Detection of Garbage Disposal from a Mobile Vehicle Using Image Processing

Abstract

One of the major causes of dispersal of various diseases is illegal disposal of waste materials. Littering also causes pollution and waste of resources. Therefore, to develop strategies and models to prevent littering is one of the interesting research area. The presented exposition focusses on real-time detection of roadside littering done by passengers in mobile vehicles. The proposed model consists of three major steps, namely, detection of vehicle, frame extraction and its analysis and detection of garbage. In the first step, the targeted vehicle frame detection is done using the Haar-cascade classifier method followed by the detection of a foreign particle motion originating from the vehicle using background subtraction and frame differencing method. Finally, the garbage is detected by calculating and analyzing the nature of curve of the foreign particle obtained from the processing of consecutive images of the target vehicle. Therefore, the experimental results indicate that if a parabolic or linear curve is observed in the motion near the detected vehicle, then it is considered to be the motion caused due to the garbage. A very high recognition rate is observed in the results.