Contour Detection of Disk Resolved Objects in Cassini ISS Image Using Deep Neural Network
-
摘要: Cassini空间探测器光学成像系统（ISS）拍摄的图像中，很多卫星呈现为面元，其轮廓检测是天体测量的重要工作.使用神经网络方法进行ISS图像中面元轮廓检测.每个ISS图像的像素分为轮廓边缘和非轮廓两类.使用神经网络框架TensorFlow，输入每个像素的9个特征，输出每个像素的分类.利用约3.6万个像素训练该网络，通过380幅ISS图像进行测试.与人工标记结果相比，轮廓像素检测的平均精确率为78.26%，平均召回率为73.32%.以检出轮廓像素作为输入，通过椭圆拟合得到面元的轮廓，所得轮廓与面元真实轮廓吻合良好.研究结果表明该方案能够有效检测出面元轮廓，进而给出假图像星的排除范围.
-
关键词:
- 轮廓检测 /
- 深度神经网络 /
- Cassini光学图像 /
- 边缘检测
Abstract: In the astrometry of CCD image, it is important to match image stars with catalogue stars to correct the camera's pointing. The onboard Imaging Science Subsystem (ISS) in Cassini orbiter has taken a large number of images of targets which are disk resolved. In the astrometry of these images, the false image stars are often detected in the disk. It disturbs the pointing correction of camera and decline the precision of the astrometry. Therefore, it is helpful to find the contours of the disk to remove the false image stars. One method based on deep learning is proposed to detect the contour of disk resolved object in ISS images. A convolutional neural network was set up by the framework TensorFlow, in which the input is the nine features of each pixel, and the output is the classification of each pixel: contour pixel or non-contour pixel. The neural network is trained by about 36000 pixels, and then it is used to detect the contour pixels in 380 ISS images. Compared with the contour pixels labeled by hand, the contour pixels detected by neural network have the precision of 78.26% and the recall ratio of 73.32%. It proved that the proposed method is available to find the contour of disk resolved target in ISS images.-
Key words:
- Contour Detection /
- Deep Neural Network /
- Cassini ISS images /
- Edge detection
-
[1] PORCO C, WEST R A, SQUYRES S, et al. Cassini imaging science:instrument characteristics and anticipated scientific investigations at Saturn[J]. Space Sci. Rev., 2004, 115:363-497 [2] COOPER N, MURRAY C, PORCO C, et al. Cassini ISS astrometric observations of the inner Jovian satellites, Amalthea and Thebe[J]. Icarus, 2006, 181:223-234 [3] COOPER N J, MURRAY C D, LAINEY V, et al. Cassini ISS mutual event astrometry of the mid-sized Saturnian satellites 2005-2012[J]. Astron. Astrophys., 2014, 572:A43 [4] TAJEDDINE R, COOPER N J, LAINEY V, et al. Astrometric reduction of Cassini ISS images of the Saturnian satellites Mimas and Enceladus[J]. Astron. Astrophys., 2013, 551:A129 [5] TAJEDDINE R, LAINEY V, COOPER N J, et al. Cassini ISS astrometry of the Saturnian satellites:Tethys, Dione, Rhea, Iapetus, and Phoebe 2004-2012[J]. Astron. Astrophys., 2015, 575:A73 [6] ZHANG Q F, LAINEY V, COOPER N J, et al. First astrometric reduction of Cassini Imaging Science Subsystem images using an automatic procedure:application to Enceladus images 2013-2017[J]. Mon. Not. R. Astron. Soc., 2018, 481:98-104 [7] ZHANG Q F, XIONG Y T, PENG Q Y, et al. Astrometric reduction of Cassini ISS images of Helene (in Chinese)[J]. Sci. Sin.:Phys. Mech. Astron., 2019, 49(1):96-103(张庆丰, 熊语桃, 彭青玉, 等. Helene的Cassini ISS图像的天体测量[J]. 中国科学:物理学力学天文学, 2019, 49(1):96-103) [8] ZHARKOVA V V, IPSON S S, ZHARKOVA S I, et al. A full-disk image standardisation of the synoptic solar observations at the Meudon Observatory[J]. Sol. Phys., 2003, 214(1):89-105 [9] YUAN F, LIN J B, GUO J J, et al. Automatic detection of solar features in HSOS full-disk solar images using guided filter[J]. New Astron., 2018, 59:38-44 [10] PU Junyu, ZHENG Yong, LI Chonghui, et al. A moon image center extraction algorithm adapted to moon phase change (in Chinese)[J]. J. Geomat. Sci. Technol., 2017, 34(05):461-465(蒲俊宇, 郑勇, 李崇辉, 等. 一种能适应月相变化的月球图像中心提取算法[J]. 测绘科学技术学报, 2017, 34(05):461-465) [11] PU Junyu, ZHENG Yong, CHEN Shaojie, et al. Geocentric position extraction algorithm of Earth sensor image of lunar rover (in Chinese)[J]. Chin. J. Space Sci., 2018, 38(2):239-248(蒲俊宇, 郑勇, 陈少杰, 等. 月球车地球敏感器图像地心位置提取算法[J]. 空间科学学报, 2018, 38(2):239-248) [12] LIU Han, GUO Yong, ZHENG Gang, et al. Edge detection based on least squares support vector machines[J]. Acta Electron. Sin., 2006, 34(7):1275-1279(刘涵, 郭勇, 郑岗, 等. 基于最小二乘支持向量机的图像边缘检测研究[J]. 电子学报, 2006, 34(7):1275-1279) [13] XIA Wei, WANG Ping, MA Jiayan, et al. Image edge detection based on image information measure and extreme learning machine[J]. Microcomput. Appl., 2017, 33(11):74-78 [14] LU Hongtao, ZHANG Qinchuan. Application of deep convolutional neural network in computer vision[J]. J. Data Acq. Process., 2016, 31(1):1-16 [15] JIN Tao, ZHANG Yongai. Digital recognition based on deep learning in TensorFlow platform[J]. Inform. Technol. Network Secur., 2018, 37(4):78-82 [16] GONZALEZ R C, WOODS R E. Digital Image Processing[M]. 3rd Edition. America:Prentice Hall Press, 2011 [17] GAO Fei, HUANG Qihong, CHENG Weidong, et al. Research of improvement of Sobel operator on edge detection[J]. Digital Technol. Appli., 2016, 2016(1):143-145 [18] XIA Jing. A Comparative Study of Ellipses Fitting Methods[D]. Guangzhou:Jinan University, 2007 -
-
计量
- 文章访问数: 1745
- HTML全文浏览量: 239
- PDF下载量: 126
-
被引次数:
0(来源:Crossref)
0(来源:其他)
下载:
