data&code

A very small number of former researches on the accessibility of public facilities focused on sports facilities. Besides, detecting sports facilities and optimizing the distribution of sports facilities are rarely involved. This study addresses the challenge of enhancing the accessibility and equality of outdoor sports facilities through a novel framework based on multisource sensing data. It aims to expand the way of obtaining outdoor sports facilities AOI through deep learning object detection methods using remote sensing imagery, and bridge the gap in current research that often overlooks equality in optimizing the accessibility of sports infrastructure. By harmonizing the expansion of accessibility with the imperative of equality and balancing the use of existing sports facilities against the need for new developments, this framework proposes a novel approach to optimizing the spatial accessibility of outdoor sports amenities. The study identified outdoor sports facilities in Shanghai using advanced deep-learning techniques on remote sensing data. It then developed a greedy heuristic algorithm based on the Gaussian Two-Step Floating Catchment Area method and Gini coefficient analysis for evaluating and optimizing facility accessibility and fairness. The recognition method achieved a precision and recall rate of 88% and 96%, respectively. Notably, the optimization efforts resulted in a 73% increase in accessibility while significantly reducing the Gini coefficient from 0.58 to 0.34, demonstrating the framework's effectiveness in ensuring equal access to sports facilities.