This project is mainly focus to detect the abnormality of the lung tissue, because there are above 150 type of lung tissue disorder occurred. To reduce this disorder, we have first determined the abnormality of the lung tissue and then the appropriate treatment is given by the clinical work. So some of the existing methods is implemented to identify the defect by using the imaging system. In this proposed method, the rotation-invariant Gabor-local binary patterns (RGLBP) texture descriptor and multi co-ordinate histogram of oriented gradients (MCHOG) is an algorithm is used to extract the feature from the acquired HRCT image. To obtain the better image quality, as well as the better result, the dataset image is taken for the detection of the abnormality. The image is cropped from the different location of the lung tissue, so that the state of abnormality is classified by using the SVM classifier. In this process, the SVM classifier is used to classify the state of the abnormality by classifying the test features and the train features. In the classification technique, the test feature and the trainfeature plays the major role for the classification purposes. The test features is nothing but the features which is obtained from the test image, then the trainfeature is nothing but the feature which is obtained from the train image. The train image is the image which is located in the dataset image otherwise the database image. The features of the LBP, Gabor, and the HOG features are combined together for the both test feature as well as the train features. By the combination of those three features, the better accuracy of the classification is obtained.