This project aims to improve the detection of pneumonia by using convolutional neural networks (CNNs) with generative AI (SRGAN), providing a solution to the global health concern of inaccurate and delayed diagnoses. Our aim is to create an accurate model that efficiently detects pneumonia in medical images. Pneumonia diagnosis is challenging due to the limitations of existing methods, which struggle to identify intricate patterns in chest X-rays, leading to misdiagnosis. We suggest using convolutional neural networks (CNNs) to solve this problem. CNNs are a powerful deep learning method that has shown impressive results in identifying images. VGG16 architecture is used for better accuracy. By leveraging CNNs, we can increase the precision of pneumonia diagnosis and reduce the number of misdiagnoses, ultimately leading to better patient outcomes. Accuracy of the model is increased with the use of SRGAN. SR is the supersampling of low-resolution images to higher resolution while minimizing information distortion. We follow a specific methodology that involves gathering and preparing a wide-ranging chest X-ray dataset. Automated diagnostic tools have the potential to revolutionize pneumonia detection, providing a valuable resource for healthcare professionals.