Continent Based Geo-Landmark Detection and Recognition Using CNN

Continent Based Geo-Landmark Detection and Recognition Using CNN

Prof. Priyanka Deshmane1, Suyog Palav2, Tejas Pichad3, Sanket Shelar4

1,2,3,4Department of Computer Engineering

1,2,3,4University of Mumbai, Maharashtra, India

priyanka.deshmane8218@gmail.com1, suyogpalav5@gmail.com2, tejaspichad2002@gmail.com3, sanketshelar098@gmail.com4.

ABSTRACT

In the modern world, we need the development of an automated system that can accurately detect landmarks and provide essential services is important. The 'Continent-based Geo-Landmark Detection and Recognition using CNN" project develop the Continent-based Landmark Detection system using deep learning, specifically Convolutional neural network (CNNs). The system will be capable of identifying landmarks from uploaded images. It will provide a short summary about the identified landmarks, and extract geographic data such as landmark address, latitude and longitude and display them on the map. Additionally, the system will utilize location information (i.e. Latitude and Longitude) to locate 17 essential services such as Road Network, Hospitals, Bus Stops, Railway Station, ATM's, Religious Institutions (Temple, Masjid, Church and more) etc.  around the identified landmark within a specified radius, such as 1000 meters (i.e. 1 kilometer).

Keywords: CNN; Continents; Geo-landmark; Detection and Recognition; Convolution neural network; Deep Learning; Python.