Detect and Evaluate Visual Pollution on Street Imagery Taken of a Moving Vehicle

Visual pollution is a growing problem in urban areas. It important for environmental management to identify, formalize, measure and evaluate visual pollution. This paper presents study on the development of an automated system classification using street images taken from moving vehicle. The proposed uses convolutional neural networks classify different types pollutants such as graffiti, faded signage, potholes, litter, construction zones, broken poor lighting, billboards, road sand, sidewalk clutter, unmaintained facades.In this study, we utilized large dataset raw sensor camera inputs gathered fleet multiple vehicles specific geographical area. Our aim was develop that simulate human learning these images. successful implementation would be significant contribution planning strengthening communities worldwide. Additionally, it could lead creation "visual score/index" areas, which serve new metric management. findings, present paper, will valuable addition academic community field computer vision applications.