the case of complete parametric certainty, the overall misclassification probability associated with aforementioned BDF is derived. This is the extension of the previous one to the CAR case. Spatial weights based on inverse of Euclidean distance and the second and third order neighbourhood schemes on regular 2-dimensional lattice are used for illustrative examples.
The effect of the spatial sampling design, Mahalanobis distances and prior probabilities on the performance of proposed classification procedure is numerically evaluated.