home  Home      home  ISPRS Foundation      sitemap  Sitemap      search  Search    
 
July 21, 2018 
Scientific Initiatives 2015

In accordance with the statutory mission and activities of ISPRS, the Society provides funds to support scientific and other initiatives, which will further improve its international status in the field of the photogrammetry, remote sensing and spatial information sciences, and will therefore benefit all ISPRS members. To this end, a call for new ISPRS Scientific Initiatives was issued in October 2014 and 11 teams of researchers from working groups across all of the eight ISPRS Technical Commissions submitted applications. Comments were sought from the relevant Technical Commission Presidents, and a sub-committee of three Council members was convened to independently review the proposals on the basis of scientific excellence. The sub-committee’s recommendation to fund seven projects to a total of CHF 33,000 Swiss Francs was accepted by ISPRS Council at its 2014 meeting in Hyderabad. The following briefly describes the outcomes of each project, with full reports available on the ISPRS website.


Completion of the ISPRS Benchmark for Multi-platform Photogrammetry

Principal Investigator: Francesco Nex; FBK Trento, Italy

This project, conducted in conjunction with EuroSDR, aimed to collect and share state-of-the-art data over urban areas to assess the performance of image orientation and matching algorithms in the handling of images captured from different platforms. Oblique aerial images, UAV and terrestrial data of two selected areas have been acquired and processed for delivery to participants. The city center of Dortmund and the area of the Zeche Zollern Museum (Dortmund, Germany) were chosen as test fields. Due to the large amount of data, different scenarios have been established for the orientation and matching tasks. The orientation and matching algorithms are assessed using reference points, Terrestrial and Airborne Laser Scanning that have also been acquired. The benchmark is now ready for participants, with many researchers having already downloaded the data. Their results will be published on a dedicated webpage.
 
Full Report


ISPRS Open Data Challenge 2015

Principal Investigator: Qing-Quan Li; Shenzhen University, China

Lasting for over five months, the 2015 ISPRS Open Data Challenge started in June 2015 and concluded in December 2015, with 59 teams from China, Hong Kong, Japan, USA and the UK participating. The challenge released the GPS trajectories of over 1,000 buses, as well as millions of records for associated transit smart cards. The top-10 teams from a qualification round obtained a copy of the dataset and were invited to attend the final of the competition where they had to design creative projects using the data and any other open source data and software. In the final, seven experts from transportation, computer science, and IT companies were invited as judges. The team from Zhejiang University of Technology (RunBunny) won the first prize, two teams (Vi-Yoghourt, WHUGCM) shared the second place prize, and three teams (Moonlight, TYUT_GIS_LAB, and Mango) shared third place. The challenge also open sourced the codes and algorithms from the top-10 teams in the qualification round at GitHub: http://citysensing.cn/opencup/opensources.html. For more information about the 2015 ISPRS Open Data Challenge, please visit the Challenge website: http://citysensing.cn/opencup/home.html.
 
Full Report


ISPRS Benchmark on Urban Object Detection and 3D Building Reconstruction

Principal Investigator: Markus Gerke; University of Twente, The Netherlands

A major yet unsolved research topic in photogrammetry and remote sensing is automated extraction of urban objects from high-resolution data. This project aimed at promoting 2D semantic segmentation that assigns object-class labels of multiple relevant urban categories to image pixels. It can be considered as a first step towards the long-term goal of completely automating mapping. In the first year of this project, the Vaihingen data set was prepared and the 2D semantic labeling challenge was launched. In the second year, the major labor work went into the creation of data for the extension of the 2D semantic labelling benchmark for Potsdam, Germany. As of mid-December 2015, 20 submissions were made for the Vaihingen dataset, and two for the Potsdam semantic labeling challenge.
 
Full Report


Global Status of Land Cover Mapping and Geospatial Database Updating

Principal Investigator: Gottfried Konecny; Leibniz Universität Hannover, Germany

Since 2012, ISPRS WG IV/2 has concentrated its efforts on a global survey of topographic mapping and geodatabase updating, initiated by ISPRS in cooperation with UNGGIM at the UN Statistics Division, New York. A questionnaire comprising 27 questions was sent to all UN member states and 115 responses were received. For each of the 27 questions, an interactive viewer was programmed permitting analysis of the results. The project concluded with the publication of the final report in 2015 and the delivery of the database to UNGGIM and GEO.
 
Full Report


Advances in the Development of an Open-source Photogrammetric Tool

Principal Investigator: Diego González-Aguilera; University of Salamanca, Spain

Photogrammetry is facing new challenges and changes and the Scientific Community is replying with new algorithms and methodologies for the automated processing of imagery. However, such advances are often limited to in-house research activities due to the intrinsic difficulties of reliably and efficiently implementing such solutions and/or the prior theoretical knowledge required for a proper use. Therefore non-expert users outside the field of photogrammetry have difficulties to access these solutions and apply them to their specific applications to support their problem solving and decision making. With the aim of providing a solution in this context, an open photogrammetric platform, called GRAPHOS (inteGRAtedPHOtogrammetric Suite), has been developed. GRAPHOS faciliatates the creation of dense and metric 3D point clouds from images and it encloses robust photogrammetric and computer vision algorithms with the aim of achieving increased automation and flexibility, as well as improved quality. Last but not least, the educational component has been reinforced with didactic explanations about algorithms and their performance. The expert user is able to test different parameters and configurations in order to assess and compare different approaches.
 
Full Report


Assessment of Learning Pedagogy in GeoInformatics

Principal Investigator: Anjana Vyas; CEPT University, India

New technology plays a vital role in education and training. The use of this technology in the natural and built environment helps to improve quality of life. This project has studied the prevailing pedagogy of learning and discussed the issues and challenges against theories and practices for academic institutions, students, and the geospatial industry.
 


A Method for Vicarious Radiometric Calibration of Orbital Hyperspectral Sensors

Principal Investigator: Eyal Ben Dor; Tel Aviv University, Israel

To make effective use of satellite hyperspectral remote sensing (HRS), data must first be converted into physical and chemical properties to enable a quantitative analysis. As HRS sensor performance may vary in time and space, a vicarious calibration (VC) method to retrieve reliable and accurate at-sensor radiance values is needed. This research has led to the development of VC via a unique method for assessing cross-calibration/validation of space-based hyperspectral data. Best practices for cal/val campaigns were promoted and a protocol on how to evaluate and correct possible radiometric drift in orbital HRS sensors was developed. The study involved the use of airborne and spaceborne data, as well as the preparation of special ground targets composed of agricultural nets with varying densities to simulate radiance intensity. Two areas in the desert region of Israel were selected to examine the VC methodology. The results showed a promising capability to assess the quality of HRS data, as well as to rectify the data by radiance correction, thereby generating better thematic products.
 
Full Report


The total ISPRS funding of Scientific Initiatives for 2014 and 2015 exceeded 80,000 Swiss Francs. Reports on projects funded by the 2014 Scientific Initiative call can also be found on the ISPRS website.