Evaluating spatial frequency response (SFR) in natural scenes is crucial for understanding camera system performance and its implications for image quality in various applications, including machine learning and automated recognition. Natural Scene derived Spatial Frequency Response (NS-SFR) represented a significant advancement by allowing for direct assessment of camera performance without the need for charts, which have been traditionally limited. However, the existing NS-SFR methods still face limitations related to restricted angular coverage and susceptibility to noise, undermining measurement accuracy. In this paper, we propose a novel methodology that can enhance the NS-SFR by employing an adaptive oversampling rate (OSR) and phase shift (PS) to broaden angular coverage and by applying a newly developed adaptive window technique that effectively reduces the impact of noise, leading to more reliable results. Furthermore, by simulation and comparison with theoretical modulation transfer function (MTF) values, as well as in natural scenes, the proposed method demonstrated that our approach successfully addresses the challenges of the existing methods, offering a more accurate representation of camera performance in natural scenes.