Four component polarimetric SAR image decomposition Experiment

420 shown in Fig. . This setup achieves a stepped frequency continuous radar SFCW configuration and wide‐band observation since the antennas can be used in frequency range from Mz until Gz. The inverse SAR SAR method, where target rotates in static place while SAR collecting the receiving data, was adopted to this observation. Because the target is in the antenna footprint all time during the measurement, this method is considered as spotlight mode and can obtain high resolution images compared to stripmap GB‐SAR measurement. Fig. 3. Generated SAR images with 50 ° off‐nadir angle obtained in Sep.14. a HH b HV c VV Fig. 4. Decomposed SAR images with 50° off‐nadir angle obtained in Sep.14. a Surface scattering b Double bounce scattering c Volume scattering d Helix scattering Fig. 5. The normalized average power. a off‐nadir 72°. b off‐nadir 50°. Fig. 6. Relative scattering components of Ps, Pd, Pv, and Pc. a off‐nadir 72°. b off‐nadir 50°. 421

4. Experimental results and discussion

The generated SAR images with ° off‐nadir angle which is obtained in Sep. are shown in Fig. as representative of many images. These images are normalized to dB as choosing a maximum value among three images. , V, and VV polarization images are placed in Fig. respectively. Fig. shows the clear dots which represent the planted place of paddy rice. n addition, this figure shows highest scattering from VV‐ polarization because paddy rice stands as vertical shape. Fig. shows the images which is decomposed to each power component of Ps surface scattering , Pd double‐bounce scattering , Pv volume scattering , and Pc helix scattering from SAR images of Fig. . Furthermore, we calculated average power values from each decomposed power image and performed normalization to the maximum power in whole values, shown in Fig. , and scattering component rate is shown in Fig. . From these figures, the obvious difference between two off‐nadir angles are observed. The measurements with ° off‐nadir angle indicate that the surface and double‐bounce scattering are dominant among the scattering components while the measurements with ° show the highest volume scatterings and unignorable high helix scattering Figure . This result indicates that the helix scattering is also one of the important factor for paddy rice monitoring, although many literatures have discussed paddy rice monitoring without considering of Pc scattering until now. Although the seasonal results show strong dependency on the off‐nadir angle, characterizing the paddy rice according to the observation time is not enough. Therefore, further analysis such as alpha decomposition is planted to conduct.

5. Conclusion

Seasonal paddy rice observations have been conducted by using pol‐GB‐SAR with two specific off‐nadir angles. The paddy rice which are planted inside of the polystyrene box are used in the measurement. To perform the polarimetric analysis, four components polarimetric SAR decomposition technique is processed in each SAR images. The seasonal results show high dependency on the off‐nadir angle in the paddy rice observation and unignorable helix scattering effect for observing paddy fields. owever, the decomposition analysis is not enough to characterize the paddy rice according to the observation date. Therefore, more analysis will be performed as a next step of this study. Acknowledgements This work was supported in part by the European Space Agency Earth Observation Category under Grant , by the th Japan Aerospace Exploration Agency JAXA ALOS Research Announcement under Grant , by the th JAXA ALOS Research Announcement under Grant , by the Japanese Government National Budget Special Budget for Project FY under Grant , Taiwan National Space Organization under Grant NSPO‐S‐ ; ndonesian Bhimasena, and by the Chiba University Strategic Priority Research Promotion Program FY .. References ayashi, N., and Sato, M. . Measurement and analysis of paddy field by polarimetric GB‐SAR. n 2009 IEEE International Geoscience and Remote Sensing Symposium Vol. , pp. V‐ 8 . Cape Town, EEE. 422 Yamaguchi, Y., Moriyama, T., shido, M., and Yamada, . . Four‐component scattering model for polarimetric SAR image decomposition, IEEE Transactions on Geoscience and Remote Sensing , Vol. , No.8, pp. ‐ . EEE, USA. Li, K., Brisco, B., Yun, S., and Touzi, R. . Polarimetric decomposition with RADARSAT‐ for rice mapping and monitoring, Canadian Journal of Remote Sensing, Vol. 8, No. , pp. ‐ . Canadian Science Publishing, Canada Joint Scientific Symposium IJJSS 2016 Chiba, 20‐24 November 2016 423 Exploring Attractive Landscape Elements and Sceneries of Bukit Kucing Forest Tanjungpinang by using Visitors-Employed Photography Method Akhmad Arifin adi a , Yusuke Mizuuchi b , Katsunori Furuya c a Bogor Agricultural University, Jl. Meranti Faperta IPB Dramaga Bogor 16680, Indonesia b The University of Tokyo, 489‐0031 Aichi‐ken Seto‐shi, Goizukacho 11‐44, Japan c Chiba University, 271‐8510 Chiba‐ken Matsudo‐shi 648, Japan Abstract The Bukit Kucing Forest BKF is a protected areas preserved in the middle of Tanjungpinang City, Indonesia. As an open protected area, the BKF is being managed for public recreation instead of wildlife protection and forest conservation. The visitor-employed photography VEP method was implemented in this research by involving 61 university students who study in universities in Tanjungpinang city. The participants were asked to walk following predetermined tracks and capture landscape objects or sceneries that impressed them during walking. Each participants used their own camera or smartphone with camera and each of them were attached a GPS logger to record their path and the time it took. All photos captured by participants were copied to researchers desktop in order to be object of the research. Each photos obtained from respondents were matched between captured time and walking time to obtained geo-tagged photos. There were 1,259 geo-tagged photographs obtained from respondents. By using density analysis in ArcGIS, it identified the concentration of photos captured places. The finding shows there are 7 clusters of landscape character captured by participants where each clusters has different trend locations of photos points’ density. It shows a valuable input for planning and management of BKF related with attractive landscape views and elements with its trend locations. However, the most preferred landscape is still a question in this research. The next research of participant’s travelling time is needed in order to investigate the most preferred landscape and its observation places in relation with observation time. Keywords VEP; Landscape; Scenery: Preference; Forest; Tanjungpinang

1. Introduction

The perception of beauty of a landscape among professional planners, managers and users is sometimes different DeLucio Mugica, ; ofmann, Westermann, Kowarik, Van der Meer, . What landscape planners and managers think about interesting places and objects sometimes different with user preferences. Therefore the Corresponding author. Tel.: +8 ‐ ‐ ‐ . E ‐mail address: arifin_hadiipb.ac.id