Establishment of a Conceptual Framework to Detect Automatically the Urban Planning Offences Using Deep Learning and RGB orthophotos: Study Case of Morocco

Unregulated Housing (UrH) is a widespread urban phenomenon in Morocco, largely driven by rapid population growth and accelerated urbanization. It has expanded mainly on the outskirts of cities and within housing developments that already benefit from basic infrastructure and superstructure services. In response to this challenge, public authorities have adopted several urban planning instruments, particularly the Land Management Plan (LMP). According to Law No. 12-90 on urban planning, the LMP seeks to regulate urban expansion, improve the architectural and aesthetic quality of the built environment, and preserve the overall coherence of developed areas. As a legally binding planning document, the LMP establishes strict land-use regulations, and any breach of these rules constitutes an offence. Traditionally, detecting such violations requires on-site inspections by control officers, followed by the preparation of official reports submitted to the competent legal authorities. However, recent advances in aerial image acquisition and processing technologies provide powerful tools to improve and facilitate the monitoring of urban planning compliance. This paper proposes a conceptual framework that integrates artificial intelligence with urban planning regulations to enable the automatic detection of urban planning offences using RGB orthophotos covering areas subject to a Land Management Plan, relying on deep learning techniques.