Wavelet Transform Based Pneumonia Classification of Chest X- Ray Images

Pneumonia is a contagious disease that inflames the sac of air in either of our lungs. It occurs mainly due to microbial infection. In proposed work, we have used normal and abnormal (infected with pneumonia) data set of chest x-ray for detection of pneumonia. With Wavelet Transform method and Gray level Co- Occurrence Matrix (GLCM) , total 6 time-frequency domain features are extracted for approximately 3000 X-Ray images of data set of normal and infected subjects.GLCM is mainly used for extracting texture features like contrast, energy, correlation & homogeneity. Then we form feature matrix and train our classifier. Support Vector Machine (SVM) and K-nearest neighbors (KNN) scheme is adopted to detect the pneumonia. There are various type of SVM and KNN based on Kernel. From these, SVM and KNN, we achieved accuracy 94.6% with cubic SVM and 92.6% with weighted KNN. The results of using the Wavelet Transform based SVM classification are quite promising when compared to another classification techniques as K-NN classification.