Stream-computing of High Accuracy On-board Real-time Cloud Detection for High Resolution Optical Satellite Imagery

doi:10.11947/j.JGGS.2019.0206

Abstract

Abstract:

This paper focuses on the time efficiency for machine vision and intelligent photogrammetry, especially high accuracy on-board real-time cloud detection method. With the development of technology, the data acquisition ability is growing continuously and the volume of raw data is increasing explosively. Meanwhile, because of the higher requirement of data accuracy, the computation load is also becoming heavier. This situation makes time efficiency extremely important. Moreover, the cloud cover rate of optical satellite imagery is up to approximately 50%, which is seriously restricting the applications of on-board intelligent photogrammetry services. To meet the on-board cloud detection requirements and offer valid input data to subsequent processing, this paper presents a stream-computing of high accuracy on-board real-time cloud detection solution which follows the “bottom-up” understanding strategy of machine vision and uses multiple embedded GPU with significant potential to be applied on-board. Without external memory, the data parallel pipeline system based on multiple processing modules of this solution could afford the “stream-in, processing, stream-out” real-time stream computing. In experiments, images of GF-2 satellite are used to validate the accuracy and performance of this approach, and the experimental results show that this solution could not only bring up cloud detection accuracy, but also match the on-board real-time processing requirements.

Key words: machine vision; intelligent photogrammetry; cloud detection; stream computing; on-board real-time processing

Mi WANG,Zhiqi ZHANG,Zhipeng DONG,Shuying JIN,Hongbo SU. Stream-computing of High Accuracy On-board Real-time Cloud Detection for High Resolution Optical Satellite Imagery[J]. Journal of Geodesy and Geoinformation Science, 2019, 2(2): 50-59.

Figures/Tables 16

Fig.1

Fig.2

Fig.3

Fig.4

Tab.1

Fig.5

Fig.6

Tab.2

Fig.7

Fig.8

Tab.3

Tab.4

Tab.5

Tab.6

Fig.9

Tab.7

References 26

[1]	LI Deren . Towards Geo-spatial Information Science in Big Data Era[J]. Acta Geodaetica et Cartographica Sinica, 2016,45(4):379-384.
[2]	LI Deren, SHEN Xin, GONG Jianya , et al. On Construction of China’s Space Information Network[J]. Geomatics and Information Science of Wuhan University, 2015,40(6):711-715, 766.
[3]	ZHANG Bing . Intelligent Remote Sensing Satellite System[J]. Journal of Remote Sensing, 2011,15(3):415-431.
[4]	LI Deren, WANG Mi, SHEN Xin , et al. From Earth Observation Satellite to Earth Observation Brain[J]. Geomatics and Information Science of Wuhan University, 2017,42(2):143-149.
[5]	DAVIS C O, HORAN D M, CORSON M R. On-orbit Calibration of the Naval EarthMap Observer (NEMO) Coastal Ocean Imaging Spectrometer (COIS) [C]//Proceedings Volume 4132, Imaging Spectrometry Ⅵ. San Diego, CA: SPIE, 2000: 250-259.
[6]	VISSER S J, DAWOOD A S . Real-time Natural Disasters Detection and Monitoring from Smart Earth Observation Satellite[J]. Journal of Aerospace Engineering, 2004,17(1):10-19.
[7]	HUANG Jingjin, ZHOU Guoqing . On-board Detection and Matching of Feature Points[J]. Remote Sensing, 2017,9(6):601.
[8]	WU Yuanfeng, GAO Lianru, ZHANG Bing, et al. Embedded GPU Implementation of Anomaly Detection for Hyperspectral Images [C]//Proceedings of SPIE 9646, High-performance Computing in Remote Sensing V. Toulouse, France: SPIE, 2015: 964608.
[9]	HOU Shuwei, SUN Wenfang, ZHENG Xiaosong . Overview of Cloud Detection Methods in Remote Sensing Images[J]. Space Electronic Technology, 2014,11(3):68-76, 86.
[10]	DINGUIRARD M, SLATER P N . Calibration of Space-multispectral Imaging Sensors:a Review[J]. Remote Sensing of Environment, 1999,68(3):194-205.
[11]	DINGUIRARD M, HENRY P . Calibration of SPOT HRV Cameras[J]. Remote Sensing of Environment, 1993.
[12]	GELLMAN D I, BIGGAR S F, DINGUIRARD M C, et al. Review of SPOT-1 and-2 Calibrations at White Sands from Launch to the Present [C]//Proceedings Volume 1938, Recent Advances in Sensors, Radiometric Calibration, and Processing of Remotely Sensed Data. Orlando, FL: SPIE, 1993: 118-125.
[13]	PAHLEVAN N, SCHOTT J R . Characterizing the Relative Calibration of Landsat-7(ETM+) Visible Bands with Terra (MODIS) over Clear Waters:the Implications for Monitoring Water Resources[J]. Remote Sensing of Environment, 2012,125:167-180.
[14]	PAN Zhiqiang, GU Xingfa, LIU Guodong , et al. Relative Radiometric Correction of CBERS-01 CCD Data Based on Detector Histogram Matching[J]. Geomatics and Information Science of Wuhan University, 2005,30(10):925-927.
[15]	YI Lina, Zhang Guifeng, WU Zhaocong . A Scale-Synjournal Method for High Spatial Resolution Remote Sensing Image Segmentation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2012,50(10):4062-4070.
[16]	LI Deren, ZHANG Guifeng, WU Zhaocong , et al. An Edge Embedded Marker-Based Watershed Algorithm for High Spatial Resolution Remote Sensing Image Segmentation[J]. IEEE Transactions on Image Processing, 2010,19(10):2781-2787.
[17]	REN X, MALIK J . Learning A Classification Model for Segmentation [C]//Proceedings of the 9th IEEE International Conference on Computer Vision. Nice, France: IEEE, 2003: 10-17.
[18]	VINCENT L, SOILLE P . Watersheds in Digital Spaces:An Efficient Algorithm Based on Immersion Simulations[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1991,13(6):583-598.
[19]	FELZENSZWALB P F, HUTTENLOCHER D P . Efficient Graph-based Image Segmentation[J]. International Journal of Computer Vision, 2004,59(2):167-181.
[20]	ACHANTA R, SHAJI A, SMITH K , et al. SLIC Superpixels Compared to State-of-the-art Superpixel Methods[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012,34(11):2274-2282.
[21]	DONG Zhipeng, MEI Xiaoming, CHEN Jie , et al. Image Segmentation of High Resolution Remote Sensing Imagery Based on Simple Linear Iterative Clustering[J]. Remote Sensing Information, 2016,31(6):83-88.
[22]	FANG Liuyang, WANG Mi, LI Deren , et al. CPU/GPU Near Real-time Preprocessing for ZY-3 Satellite Images:Relative Radiometric Correction, MTF Compensation, and Geocorrection[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,87:229-240.
[23]	CHAPMAN B, JOST G, VAN DER PAS R . Using OpenMP:Portable Shared Memory Parallel Programming[M]. Cambridge: MIT Press, 2007.
[24]	KIRK D B, HWU W M W . Programming Massively Parallel Processors:A Hands-on Approach[M]. 3rd ed. San Francisco, CA: Morgan Kaufmann, 2016.
[25]	NVIDIA JETSON . A Little Goes A Long Way[EB/OL]. [ 2017- 07- 31]. http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html .
[26]	TAN Kai, ZHANG Yongjun, TONG Xin , et al. Automatic Cloud Detection for Chinese High Resolution Remote Sensing Satellite Imagery[J]. Acta Geodaetica et Cartographica Sinica, 2016,45(5):581-591.

Algorithm	Computing Load	Parallelizability	Iterative	Correlationship
Radiometric correction	Light	High(by Pixel)	No	No
RGB To LAB	Light	High(by Pixel)	No	No
Seed point initialization	Heavy	Middle(by Block)	Yes	Middle
Similarity measure	Heavy	High(by Pixel)	Yes	High
Isolated pixel connection	Middle	Low	No	High
Superpixel determination	Light	Middle(by Block)	No	Low

No.	Location	Resolution	Image size	Cloud type
1	Lon32.3, Lat 3.3	4m	7300×6908	Thick
2	Lon115.4 Lat 41	4m	7300×6908	Thin

	Experiment1		Experiment2
	multi-spectral threshold method	SLIC method	multi-spectral threshold method	SLIC method
PR	0.8246	0.8745	0.9660	0.9735
ER	0.0347	0.0300	0.0461	0.0451

No.	Processing time/s
No.	multi-thread	CUDA	multi-thread+CUDA
0	20.785	23.971	11.303
1	20.297	22.575	10.754

Data segment	Generation time/s	Processing time/s	Module
Set 0, Seg. 0	0.403	1.439	0
Set 0, Seg. 1	0.403	1.501	1
Set 0, Seg. 2	0.403	1.462	2
Set 0, Seg. 3	0.403	1.454	3
Set 0, Seg. 4	0.403	1.443	0
Set 0, Seg. 5	0.403	1.427	1
Set 0, Seg. 6	0.403	1.483	2
Set 0, Seg. 7	0.396	1.417	3
Set 1, Seg. 0	0.399	1.470	0
Set 1, Seg. 1	0.399	1.421	1
Set 1, Seg. 2	0.399	1.418	2
Set 1, Seg. 3	0.399	1.449	3
Set 1, Seg. 4	0.399	1.462	0
Set 1, Seg. 5	0.399	1.399	1
Set 1, Seg. 6	0.399	1.423	2
Set 1, Seg. 7	0.394	1.408	3