<?xml version="1.0" encoding="UTF-8"?><xml><records><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>17</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Koga, Yohei</style></author><author><style face="normal" font="default" size="100%">Miyazaki, Hiroyuki</style></author><author><style face="normal" font="default" size="100%">Shibasaki, Ryosuke</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">A Method for Vehicle Detection in High-Resolution Satellite Images that Uses a Region-Based Object Detector and Unsupervised Domain Adaptation</style></title><secondary-title><style face="normal" font="default" size="100%">Remote Sensing</style></secondary-title></titles><dates><year><style  face="normal" font="default" size="100%">2020</style></year></dates><urls><web-urls><url><style face="normal" font="default" size="100%">https://www.mdpi.com/2072-4292/12/3/575</style></url></web-urls></urls><volume><style face="normal" font="default" size="100%">12</style></volume><pages><style face="normal" font="default" size="100%">575</style></pages><language><style face="normal" font="default" size="100%">eng</style></language><abstract><style face="normal" font="default" size="100%">Recently, object detectors based on deep learning have become widely used for vehicle detection and contributed to drastic improvement in performance measures. However, deep learning requires much training data, and detection performance notably degrades when the target area of vehicle detection (the target domain) is different from the training data (the source domain). To address this problem, we propose an unsupervised domain adaptation (DA) method that does not require labeled training data, and thus can maintain detection performance in the target domain at a low cost. We applied Correlation alignment (CORAL) DA and adversarial DA to our region-based vehicle detector and improved the detection accuracy by over 10% in the target domain. We further improved adversarial DA by utilizing the reconstruction loss to facilitate learning semantic features. Our proposed method achieved slightly better performance than the accuracy achieved with the labeled training data of the target domain. We demonstrated that our improved DA method could achieve almost the same level of accuracy at a lower cost than non-DA methods with a sufficient amount of labeled training data of the target domain.</style></abstract></record><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>10</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Itabashi, Koichiro</style></author><author><style face="normal" font="default" size="100%">Miyazaki, Hiroyuki</style></author><author><style face="normal" font="default" size="100%">Iwao, Koki</style></author><author><style face="normal" font="default" size="100%">Nakamura, Kazuki</style></author><author><style face="normal" font="default" size="100%">MATSUOKA, Masashi</style></author><author><style face="normal" font="default" size="100%">Shibasaki, Ryosuke</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">A Method for Constructing Urban Extent Map from ALOS/PALSAR Satellite Data</style></title><secondary-title><style face="normal" font="default" size="100%">32nd Asian Conference on Remote Sensing</style></secondary-title></titles><keywords><keyword><style  face="normal" font="default" size="100%">ALOS/PALSAR</style></keyword><keyword><style  face="normal" font="default" size="100%">classification</style></keyword><keyword><style  face="normal" font="default" size="100%">microwave sensor</style></keyword><keyword><style  face="normal" font="default" size="100%">urban extent map</style></keyword></keywords><dates><year><style  face="normal" font="default" size="100%">2011</style></year><pub-dates><date><style  face="normal" font="default" size="100%">2011</style></date></pub-dates></dates><urls><web-urls><url><style face="normal" font="default" size="100%">http://a-a-r-s.org/acrs/index.php/acrs/acrs-overview/proceedings-1?view=publication&amp;task=show&amp;id=1005</style></url></web-urls></urls><pub-location><style face="normal" font="default" size="100%">Taipei</style></pub-location><pages><style face="normal" font="default" size="100%">TS9-1</style></pages><language><style face="normal" font="default" size="100%">eng</style></language><abstract><style face="normal" font="default" size="100%">Currently, global urban extent map of high accuracy and high resolution have been constructed mainly using optical sensor including ASTER/VNIR. However, there are some regions where urban areas are not correctly detected due to cloud cover and similar reflectance among land cover classes. To solve the problems, we used microwave sensor images of ALOS/PALSAR, which has an advantage in enabling observation in all weather conditions. This study aims at examining the possibility of using ALOS/PALSAR images as an alternative data resource for constructing urban extent map. Firstly, to determine useful ALOS/PALSAR observation mode, we examined how often ALOS/PALSAR images are taken in the regions for which an existing method using ASTER/VNIR images could not detect urban area correctly. Secondly, we collected ALOS/PALSAR satellite images, and examined effect of local-incident-angle-corrected images of ALOS/PALSAR taken by Fine Resolution Mode which can reduce distortion of pixel values due to local incident angle. We also performed unsupervised classifications on the ALOS/PALSAR and local-incident-angle-corrected images. Finally, we discussed ground truth datasets for image classification.</style></abstract></record><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>10</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Itabashi, Koichiro</style></author><author><style face="normal" font="default" size="100%">Miyazaki, Hiroyuki</style></author><author><style face="normal" font="default" size="100%">Iwao, Koki</style></author><author><style face="normal" font="default" size="100%">Nakamura, Kazuki</style></author><author><style face="normal" font="default" size="100%">Shibasaki, Ryosuke</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">A method for detecting and mapping Urban Area by ALOS/PALSAR data</style></title><secondary-title><style face="normal" font="default" size="100%">31st Asian Conference on Remote Sensing</style></secondary-title></titles><keywords><keyword><style  face="normal" font="default" size="100%">ALOS PALSAR urban extent map high resolution</style></keyword></keywords><dates><year><style  face="normal" font="default" size="100%">2010</style></year><pub-dates><date><style  face="normal" font="default" size="100%">2010</style></date></pub-dates></dates><pub-location><style face="normal" font="default" size="100%">Hanoi, Vietnam</style></pub-location><language><style face="normal" font="default" size="100%">eng</style></language><abstract><style face="normal" font="default" size="100%">Currently, global urban extent map of high accuracy and resolution have been constructed mainly using optical sensor including ASTER/VNIR. However, there are some regions where urban area is not detected because of cloud cover and similar reflectance among land cover classes. In the present work, by using ALOS/PALSAR, a microwave sensor, we proposed a method for detecting urban area which cannot be detected by ASTER/VNIR optical sensor and developing urban extent map in high accuracy and resolution. We mainly used satellite images taken by Fine Resolution Mode of ALOS/PALSAR. Local-incident-angle corrected images by Fine Resolution Mode were used for this method. The proposed method consists of sampling pixel values and ground truth data at urban and non-urban area from ALOS/PALSAR images; constructing classifier based on the pixel values and ground truth data; and classifying pixels into urban or non-urban area. We compared the results with urban extent map derived from ASTER/VNIR optical sensor images, and evaluated the possibility of using ALOS/PALSAR data for developing urban extent map. In addition, we examined accuracy improvement of detecting urban area using both ASTER/VNIR and ALOS/PALSAR images. The proposed method could classify regions which were misclassified by ASTER/VNIR optical sensor images, and develop urban extent map in high accuracy and resolution.</style></abstract></record><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>17</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Miyazaki, Hiroyuki</style></author><author><style face="normal" font="default" size="100%">Tanaka, Ayako</style></author><author><style face="normal" font="default" size="100%">Iwao, Koki</style></author><author><style face="normal" font="default" size="100%">Shibasaki, Ryosuke</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">A Method for Developing Urban Extent Map of High Accuracy and Resolution by Integrating ASTER/VNIR Images and Existing Urban Extent Maps</style></title><secondary-title><style face="normal" font="default" size="100%">Journal of the Japan Society of Photogrammetry and Remote Sensing</style></secondary-title></titles><dates><year><style  face="normal" font="default" size="100%">2009</style></year></dates><urls><web-urls><url><style face="normal" font="default" size="100%">http://ci.nii.ac.jp/naid/10025572312/en/</style></url></web-urls></urls><number><style face="normal" font="default" size="100%">2</style></number><publisher><style face="normal" font="default" size="100%">日本写真測量学会</style></publisher><volume><style face="normal" font="default" size="100%">48</style></volume><pages><style face="normal" font="default" size="100%">82–96</style></pages><language><style face="normal" font="default" size="100%">eng</style></language></record></records></xml>