Illuminating West Africa: Enhancing Electricity Reliability with a Pole Component Detection Dataset

Reliable and sustainable energy access is pivotal to socio-economic development and is an integral part of the United Nations' Sustainable Development Goal (SDG) 7. Despite this, ensuring a consistent electricity supply, especially in regions like West Africa, is a considerable challenge. In response, we introduce a novel dataset for mapping electrical networks using object recognition techniques. We uniquely created our dataset, which comprises 9,633 labeled instances within 872 images, all captured and annotated by the authors. These images cover seven critical object categories related to electricity distribution. The dataset allows for the mapping and visual inspection of physical infrastructure and overhead distribution lines, thereby promoting enhanced maintenance strategies and pinpointing areas in need of augmented public infrastructure investment. This is essential for ensuring service quality and resilience against extreme weather events. Our experiments present the performance of various existing detection algorithms on this self-constructed dataset. In fostering affordable, reliable, and sustainable energy access in West Africa, our work contributes towards the achievement of SDG 7 and can offer valuable insights for related challenges elsewhere. Overall, our research underscores the potential of machine learning and object recognition techniques in resolving critical energy infrastructure issues and propelling sustainable development goals.