Vision-Based Parking Lots Management System Using an Efficient Adaptive Weather Analytic Technique

With rapid race in this era of technologies, Smart Parking in urban areas (Smart Cities) is becoming a big concern for its importance from environmental and economic aspects. Smart parking systems play an important role to improve the city life in terms of reducing the air pollution, gas emission, traffic congestion; it alleviates the time consuming for motorists to easily allocate available parking space lots. This project will provide a Vision-Based approach for easily detection and classification of empty and non-empty parking lots. Firstly, prelocalized coordinates configuration will be done to determine the places of vehicle parking lots. Secondly, an adaptive weather analytic technique that depended on first statistics (i.e., the mean & variance) criteria was adopted in this work to measure and compensate the weather condition changes using information of RGB chromatic channels. Finally, parking spaces inspection technique is used that based on some pixels of empty road and parking areas to detect and classify vehicle occupancy in these lots. The proposed system applied on some common parking standard dataset (PKLot) samples; and the results were promising under different scenarios of weather conditions.