The characterization and assessment of remote sensing image quality is an important research contents in the field of remote sensing. At present, the objective image assessment method in remote sensing engineering application is mainly based on the characterization and evaluation of imaging system performance indicators, such as modulation transfer function (MTF), signal-to-noise ratio (SNR), etc. However, this kind of method ignores the analysis and estimation of image quality from the perspective of subjective interpretation, which restricts the research of image quality assessment method based on image interpretation degree. In recent years, no-reference image quality assessment has attracted increasing attention, and a number of image quality assessment algorithms with visual relevance have emerged. This kind of algorithm uses the NSS (natural scene statistics) characteristics of natural image to evaluate the image quality from multiple dimensions. It has the advantages of fast processing speed and strong versatility. On the one hand, this kind of method provides a new idea for the remote sensing image quality evaluation task, but also opens the limitation of the current MTF index evaluation method in the image content. On the other hand, whether this method can replace the traditional imaging performance index evaluation method in practical remote sensing application remains to be analyzed and demonstrated. This study focuses on the analysis of the correlation between NSS image feature parameters and MTF index in the problem of image definition representation. The results of correlation calculation on two different image test sets show that NSS image feature parameters and MTF index have strong correlation. This conclusion provides theoretical support and guiding significance for the promotion and implementation of no-reference image quality assessment method in engineering application.