Machine Learning has made it possible to do things we only thought was possible by a human. Abilities like detecting and recognizing particular objects from a view are now done by computers through training them. However, the domain to which a machine can recognize objects is limited to how much the user has trained it. This ability can be seen being used by humans in various fields to recognize and identify objects like face detection, color classification, fruits classifications, and many more fields. My project leans towards recognizing monuments and buildings that are historical and religious. Tourists that visit some historical places might not be aware of the building or statues they are observing. So, allowing them to click the picture they are curious to know about and uploading on our application, the application recognizes the object being displayed in the picture and gives the user information on what they are actually looking at. The system is trained with images of such monuments so that when a similar image is shown to it by the user, it can identify that object in the image. Basically, there is an application through which the user can click the pictures and the application detects the object in the image (considering the object to be recognized is trained to the system through supervised learning). Images are classified using certain classification algorithms such as SVM, CNN, and a few others. At first, they are cleansed and augmented through python scripts, and then they are trained in classifier built in TensorFlow Keras. Then, an application is created and the classifier is used by the application to recognize the monuments clicked from the application itself.