Cleanliness plays a crucial role in developing smart cities and cleaning garbage in urban areas has become a challenge to local governing bodies. To tackle the situation, we propose an urban street cleanliness assessment approach in three phases using advanced technology i.e., Mobile Edge servers, R-CNN Deep learning & assessment of the approach. Initially, high resolution cameras will be installed on the vehicles to collect the street images. Mobile edge servers are used to store and extract street image information temporarily. Secondly, the primary data collected from Mobile Edge servers were transmitted to the cloud data center for thorough analysis via available city networks. Also, Faster Region-based Convolutional Neural Network is used to identify the strategically located street garbage locations & categories. The results will help the city managers to arrange necessary clean-up personnel effectively & efficiently.