激光与光电子学进展, 2019, 56 (8): 081008, 网络出版: 2019-07-26
真彩微光夜视图像融合算法 下载: 1124次
Image Fusion Algorithms for True Color Low Light Level Night Vision
图像处理 夜视 真彩微光夜视 融合算法 客观评价 图像融合质量 image processing night vision true-color low-light-level night vision fusion algorithm impersonal evaluation image fusion quality
摘要
简要阐述了基于全波与三波段的真彩色微光夜视系统原理,结合图像融合的一般算法,研究了加权平均法、基于线性变换增强的Brovey法、 HIS(色调、亮度和饱和度)空间法及基于边缘分割的HIS法4种真彩微光夜视图像融合算法,详细阐述了融合算法的实现方法和过程。研究结果表明,利用基于线性变换增强的Brovey法,得到场景一和场景二的融合图像的综合客观评价指标值,分别为27.9647、31.2756,均大于其余3种算法得到的值。在这4种融合算法中,由基于线性变换增强的Brovey法得到的融合图像视觉效果最优。
Abstract
The principle of a true color low light level night vision system based on full-wave and three-wave bands is briefly described. Combining with the general image fusion algorithms, four image fusion algorithms for low-light-level night vision, including the weighted average method, the Brovey method based on linear transformation enhancement, the HIS (hue, intensity and saturation) space method, and the HIS method based on edge segmentation are studied. The realization methods and processes of these fusion algorithms are described in detail. The research results show that by the Brovey method based on linear transformation enhancement, the comprehensive objective evaluation indexes of fusion images for Scene 1 and Scene 2 are 27.9647 and 31.2756, respectively, larger than those by the other three algorithms. Among these four fusion algorithms, the visual effect of fusion images obtained by the Brovey method based on linear transformation enhancement is the best.
蒋云峰, 武东生, 黄富瑜. 真彩微光夜视图像融合算法[J]. 激光与光电子学进展, 2019, 56(8): 081008. Yunfeng Jiang, Dongsheng Wu, Fuyu Huang. Image Fusion Algorithms for True Color Low Light Level Night Vision[J]. Laser & Optoelectronics Progress, 2019, 56(8): 081008.