In traditional computer vision tasks, deep learning has shown remarkable results. Classifying bacteria is important in medical field since it helps with the diagnosis and treatment of many different types of ailments. Image classification algorithms have not usually been used in the traditional procedures used by clinical specialists to classify data. Manually classifying bacteria requires a lot of human labour and takes a long time. It is now possible to identify microbes using novel machine learning algorithms that operate on computers. The deep neural network (DNN) is one such exciting technology that is widely employed for image classification. Deep Learning is now widely employed in many image processing jobs and is acknowledged as a potent feature-extraction tool to efficiently address nonlinear challenges. This research uses Convolutional Neural Network (CNN), one of the DNN versions and an efficient approach for classification problems, to categorize microorganisms. In this study, it uses deep learning techniques to the problem of bacterial image classification. It uses CNN model ResNet-50 to classify bacterial images into twenty groups that are notable in medicine. The performance is measured in terms of accuracy parameter. The results help to identify the superiority of proposed model as compared to conventional model.