基于多特征信息融合的彩色图像复杂度研究

Color Image Complexity based on Multi-Feature Information Fusion

针对图像复杂度广泛应用却无统一标准的现状,构建基于彩色图像复杂度敏感模型的评价体系。以人类视觉系统为基准,选择颜色熵、能量、对比度、相关性、同质性和边缘比率特征,结合遗传算法和BP神经网络,构造了多特征信息融合的复杂度敏感模型。同时求解各个指标权重,创建图像复杂度与评价体系之间的定量机制。实验结果表明模型符合人眼视觉系统,且能够得到准确的彩色图像复杂度,为基于图像、视频等通信技术提供了有效参考依据。

Aiming at the current situation that image complexity is widely used but there is no uniform standard, an evaluation system based on color image complexity sensitive model is constructed. On the basis of human visual system, the features of color entropy, energy, contrast, correlation, homogeneity and edge ratio are selected, and the complexity-sensitive model of multi-feature information fusion is established by combining genetic algorithm and BP neural network. At the same time, the weights of each index are solved to create a quantitative mechanism between the image complexity and the evaluation system. The experimental results indicate that the model is in line with the human visual system and can obtain accurate color image complexity. It provides an effective reference for image-based and video-based communication technologies.