基于复杂性测度的零件序列图像融合方法研究

通过分析零件序列图像的特点,研究了基于C0复杂性测度的图像融合方法,解决了传统图像融合方法(如空间域、变换域方法)都是将非线性的二维图像信号近似视为线性系统的局限性.仿真结果表明,采用该方法,能在融合性能的主观评价和客观评价等指标上,都优于目前普遍使用的图像融合方法,表明其具有广泛的应用前景.

Study of parts sequence images fusion based on complexity measure

A new image fusion method based on the Co complexity measure was proposed by analyzing the property of parts sequence images. Image fusion methods, including spatial and transforming domain method, assumed the two-dimension image as a linear system. Thus the new method solved this problem in a way derived from the linearity scientific field, ignoring the fact that the linear method dealing with the nonlinear two-dimensional image signal is just a maximum likelihood approach. This new method is proved to be promised by the experiment by the fusion performance in the subjective evaluation and objective evaluation indicators. It is superior to the current widespread use of the image fusion method, demonstrating the broad application prospects.