Abstract: A new approach based on quantum theory for the decomposition of image was proposed in this paper. An image can be decomposed into a series of characteristic sub-images according to the relation between the pixels built by means of a superposition of states in quantum mechanics in this approach. The characteristics of these sub-images were analyzed and a rule used to generate an operator for the detection of image edge was given on the basis of this. A new algorithm for image edge detection was proposed with the rule. The computer simulation results show the effectiveness of the rule and the superiority of the algorithm proposed here.

Key words: quantum-inspired, image decomposition, characteristic sub-image, algorithm production rule, edge detection

摘要： 针对数字图像的分解问题，研究了一种基于量子理论的新分解方法,该方法借鉴量子叠加态建立图像像素之间的关联,将一幅数字图像分解为一系列特征子图的叠加。分析了特征子图的意义，并在此基础上给出了一种边缘检测算子的生成规则。利用该规则构造的一种新的图像边缘检测算法被提出，计算机仿真结果验证了该规则的有效性和提出的边缘检测算法的优越性。

关键词: 量子衍生, 图像分解, 特征子图, 算法生成规则, 边缘检测