基于显著性映射与兴趣点凸壳的图像融合算法

量子电子学报

基于显著性映射与兴趣点凸壳的图像融合算法

王鑑航1，张广宇2，马明金1

1吉林交通职业技术学院，电子信息学院，吉林长春 130012; 2长春理工大学，国有资产管理处，吉林长春 130022

出版日期:2017-09-28 发布日期:2019-06-13
通讯作者: 王鑑航(1973-)，吉林长春人，硕士，高级实验师, 研究方向为计算机图像、计算机应用。 E-mail:wangJHanG1973jt@163.com
作者简介:王鑑航(1973-)，吉林长春人，硕士，高级实验师, 研究方向为计算机图像、计算机应用。 E-mail: wangJHanG1973jt@163.com
基金资助:
Supported by Science and Technology Department of Jilin Province (吉林省科技厅资助项目, 20140101206JA)

Image fusion algorithm based on saliency mapping and interest Points convex hull

Wang Jianhang1 Zhang Guangyu2 Ma Mingjin1

1 School of electronic information, Jilin Communications Polytechnic, Changchun 130012, China; 2 Department of State-owned assets management, Changchun University of Science and Technology, Changchun 130022, China

Published:2017-09-28 Online:2019-06-13

摘要/Abstract

摘要： 提出了一种基于显著性特征的可见光与红外图像融合算法来改善目标的融合质量。引入显著检测器对红外图像进行处理，生成显著映射；通过进一步分析红外图像并检测兴趣点，提取图像中的显著兴趣点；通过计算显著兴趣点的凸壳确定显著区域；利用显著兴趣点凸壳对初始显著映射进行优化，使目标定位更加精确。根据区域映射获取可见光图像的背景区域；根据不同的融合准则对目标、背景区域进行融合，获得最终的融合图像。结果表明与当前可见光图像融合技术相比，所提算法在标准差、联合熵与边缘信息因子等指标方面具有优势，其融合图像的细节纹理更清晰。

关键词: 图像处理, 图像融合, 显著性映射, 兴趣点凸壳, 融合准则

Abstract: A visible and infrared image fusion algorithm based on saliency features is proposed to improve fusion quality of objects. The saliency detector is introduced to process the infrared image for generating the saliency map. The salient interest points are extracted in the image by further analyzing the IR image and detecting interest points. The saliency areas are identified by calculating the convex hull of salient interest points, and the target positioning becomes more accuracy by using salient interest points to optimize the initial salient mapping. The background area of visible light image is obtained according to the region mapping, and the final fusion image is obtained by using different fusion rules to fuse the target and background regions. Results show that this algorithm has advantages with clearer detail texture of fusion image in standard deviation, joint entropy and edge information factor indexes compared with the common visible image fusion technique,.

Key words: Image processing, Image fusion, Saliency mapping, Interest points convex hull, Fusion criterion

王鑑航1，张广宇2，马明金1. 基于显著性映射与兴趣点凸壳的图像融合算法[J]. 量子电子学报.

Wang Jianhang1 Zhang Guangyu2 Ma Mingjin1 . Image fusion algorithm based on saliency mapping and interest Points convex hull[J]. Chinese Journal of Quantum Electronics.

参考文献

[1] Zhang Lei, Jin Longxu, Han Shuangli. Fusion of infrared and visible images based on non-sampled Contourlet transform and region classification[J]. Optics and Precision Engineering（光学精密工程）, 2015, 23 (3): 810-818（in Chinese）. [2] Zambotti Villela L, Yamasaki S C， Villarroel J S, et al. Novel fusion method for visible light and infrared images based on NSST-SF- PCNN [J]. Infrared Physics and Technology, 2014, 65(7): 103-112. [3]Zhou Qaing, Zhao Jufeng, Feng Huajun, et al. Infrared polarization image fusion with non-sampling shearlets[J]. Journal of Zhejiang University (Engineering Science Edition) (浙江大学学报(工学版)), 2014, 48(8): 1508-1516（in Chinese）. [4] Zheng Hong, Zheng Chen, Yan Xiusheng. Visible light and infrared image fusion algorithm based on shear wave transform [J]. Chinese Journal of Scientific Instrument(仪器仪表学报), 2015, 33 (7): 1613-1619（in Chinese）. [5] Li Changxing, Wang Zhiqiang. Infrared and visible image fusion based on visual features[J]. Journal of Xi'an University of Posts and Telecommunications(西安邮电大学学报), 2016, 21 (6): 98-102(in Chinese). [6]Jin Haiyan, Xia Qian, Wang Yanyan. Fusion of visible and infrared images using multi-objective evolutionary algorithm based on decomposition [J]. Infrared Physics and Technology, 2015, 7(1): 151-158. [7]Bavirisetti D P, Dhuli R. Two-scale image fusion of visible and infrared images using saliency detection[J]. Infrared Physics and Technology, 2016, 76(10): 52-64. [8] Ni Weiping, Bian Hui, Yan Weidong. Sequence image mosaic based on interest point detection and affine transform model [J]. Laser and Infrared（激光与红外）, 2014, 39 (3): 330-334（in Chinese）. [9] Tian Juanjuan. Research on Image Retrieval Based on Interest Point Oriented Gradient Histogram(基于兴趣点方向梯度直方图的图像检索技术研究)[D]. Shanxi: Taiyuan University of Technology, 2015: 23-35. [10]Liu Xiaonan, Luo Yuwen, Yin Silu. Image quality assessment based on saliency detection[J]. Biometric Recognition, 2016, 39(2): 349-356. [11] Garrigues M, Manzanera A, Bernard T M. Video Extruder: A semi-dense point tracker for extracting beams of trajectories in real time [J]. Journal of Real-Time Image Processing, 2016, 11(4): 785-798. [12] Xu Yin, Zhai Yikui, Gan Junying. An image set and convex hull optimization algorithm for finger joint pattern recognition[J]. Signal Processing(信号处理), 2014, 30 (8): 930-936(in Chinese). [13]Sun Xiaoliang, Su Ang, Chen Shengyi. Objectness to assist salient object detection[J]. IET Image Processing, 2016, 10(5): 391-397. [14] Yang Guang, Tong Tao, Lu Yansong. Infrared and visible image fusion based on multi features[J]. Optics and Precision Engineering(光学精密工程), 2014, 22 (2): 489-496(in Chinese). [15] Peng Zhenming, Jing Liang, He Yanmin. Multi focus image super resolution fusion based on multiscale sparse dictionary[J]. Optics and Precision Engineering(光学精密工程), 2014, 21(1): 169-176(in Chinese). [16]Chai Yi，Li Huafeng，Zhang Xiaoyang．Multi-focus image fusion based on features contrast of multi-scale products in non-subsampled contourlet transform domain[J]. Optik-International Journal for Light and Electron Optics，2014，123( 7) : 569-581． [17]Charbal A，Dufour J E, Hild F, et al. Hybrid stereocorrelation using infrared and visible light cameras[J]. Experimental Mechanics, 2016, 56(5): 845-860.