Pulmonary diseases are very dangerous and rapidly spreading diseases in Ethiopia. Frequently occurring pulmonary disease are very serious and infectious disease. These are cancer, tuberculosis and pneumonia and have different varieties and detected on the basis of disease features (properties). In sight of this, chest x-ray is one of the key techniques of diagnosis. It has been popular and many researchers are interested in this research area and different approaches have been proposed for the pulmonary diseases and decease classification. A digital image analysis technique was developed to detect different varieties of frequently occurring pulmonary disease based on type of disease. X-ray images are inputs for detection process. Training and sample x-ray image are taken from LC (100), PTB (010) and PN (001) categories. Four morphological features such as area, perimeter, equivalent distance and roundness, and ten texture features such as mean, standard deviation, smoothness, correlation, entropy, contrast, dissimilarity, skewness, homogeneity and energy features vector are extracted from disease x-ray images. The total images taken from the three above mentioned pulmonary diseases are 441 covering the period from 2013 to 2014. Out of these images, 289 image data set are taken from Adinas higher clinic and the rest 152 are taken from Black lion referral hospital. 70% of image data set is used for training, 15% for validation and the remaining 15% images for testing purpose. In this study, naïve Bayes and neural network classifier with sigmoid activation function multilayer perceptron architecture using gradient descent learning method (is the optimization technique which is the mathematical basis for the back-propagation algorithm) are used. The overall system detection accuracy of naïveBayes and ANN are 90.9% and 95.5% respectively. The major reason for misclassification is the effect of the quality and inappropriate labelled image data set. As a recommendation, the researcher should use properly threated pulmonary disease xray image data set having good quality.