To address the prevalent social and psychological health issues stemming from the inadequate expression of architectural emotion, an empirical study was conducted on the visual perception characteristics of the rural landscape in Linjiahe Village, Southern Hebei Province, leveraging eye-tracking technology. By establishing an evaluation system encompassing four key elements: roads, boundaries, nodes, and landmarks, and adopting a methodology that integrates subjective and objective analyses, the study identified notable issues in the current rural landscape, including the absence of a visual focus and attention dispersion. Importantly, optimized design was found to substantially enhance participants’ visual attention in critical scenes. Furthermore, the research unveiled a clear interval effect between spatial feature parameters, such as green vision ratio and enclosure degree, and visual perception outcomes. Specifically, optimal visual perception was achieved with a green vision ratio ranging from 20% to 40%, an enclosure degree of 50% to 60%, an openness of 12% to 26%, and a visual discriminability of 10% to 31%. This discovery offers a quantitative evaluation approach and optimization strategies grounded in visual perception for rural landscape design.