Taiwan, known for its diverse agricultural output, is facing the rapid loss of arable land due to urbanization and industrial development. To address this, the Taiwan Agriculture Research Institute (TARI) collaborated with Professor Chi-Kuei Wang from the National Cheng Kung University (NCKU) to utilize advanced remote sensing and artificial intelligence (AI) technologies. The project aimed to accurately assess the current state of arable land and quantify the loss due to construction. Using very high-resolution satellite imagery from the Airbus Pleiades constellation, processed with ENVI software and its Deep Learning module, the research team analysed and identified over 168,000 building structures in agricultural areas. This effort provided a precise calculation of land converted from agricultural use to construction, serving as a basis for future land use policy decisions by the Taiwanese government.