The work aims to further understand the development process and research hotspots of no-reference image quality assessment and provide reference for subsequent related research. CiteSpace literature visualization software was used to conduct basic analysis, co-citation analysis and Keywordanalysis on 1 712 items of literature retrieved from Web of Science from 2000 to 2021. The development characteristics of no-reference image quality assessment were obtained by analyzing the visualization atlas. The analysis results showed that no-reference image quality assessment was currently in a stage of rapid development, and countries all over the world had made achievements in this field. At present, there were more mature no-reference image assessment algorithms, but there was still a gap in accuracy compared with subjective assessment. In the future, researchers should combine today's artificial intelligence technology to promote the transition from high-speed development to high-quality development of no-reference image quality assessment.