AI in Remote Sensing: An Introduction to the Current State
April 23, 2022, 19:55 – 20:10
Prof. Qihao WENG
- Associate Director of Research Institute for Land and Space (RILS)
- Chair Professor of Geomatics and Artificial Intelligence Department of Land Surveying and Geo-Informatics
Prof. Qihao Weng, a Foreign Member of the Academy of Europe (Academia Europaea), and a Fellow of IEEE, AAAS, ASPRS, and AAIA, is a Chair Professor at the Hong Kong Polytechnic University, and worked as the Director of the Center for Urban and Environmental Change and a Professor of Geography at Indiana State University, 2001-2021, and as a Senior Fellow at the National Aeronautics and Space Administration from 2008 to 2009. He received his Ph.D. degree in geography from the University of Georgia in 1999. Weng is the Lead of GEO Global Urban Observation and Information Initiative, 2012-2022, and an Editor-in-Chief of ISPRS Journal of Photogrammetry and Remote Sensing. Additionally, he serves as the Series Editor of Taylor & Francis Series in Remote Sensing Applications, and Taylor & Francis Series in Imaging Science. Weng has been the Organizer and Program Committee Chair of the biennial IEEE/ISPRS/GEO sponsored International Workshop on Earth Observation and Remote Sensing Applications conference series since 2008, a National Director of American Society for Photogrammetry and Remote Sensing from 2007 to 2010, and a panelist of U.S. DOE’s Cool Roofs Roadmap and Strategy in 2010.
This talk will discuss the current state of the artificial intelligence in remote sensing. Machine learning for remote sensing image analysis, including traditional machine learning models and the popular deep neural networks, will also be introduced. In addition, application examples of machine learning to satellite-based Earth observation data will be presented.
翁齊浩教授是歐洲科學院外籍院士、AAAS會士、IEEE會士、AAG會士、ASPRS會士、AAIA會士，現任香港理工大學地理信息學和人工智能講座教授、曾任美國印第安納州立大學城市與環境變化中心主任和教授和美國航天局高級研究員。 1999年獲美國喬治亞大學地理學博士學位。現為地球觀測組織的全球城市觀測和信息系統項目負責人，任《國際攝影測量與遙感學會期刊》主編、國際攝影測量與遙感學會基金會董事，曾任美國地理學家協會（AAG）中國地理學專業委員會主席和美國攝影測量和遙感學會（ASPRS）全國理事。翁教授的研究側重於遙感科学和技術在城市環境與生態系統中的應用、土地利用和土地覆蓋的變化和城市化的環境效應等，共發表263篇論文，14本英文書籍，谷歌學術引用次數25800餘次，H-指數達72，得到了美國國家科學基金會和美國航天局等部門的研究資助。翁教授是城市遙感領域的領軍人物之一，在遙感技術應用於城市環境、城市化發展的研究中做出了系統性的創新貢獻，開創了一系列遙感地學分析方法和算法，完善城市熱島、城市擴張和環境可持續性的理論研究體係與方法論。早在1992年翁教授出國前，著名科學家錢學森院士親自與他聯繫，索取探討其地理哲學論著。翁教授入選2021年香港特區首批傑出創科學人計劃（Global STEM Professorship）、獲得了2020年美國地理學會傑出學術成就獎、2019年泰勒和弗朗西斯終身成就獎、2019年日本學術振興會傑出科學家獎和2011年遙感科學傑出貢獻獎。 2018年當選美國電氣電子工程師協會 (IEEE)會士，2019年當選美國科學促進會（AAAS）會士，2020年當選美國攝影測量和遙感學會（ASPRS）會士，2021年當選歐洲科學院外籍院士、美國地理學家協會 (AAG)會士和亞太人工智能學會（AAIA）會士。
遙感技術與實地調查相結合，可用於觀測、監測、測量和模擬構成自然和人類生態系統的許多組成部分。在社會需求和衛星和傳感器技術和圖像處理技術（如人工智能）的推動下，我們目睹了全球研發、技術轉讓和工程应用的巨大增長。自20世紀90年代以來，商業衛星、高光譜成像、激光雷達和雷達技術得到發明或重新發明。與互聯網技術和地理信息系統的集成導致了谷歌地球等網絡上地理參考信息的出現。這些新的"地理虛擬世界"，結合GPS，移動地圖和現代電信技術，已引起政府、行業和公眾的極大興趣。遙感 “大數據” 的增長伴隨著免費中低分辨率衛星圖像而发展。隨著衛星圖像採集的時間分辨率不斷提高，特別是那些具有高空間分辨率的圖像，導致任何地点的圖像積累增加，為時間序列分析和場景動態的實時評估創造了可能性。自 1992 年以來，衛星收集的夜間燈光圖像已越來越多地用作人類住區和活動的指標，利用時間序列夜間燈光圖像來監測城市增長，分析社會經濟和環境變量的動態，並評估城市化的環境後果。
本講座拟探討使用白天和夜間的時間序列卫星圖像，通過比较大数据和小数据方法來分析大灣區的城市化过程，討論两個關鍵科學問題：（1）時間序列圖像如何用於表徵和量化城市化進程？ (2) 是否可以提取夜間燈光特徵來增強白天圖像，以繪製和表徵各種類型的城市土地利用變化？
除咗學術性的討論之外， 准備高考或系度考慮日後升學問題的同學 也歡迎參加这次講座， 因為屆時還會介紹2022內地招生計劃、 流程跟入學要求。 喺最新公布的 2022年QS世界大學學科排名中， 建設及環境學院的兩個學科 「建築/建築環境」 及 「土木及結構工程」 均位列全球第15位！再次印證學院實力雄厚！
- 會議平台: ZOOM
- 课程網址: https://www.polyu.edu.hk/rils/news-and-events/event/2022/5/20220507_polyu-cloud-lecture-hall_prof-qihao-weng/?sc_lang=tc
- 註冊網址: https://www.polyu.edu.hk/pfs/index.php/666139?lang=zh-Hans
Where does Remote Sensing Belong: Technology, Science, AI, or …?
September 27, 2022, 18:00-19:00
Prof. Qihao Weng
Chair Professor of Geomatics and Artificial Intelligence Department of Land Surveying and Geo-Informatics
Dr. Qihao Weng is a Chair Professor of Geomatics and Artificial Intelligence at the Hong Kong Polytechnic University since July 2021. He worked as the Director of the Center for Urban and Environmental Change (July 2004-2021) and a Professor at Indiana State University, 2001-2021, and was a Senior Fellow at the NASA Marshall Space Flight Center from Dec. 2008 to Dec. 2009. He received his Ph.D. degree in geography (concentrations: remote sensing, GIS, and environment) from the University of Georgia in 1999. Weng is currently the Lead of the Group on Earth Observation’s (GEO) Global Urban Observation and Information Initiative and serves as an Editor-in-Chief of the ISPRS Journal of Photogrammetry and Remote Sensing. Additionally, he works as the Series Editor of Taylor & Francis Series in Remote Sensing Applications and Taylor & Francis Series in Imaging Science. Weng has been the Organizer and Program Committee Chair of the biennial IEEE International Workshop on Earth Observation and Remote Sensing Applications conference series since 2008, a National Director of the American Society for Photogrammetry and Remote Sensing (ASPRS) from 2007 to 2010, a Trustee of the ISPRS Foundation since 2021, and a panelist of U.S. DOE’s Cool Roofs Roadmap and Strategy committee in 2010.
In 2008, Weng received a prestigious NASA senior fellowship. He received the Outstanding Contributions Award in Remote Sensing in 2011, the AAG E. Willard and Ruby S. Miller Award in 2015, the 2020 AAG Distinguished Scholarship Honors Award, and Outstanding Service Award of AAG China Geography Specialty Group in 2021, all from the American Association of Geographers (AAG). Furthermore, in 2019, he was given a Taylor & Francis Lifetime Achievements Award, and a fellowship from Japan Society for the Promotion of Science under the “JSPS Invitational Fellowships for Research in Japan” (Short-term S [E], formerly known as “JSPS Award for Eminent Scientists”). In 2021, Weng was awarded a Global STEM Professorship by the Hong Kong government. At Indiana State University, he was selected as a Lilly Foundation Faculty Fellow in 2015. In the following year, he also received the Theodore Dreiser Distinguished Research Award. Additionally, Weng was the recipient of 2010 Erdas Award for Best Scientific Paper in Remote Sensing (1st place) and the 1999 Robert E. Altenhofen Memorial Scholarship Award, which were both awarded by the ASPRS. He was also awarded the Best Student-Authored Paper Award by the International Geographic Information Foundation in 1998. Weng has been invited to give more than 120 talks by worldwide organizations and conferences and is honored with a visiting professorship in a dozen of universities in Japan, France, and China. Weng was elected a fellow of the Institute of Electrical and Electronics Engineers (IEEE) in 2018, the American Association for the Advancement of Science (AAAS) in 2019, the ASPRS in 2020, the AAG and the Asia-Pacific Artificial Intelligence Association (AAIA) in 2021. He is an elected foreign member of Academia Europaea (The Academy of Europe) in 2021.
Weng’s research focuses on remote sensing applications to urban environmental and ecological systems, land-use and land-cover changes, urbanization impacts, urban sustainability, environmental modeling, and human-environment interactions. Through the serial invention of innovative techniques, methods, and theories, Weng’s research efforts have fostered growth in the geographical and environmental applications of the science and technology of remote sensing. Moreover, his research has bridged methodological gaps between geography, landscape ecology, and environmental science, uniting these related fields to design holistic solutions to many environmental, ecological, and climatic issues. Throughout his research career, Weng has authored 266 articles (journal articles, chapters, etc.) and 14 books. Per Google Scholar, as of April 2022, he has been cited more than 26000 times (H-index of 72), and 55 of his publications have more than 100 citations each. Weng’s research has been supported by funding agencies including the NSF, NASA, USGS, USAID, NOAA, National Geographic, and Microsoft.
Remote sensing eventuated with the development of flight technology, with a symbolic event of taking photographs of Paris from hot-air balloon by G. Tournachon in 1858. Aerial photography was first developed for military surveillance and reconnaissance purposes commencing the World War I and measured up during the Cold War. The development of Earth observing and weather sensors onboard satellites since the second half of the 20th century allows remote sensing to progress to a global scale and provides data for civil, research, and military usages. At the turn of the 21st century, we have witnessed great advances in remote sensing, in particularly commercial satellites and drones, with innovative sensor technologies as hyperspectral imaging, Light Detection and Ranging (LiDAR), and spaceborne radar. These technologies are not isolated at all. In fact, their integrated uses with more established aerial photography and multispectral remote sensing techniques have been the mainstream of current remote sensing research and applications.
Today, remote sensing has become an essential tool for understanding the Earth and managing human-Earth interactions. The US Department of Labor identified geotechnology, to which remote sensing is a core, as one of the three most important emerging and evolving fields, along with nanotechnology and biotechnology. By integrating with Geographic Information System (GIS), GPS, mobile mapping, Internet Technology and modern telecommunication technologies, remote sensing has contributed to the emergence of abundant geo-referenced information over the web and hands, creating a new geo-referenced “world”. How to process ever increasing remote sensing, and associated geospatial big data, has become a new challenge. In the past few years, artificial intelligence (AI) is rapidly developing and has been applied to a variety of remote sensing areas. This talk examines the interplay between remote sensing, image/signal processing, AI, and computer vision and how this interplay will likely shape the discipline of remote sensing. Three key science questions will be discussed, including: (1) Is remote sensing a technology, science, or something else? (2) Has recent development in AI re-shaped the field of remote sensing, and if so, how? and (3) Have there been any paradigm shifts in the remote sensing research? If so, how are they associated with research frontiers over time? Can we synthesize advances across those multiple frontiers to advance the remote sensing field?
- Online/ Physical (*subject to the pandemic development）
GRSS 2022 Educational Webinar
- Prof. Qihao Weng, Hong Kong Polytechnic University
- Prof. Yuei-An Liou, National Central University, Taiwan
This webinar series aims at providing students, young professionals, people wishing to continuing education, and the curious minds, with 30 to 60 minutes on-line webinar lectures.
Satellite Time Series Reconstruction and Applications by Prof. Xiaolin Zhu (The Hong Kong Polytechnic University)
Uncovering Human-Urban Environment Interaction with Multi-Source Data and Interdisciplinary Approaches by Prof. Bin Chen (The University of Hong Kong)
The Non-Physical Approach to Calibrate and Improve the Accuracy of Satellite Remote Sensing Water Vapor Based on Ground-Based GNSS Observations by Prof. Zhizhao (George) Liu (The Hong Kong Polytechnic University)
Integrating Radar and Optical Data for Monitoring Urban Development by Prof. Hongsheng Zhang (The University of Hong Kong)
Geo-Data for Making Better Decisions on Climate Change Programs by Dr. Pooja Mahapatra (Fugro)
Remote Sensing and Machine Learning
August 17-19, 2022
- To learn the basic principles and methods of remote sensing digital image analysis and machine learning;
- To acquire hands-on experience in implementing automatic image analysis methods, in performing classification, semantic segmentation, feature extraction and result evaluation; and
- To set up a computer to perform machine learning.
- 1 days of tutorials; and
- 2 days hands-on exercises on machine learning and digital image processing.
- For Tutorial and Lab Exercise at PolyU Campus.
- 20 Professional and Technical Grade Staff with basic knowledge in image processing from Hong Kong SAR government.