Integration of Ai in Gis for Identifying and Locating Illegal Waste Deposits in Algerian Municipalities

Like many municipalities around the world, Algerian municipalities are faced with the challenge of managing the collection of illegal waste deposits located across their territories. These deposits may occur both at authorized household waste collection points and at unauthorized locations. The management of these deposits presents a challenge, as their handling is incompatible with refuse trucks typically used for household waste collection or because they must be covered by collection services other than those provided by the municipality. It is therefore essential to identify and locate these deposits to ensure appropriate handling. This article aims to address this issue through an innovative solution that integrates artificial intelligence (AI) into geographic information systems (GIS). The method is based on transfer learning combined with MobileNetV2 to generate a classification model for images of illegal waste deposits at authorized and unauthorized points. This model is integrated into a plugin created with QGIS software to perform image classification, enabling the location and identification of these deposits. The model achieved an accuracy of 98% during training, and its application to images from Biskra municipality illustrates its potential effectiveness. Beyond this case study, the approach offers a scalable and adaptable solution for improving illegal waste deposit management practices in diverse municipal contexts.