Because of the progressive increase in human activity, the world's wildlife is critically threatened by the transformation of natural ecosystems into artificial substrates. The precise location of wilderness places is essential for ecological management and biodiversity preservation. There are currently few methods for identifying national scale wilderness regions by utilizing the emerging, high-precision data and taking into account regional-scale geographic characteristics when building index models. This study develops a methodology for recognizing wilderness areas from the standpoint of how human activity affects the environment. First, it takes into account the micro-perspective characteristics of how human activities respond to a complex geographical context and develops a multi-factor integrated calculation model of the intensity of human unsustainable activities. Second, we establish indexes for artificial environment and the impacts of light pollution by introducing novel data sources and developing spatial feature extraction algorithms to identify zones of long-term human activity. Final, the research indicates the possible spatial extent of wilderness regions and provides a quantitative assessment of their wilderness degree. The findings demonstrate that: (1) The wilderness recognition model based on human activities can successfully simulate the distribution of wilderness degree in China, with a spatial pattern of high in the west and low in the east. (2) China's wilderness area accounts for 22.47% of the total land area, primarily distributed in the Great and small Khingan Mountains, Qinghai-Tibet Plateau, the Inner Mongolia Plateau, Tarim Basin, and other regions. (3) This paper's coincidence rate with wilderness data based on the human footprint approach is greater than 80%, demonstrating strong identification capacity and offering reference recognition findings for certain locations.