基于可能性分布联合落影的红外图像融合算法选取

针对双模态红外图像实际融合需求中常涉及到异类差异特征协同优化融合,且现有差异特征属性无法针对性地调整融合算法进行有效驱动,导致融合效果差等问题,提出了基于可能性分布联合落影的双模态红外图像融合算法选取的方法。首先计算双模态红外图像差异特征幅值的融合有效度,利用K最近邻法得到差异特征幅值的概率密度分布,得到差异特征频次属性的分布;然后通过差异特征幅值属性与频次属性构造差异特征权重函数,建立异类差异特征权重函数与多融合算法间的可能性分布合成,得到异类差异特征权重函数多融合算法融合有效度的联合落影;最后构建融合性能指标动态选取最优融合算法。实验结果表明,本文方法所选出的最优融合算法在等级得分指标上明显优于其他算法,验证了本文将可能性分布联合落影运用于双模态红外图像最优融合算法选取中具有可行性。

Infrared Image Fusion Algorithm Selection Based on Joint Drop Shadow of Possibility Distributions

Aiming at the actual fusion needs of heterogeneous difference features collaborative optimization are often involved for dual-mode infrared images fusion,and the existing difference feature attributes cannot be targeted to adjust algorithms to drive fusion effectively,resulting in poor fusion effect,a method of infrared image fusion algorithm selection based on joint drop shadow of possibility distributions is proposed.Firstly,the fusion effectiveness of the difference feature amplitude of the dual-mode infrared image is calculated,and the probability density distribution of the difference feature amplitude and the distribution of the frequency attribute based on K-nearest neighbor method is obtained.Then the difference feature weight function through difference feature amplitude attribute and frequency attribute is constructed,and the possibility distribution synthesis between the heterogeneous difference feature weight function and fusion algorithms is established to obtain the joint drop shadow of the fusion effectiveness of heterogeneous difference feature weight function and the fusion algorithms.Finally,the fusion performance index is constructed to select the optimal fusion algorithm dynamically.The experimental results show that the optimal fusion algorithm selected by this method on the ranking score index is significantly higher than other algorithms,which the feasibility of applying joint drop shadow of possibility distributions in the optimal selection of dual-mode infrared images fusion algorithm is verified.