 This study proposes a new enhanced kernel convolution function to reduce overfitting in lung disease detection and diagnosis using chest x-ray or CT scan images. The proposed approach uses convolutional neural network, CNN, models to extract features from the common layers, which are then used as input for multiclass classification of pneumonia and COVID-19. The results show that the improved support vector machine, SVM, classifier fed with CNN features has a success rate of 99.8%. This article was authored by Ushrani Bamavarapu, Nellini Chintalopiti, and Gopi Badanini.