[1]丁永前,邢智雄,姜懿倬,等.光谱指数测量中窄带图像的光强自适应分割方法[J].南京农业大学学报,2020,43(3):574-581.[doi:10.7685/jnau.201908022]
 DING Yongqian,XING Zhixiong,JIANG Yizhuo,et al.Light intensity self-adaption segmentation method of narrow-band image in measuring spectral index[J].Journal of Nanjing Agricultural University,2020,43(3):574-581.[doi:10.7685/jnau.201908022]
点击复制

光谱指数测量中窄带图像的光强自适应分割方法()
分享到:

《南京农业大学学报》[ISSN:1000-2030/CN:32-1148/S]

卷:
43卷
期数:
2020年3期
页码:
574-581
栏目:
食品与工程
出版日期:
2020-05-10

文章信息/Info

Title:
Light intensity self-adaption segmentation method of narrow-band image in measuring spectral index
作者:
丁永前 邢智雄 姜懿倬 程浩明 徐明皓
南京农业大学工学院, 江苏 南京 210031
Author(s):
DING Yongqian XING Zhixiong JIANG Yizhuo CHENG Haoming XU Minghao
College of Engineering, Nanjing Agricultural University, Nanjing 210031, China
关键词:
归一化植被指数土壤背景干扰窄带图像分割几何变换光强自适应
Keywords:
normalized difference vegetation index(NDVI)soil background interferencenarrow-band image segmentationgeometric transformationlight intensity self-adaption
分类号:
S123
DOI:
10.7685/jnau.201908022
摘要:
[目的]本文对作物冠层区域的光强自适应性分割算法进行了研究,旨在消除光谱指数测量过程中土壤背景的干扰,克服现有光谱检测方法对结构光的依赖性。[方法]采用加装窄带滤光片的2个相机组建测量系统,并对相机的几何安装参数进行标定;将窄带图像灰度直方图的波谷灰度值与最大灰度值的比值作为归一化阈值,实现窄带图像分割的光环境自适应性。在实现近红外窄带图像中土壤和冠层分割的基础上,通过几何变换推算出可见光波段窄带图像的冠层区域,从而实现可见光窄带图像的分割。[结果]在80~140 cm高度内,每隔10 cm设置1个测量高度,每个高度采集5组平均株高为25 cm的绿萝冠层窄带图像(770和660 nm),安装660 nm滤光片的相机在完成图像采集后,将滤光片换成770 nm重新拍摄5张照片,作为660 nm图像分割效果的参考图像。结果表明:660 nm图像的分割区域与参考图像的分割区域平均重合度大于99%。同时计算了光照度为10 000~26 000 lx时拍摄的50幅绿萝770 nm窄带图像的非归一化阈值和归一化阈值,其与光照度的相关系数分别为0.586 6和0.091 6。[结论]本文提出的基于归一化阈值的分割方法对光环境的变化具有很强的适应性,综合归一化阈值和几何变换可以实现窄带图像的分割,为构建消除土壤干扰的归一化植被指数(normalized difference vegetation index,NDVI)等光谱指数提供有效的基础数据。
Abstract:
[Objectives] In order to eliminate the interference of soil background in the process of spectral indices measurement and overcome the dependence of existing spectral detection methods on structural light,the self-adaption segmentation algorithm of crop canopy region was studied in this paper. [Methods] In this paper,two cameras with narrow-band filters were used to build the measurement system,and the geometric installation parameters of the two cameras were calibrated. The ratio of the valley gray value of the narrow-band image gray histogram to the maximum gray value was used as a normalization threshold to achieve optical environment adaptability of narrow-band image segmentation. On the basis of realizing the segmentation of soil and canopy in the near-infrared narrow-band image,the canopy regions of the narrow-band images in the visible light band were derived by geometric transformation,and the narrow-band image of visible light could be segmented. [Results] In the height range of 80-140 cm,a measurement height was set every 10 cm,and each group collected 5 sets of narrow-band images(770 and 660 nm)of green radish with an average height of 25 cm. After the image acquisition was completed,the camera with the 660 nm filter was replaced with 770 nm to retake 5 photos as the reference images for the 660 nm images’ segmentation effect. The results showed that each average overlap of the segmentation region of the 660 nm image and the segmentation region of the reference image was greater than 99%. At the same time,the non-normalized thresholds and normalized thresholds of the 770 nm narrow-band images of 50 green radish photographed in the range of 10 000-26 000 lx were calculated,and their correlation coefficients with light intensity were 0.586 6 and 0.091 6,respectively. [Conclusions] The segmentation method based on normalized threshold proposed in this paper has strong adaptability to the changes of light environment. Combined with normalized threshold and geometric transformation,it can realize the segmentation of narrow-band images and provide effective basic data for constructing spectral indices which eliminate soil interference such as normalized difference vegetation index(NDVI).

参考文献/References:

[1] 倪军,姚霞,田永超,等. 便携式作物生长监测诊断仪的设计与试验[J]. 农业工程学报,2013,29(6):150-156. Ni J,Yao X,Tian Y C,et al. Design and experiments of portable apparatus for plant growth monitoring and diagnosis[J]. Transactions of the Chinese Society of Agricultural Engineering,2013,29(6):150-156(in Chinese with English abstract).
[2] 丁永前,朱艳,李杨,等. 基于主动光源的作物冠层反射光谱测量方法[J]. 农业机械学报,2013,44(12):209-214. Ding Y Q,Zhu Y,Li Y,et al. A method for measuring reflection spectrum of crop canopy based on active light source[J]. Transactions of the Chinese Society for Agricultural Machinery,2013,44(12):209-214(in Chinese with English abstract).
[3] 杨建宁,张井超,朱艳,等. 便携式作物生长监测诊断仪的性能试验[J]. 农业机械学报,2013,44(4):208-212,201. Yang J N,Zhang J C,Zhu Y,et al. Experiments on performance of portable plant growth monitoring diagnostic instrument[J]. Transactions of the Chinese Society for Agricultural Machinery,2013,44(4):208-212,201(in Chinese with English abstract).
[4] 卢少林,倪军,曹卫星,等. 基于主动光源的作物生长信息监测仪的设计与试验[J]. 农业工程学报,2014,30(23):199-206. Lu S L,Ni J,Cao W X,et al. Design and experiment for crop growth information monitoring instrument based on active light source[J]. Transactions of the Chinese Society of Agricultural Engineering,2014,30(23):199-206(in Chinese with English abstract).
[5] 丁永前,李杨,谭星祥,等. 冠层反射光谱测量中主动光源光谱稳定性控制[J]. 农业机械学报,2014,45(6):284-290. Ding Y Q,Li Y,Tan X X,et al. Control stability of active light source spectra during measuring reflectance of canopy[J]. Transactions of the Chinese Society for Agricultural Machinery,2014,45(6):284-290(in Chinese with English abstract).
[6] 李杨.主动光源式反射光谱测量仪检控电路的设计与性能测试[D]. 南京:南京农业大学,2015:12-14. Li Y. Design and performance test of detection and control circuit for reflectance spectroscopy measuring instrument based on active light source[D]. Nanjing:Nanjing Agricultural University,2015:12-14(in Chinese with English abstract).
[7] Jordan C F. Derivation of leaf area index from quality of light on the forest floor[J]. Ecology,1969,50:663-666.
[8] Huete A R. A soil-adjusted vegetation index(SAVI)[J]. Remote Sensing of Environment,1988,25(3):295-309.
[9] Gitelson A A,Gritz Y,Merzlyak M N. Relationships between leaf chlorophyll content and spectral reflectance and algorithms for non-destructive chlorophyll assessment in higher plant leaves[J]. Journal of Plant Physiology,2003,160(3):271-282.
[10] Asner G P,Martin R E. Spectral and chemical analysis of tropical forests:scaling from leaf to canopy levels[J]. Remote Sensing of Environment,2008,112(10):3958-3970.
[11] Haboudane D,Miller J R,Tremblay N,et al. Integrated narrow-band vegetation indices for prediction of crop chlorophyll content for application to precision agriculture[J]. Remote Sensing of Environment,2002,81(2/3):416-426.
[12] Zarco-Tejada P J,Berjón A,López-Lozano R,et al. Assessing vineyard condition with hyperspectral indices:leaf and canopy reflectance simulation in a row-structured discontinuous canopy[J]. Remote Sensing of Environment,2005,99(3):271-287.
[13] Yu K,Lenz-Wiedemann V,Chen X P,et al. Estimating leaf chlorophyll of barley at different growth stages using spectral indices to reduce soil background and canopy structure effects[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2014,97:58-77.
[14] Daughtry C S T,Walthall C L,Kim M S,et al. Estimating corn leaf chlorophyll concentration from leaf and canopy reflectance[J]. Remote Sensing of Environment,2000,74(2):229-239.
[15] 方博文,张晓东,陈敬义,等. 基于双目视觉的行车中障碍距离检测方法研究[J]. 机械设计与制造,2019(4):94-98. Fang B W,Zhang X D,Chen J Y,et al. Research on detecting method of obstacle distance in driving based on binocular vision[J]. Machinery Design & Manufacture,2019(4):94-98(in Chinese with English abstract).
[16] Govindaraj V,Sengottaiyan G. Survey of image denoising using different filters[J]. International Journal of Science,Engineering and Technology Research,2013,2(2):344-351.
[17] Otsu N. A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems,Man and Cybernetics,1979,9(1):62-66.

备注/Memo

备注/Memo:
收稿日期:2019-08-11。
基金项目:国家重点研发计划项目(2016YFD070030403);国家大学生创新训练计划项目(20181037090)
作者简介:丁永前,副教授,博士,主要从事农业信息化和自动化研究,E-mail:yongqiand@njau.edu.cn。
更新日期/Last Update: 1900-01-01