基于边界的图像质量评价方法

图像边界是图像质量评价的主要因素。提出了一种结合图像边界的各向异性和奇异性等因素定义边界强度，利用边界的正规性定义边界保持度，然后结合人眼视觉的掩蔽效应和凸效应来建立数学模型，以此度量图像质量。实验结果表明，该方法在对模糊图像和对图像边缘、纹理要求高的领域，其质量评价优于现有的评价方法。

Image Quality Evaluation Method Based on Boundary

The image boundary is the main factors of image quality assessment. This paper proposes a combination of image boundary anisotropic and singularity factors such as to define the boundary strength, at the same time using the boundary keeping degree of normality to define the boundary, and then combined with the human visual masking effect and convex effect to establish the mathematical model, a measure of the quality of the image. Experimental results show that the fuzzy image and the image edge, texture demanding field, its quality is better than that of the existing evaluation methods.