Researchers from the Graduate School, Universitas Padjadjaran—Krisna Ramita Sijabat and Mochamad Candra Wirawan Arief—have developed a sophisticated method for mapping urban slum areas in Bandung by merging remote sensing imagery (RSI) with street view imagery (SVI) using deep learning frameworks. Their study, “Integrating Remote Sensing and Street View Imagery with Deep Learning for Urban Slum Mapping: A Case Study from Bandung City,” demonstrates that combining both data sources with the FCN-DK (Fully Convolutional Network with Dilated Kernel) model significantly boosts accuracy—reaching 86.25%—compared to models using RSI alone. The combined approach also achieved a remarkable 97.93% recall in identifying slum areas when RSI and SVI were jointly employed.

This work directly supports several Sustainable Development Goals (SDGs): it advances SDG 11 (Sustainable Cities and Communities) by offering a more precise, scalable, and cost-effective tool for urban planning and slum upgrading; contributes to SDG 9 (Industry, Innovation, and Infrastructure) through pioneering use of AI and integrated geospatial technologies; and strengthens SDG 17 (Partnerships for the Goals) by demonstrating how academic institutions can collaborate with local governments to provide data-driven solutions that underpin equitable and resilient urban development.