As a Schmidt AI in Science Fellow at the University of Michigan, Imran aims to harness the power of artificial intelligence (AI) and satellite data to illuminate one of the least understood components of the ocean economy – small-scale coastal fisheries. These fisheries sustain over a billion people worldwide but remain largely invisible to global monitoring systems. At the same time, they operate in the very coastal zones where endangered marine species, such as dolphins, turtles, and sharks, are most abundant. Understanding where and how these activities overlap is critical for protecting both livelihoods and biodiversity.
Imran’s research integrates machine learning, computer vision, and geospatial modelling to develop algorithms capable of detecting and classifying small fishing vessels from very high-resolution satellite imagery. By combining these detections with environmental and ecological data, his work will generate the first global, high-resolution maps of coastal fishing activity and its overlap with endangered wildlife. Critically, this effort will generate two novel and actionable findings relevant to global fisheries and conservation: (1) the most comprehensive quantification of small-scale fisheries that, for the first time, will enable their ecological and economic importance to be understood at the global scale, (2) hotspots of fishing and wildlife overlap that are of maximal priority for conservation efforts.