水稻病害图像无参考质量评价方法研究

The Research of Digital Image Quality Evaluation Method for Rice Diseases

无参考图像质量评价是一个比较新的研究领域,在这一领域中如何选择好的图像以提高模式识别系统的识别效率是一个重要的研究内容。为此,以水稻病害图像为研究对象,在借鉴传统图像质量评价方法的基础上,分析了影响图像质量的部分因素;同时提出了综合像素空间边缘分布、颜色分布、色相计数、模糊度、清晰度和对比度几个方面的整体质量评价方法。实验证明:利用本文提出的方法所得出的图像质量评估得分与模式识别系统的识别结果相一致,并能显著提高模式识别系统的识别效率,为农作物病害图像质量评价提供了一种新思路。

No-reference image quality evaluation is a relatively new area of research.In this field,how to choose a good image is an important research content for enhancing the efficiency of pattern recognition system.In this paper,rice diseases images are used as an object of study,based on the traditional evaluation method of image quality,the thesis analyzes some factors that affect the quality of digital image and proposes an comprehensive evaluation method combined spatial distribution of edges,color distribution,hue count,blur,contrast and brightness.Experiments prove that image quality assessment scores marked from the method proposed in this paper are consistent with results got from pattern recognition system and the efficiency of pattern recognition system is significantly increased.It means that the proposed approach is effective.