Hyperspectral data fusion technology for weak small sea surface target
-
摘要: 海面目标受海水扰动影响，难以被宽波段光电传感器探测识别。高光谱传感器可以获取海面目标和水体的光谱数据，利用目标和海水的光谱特性差异可以有效抑制海面扰动影响，提高探测识别能力。针对高光谱数据降维融合容易丢失海面弱小目标问题，研究了弱小目标光谱数据融合方法。通过相似性分类产生类矩阵，由类矩阵主成分变换的降维投影矩阵来投影变换原有光谱数据，获得降维数据矩阵。降维矩阵通过空间变换转换到RGB彩色空间，生成伪彩色融合图像。通过远距离弱小目标和水中鱼群高光谱数据，验证了改进方法的融合性能。实验结果表明：相似性分类融合方法不仅能将高维光谱数据融合成一幅伪彩色图像，还能有效避免弱小目标融合丢失问题，提高了目标探测识别能力。Abstract: Aiming at solving problem of weak small target missing during dimension reduction and fusion of hyperspectral data, hyperspectral data fusion method for weak small target is studied in this paper. Class pixel matrix is produced by similarity measures, dimension reduction projection matrix is calculated by principal component transforming class pixel matrix, and dimension reduction data is obtained by dimension reduction projection matrix projecting the original spectral data. False-color result image is produced by RGB color space transforming from reduction data. Fusion performance of improved method is verified by hyperspectral data of weak small target and underwater fish group. Experiments show that similarity classification fusion method can not only get one false color image from high dimension hyperspectral data, but also avoid missing weak small targets from background, which enhances abilities of detection and identification, and has a strong application prospects.
-
-
![]()
图 2 靶场迷彩PCA后的前3个分量
Figure 2. First three components image after PCA in camouflage fabric
![]()
图 3 靶场迷彩宽波段电视成像与光谱融合图像对比
Figure 3. Comparision wide band TV imaging with spectral fusion image in camouflage fabric
-
[1] 苏晓丹.舰艇光电探测系统发展[J].飞航导弹, 2014(12):62-64. http://d.old.wanfangdata.com.cn/Periodical/fhdd201412014 Su Xiaodan.Ships photoelectric detection system development[J]. Winged Missiles Journal, 2014(12):62-64. http://d.old.wanfangdata.com.cn/Periodical/fhdd201412014
[2] 马优恒, 陆培国, 郭渝林, 等.舰载光电系统的环境适应性设计[J].应用光学, 2014, 35(3):371-376. http://d.old.wanfangdata.com.cn/Periodical/yygx201403004 Ma Youheng, Lu Peiguo, Guo Yulin, et al.Environmental worthiness design of shipborne electro-optical system[J]. Journal of Applied Optics, 2014, 35(3):371-376. http://d.old.wanfangdata.com.cn/Periodical/yygx201403004
[3] Shaw G A, Burke H K. Spectral imaging for remote sensing[J]. Lincoln Laboratory Journal, 2003, 14(1): 3-28. http://d.old.wanfangdata.com.cn/OAPaper/oai_pubmedcentral.nih.gov_3841904
[4] Manolakis D, Marden D, Shaw G A. Hyperspectral image processing for automatic target detection[J]. Lincoln Laboratory Journal, 2003, 14(1): 79-116. http://d.old.wanfangdata.com.cn/OAPaper/oai_doaj-articles_ec1eaca88a20f6136970335a3739acd5
[5] 余旭初, 冯伍法, 杨国鹏, 等.高光谱影像分析与应用[M].北京:科学出版社, 2013. Yu Xuchu, Feng Wufa, Yang Guopeng, et al.Hyperspectral image analysis and application[M]. Beijing:Science Press, 2013.
[6] Chang C I. Hyperspectral data processing algorithm design and analysis[M]. New Jersey:John Wiley & Sons, 2013.
[7] 朱院院, 高教波, 高泽东, 等.基于相似性分类的高光谱主成分融合方法比较[J].光学技术, 2016, 42(2):97-102. http://d.old.wanfangdata.com.cn/Periodical/gxjs201602001 Zhu Yuanyuan, Gao Jiaobo, Gao Zedong, et al.Comparison of principal component fusion method of hyperspectral image based on similarity classification[J]. Optical Technique, 2016, 42(2):97-102. http://d.old.wanfangdata.com.cn/Periodical/gxjs201602001
[8] 朱院院, 高教波, 高泽东, 等.基于多区间映射评价优选的光谱数据融合显示算法[J].应用光学, 2015, 36(5):728-734. http://d.old.wanfangdata.com.cn/Periodical/yygx201505011 Zhu Yuanyuan, Gao Jiaobo, Gao Zedong, et al.Fusion and vision algorithm of spectral data based on mapping-evaluating-optimizing methods within multi-section[J]. Journal of Applied Optics, 2015, 36(5):728-734. http://d.old.wanfangdata.com.cn/Periodical/yygx201505011
[9] Achalakul T, Taylor S. A concurrent spectral-screening PCT algorithm for remote sensing application[J]. Journal of Information Fusion, 2000(1/2): 89-97. http://cn.bing.com/academic/profile?id=2a7373d89fb9243fe5c0486660a14e57&encoded=0&v=paper_preview&mkt=zh-cn
[10] 张浚哲, 朱文泉, 郑周涛, 等.高光谱数据的相似性测度对比研究[J].测绘科学, 2013, 38(6):33-36. http://d.old.wanfangdata.com.cn/Periodical/chkx201306010 Zhang Junzhe, Zhu Wenquan, Zheng Zhoutao, et al. Comparative study on similarity measures of hyperspectral remote sensing data[J]. Science of Surveying and Mapping, 2013, 38(6):33-36. http://d.old.wanfangdata.com.cn/Periodical/chkx201306010
[11] 闻兵工, 冯伍法, 刘伟, 等.基于光谱曲线整体相似性测度的匹配分类[J].测绘科学技术学报, 2009, 26(2):128-131. doi: 10.3969/j.issn.1673-6338.2009.02.014 Wen Binggong, Feng Wufa, Liu Wei, et al.Matching and classification based on the whole comparability measure of spectral curve[J]. Journal of Geomatics Science and Technology, 2009, 26(2):128-131. doi: 10.3969/j.issn.1673-6338.2009.02.014
[12] 董广军, 张永生, 戴晨光, 等.基于信息散度特征的高光谱影像识别技术[J].仪器仪表学报, 2006, 27(6):2091-2092. http://d.old.wanfangdata.com.cn/Periodical/yqyb2006z3128 Dong Guangjun, Zhang Yongsheng, Dai Chenguang, et al. Target recognition for hyperspectral image based on SID spectral feature[J]. Chinese Journal of Scientific Instrument, 2006, 27(6):2091-2092. http://d.old.wanfangdata.com.cn/Periodical/yqyb2006z3128
[13] Fairchild M D. Color appearance models[M]. 3nd ed. West Sussex: John Wiley & Sons, 2013.
下载:
计量
- 文章访问数: 620
- HTML全文浏览量: 191
- PDF下载量: 89
陕公网安备 61011302001501号 