The objective of Program 2 was to conduct research into automated feature extraction from aerial, satellite and terrestrial imaging and ranging systems. To early 2015, the research activity within Program 2, focused on Automated Spatial Information Generation concentrated within three main projects:
- 2.01: Multimodal data acquisition and feature extraction from multi-sensor terrestrial mobile mapping systems
- 2.02: Feature Extraction from Multi-Source Airborne and Space-Borne Imaging and Ranging Data
- 2.07: Australian Woody Vegetation Landscape Feature Generation
The three projects have been completed in 2014-15, leading to a new focus in Program 2 on Rapid Spatial Analytics. The new program has undergone some initial activities in 2014-15, but begins in earnest in 2015-16. A summary of the completed activities in the Automated Spatial Information Generation program is provided.
Project 2.01 researched and developed techniques and algorithms for the detection, recognition and analysis of low-level and high-level features from mobile mapped data along urban and rural transport corridors. Mobile mapping data is acquired from near ground vehicle sensors that gather information at a high resolution. Low-level features include edges, corners, planar surfaces, and high level features include buildings, road signs and the transport corridor reserve.
The research used fused image, video and laser data acquired from mobile platforms to recognise, locate and measure the various features of interest. The 2.01 project team, under the direction of the Project Leader, Prof. Geoff West, comprised one research associate, Dr David Belton, and four PhD students Dr. Abdul Nurrunabi (graduated), Richard Palmer, Michael Borck, and Cynthia Yu. Project 2.01 has now been successfully concluded in terms of the research carried out and the various outputs. Matching up the research outcomes with utilisation by the project partners, however, is still underway, with the progress in this area being slower than originally anticipated.
Project 2.02 advanced the technology and practise of automated processes within topographic mapping, geo-database generation and updating, and change detection, through enhanced feature extraction from airborne and space-borne imagery and from airborne laser ranging (LiDAR) data. The project concentrated upon automated 3D reconstruction of man-made objects and vegetation parameters through development of new mathematical models, enhanced algorithms, improved computational processes and new software systems for integrated processing of imagery and ranging data, improved higher-level feature extraction, and robust object reconstruction.
Three PhD students were engaged on the project, under the supervision of the Project Leader, Dr Chunsun Zhang. Dr Mohammad Awrangjeb was also engaged on the project as a part-time research associate. Dr Ebadat Parmehr and Dr Yuxiang He completed their PhDs in 2014-15.
A number of the developed algorithms were both implemented in the CRCSI-developed Barista software, and made available in the form of software libraries for integration by industry partners. Project 2.02 was wound up in early 2015, having achieved its principal objectives and met its important commonwealth milestones.
The final of the three major Program project, Project 2.07 was also successfully completed in May 2015. The project developed processes to characterise woody vegetation ecosystems through automated feature generation, using a combination of ground (field), airborne and space-borne image and ranging data. The Woody Vegetation Project produced tools and procedures to auto-generate landscape level woody vegetation features, such as spatial layers, from field and remote sensing woody vegetation data. The metrics are assessed to inform carbon accounting, biodiversity and ecosystem health, and fire management. The open source tools have been developed for sustainable land management decision-making and monitoring, mapping and natural resource management activities. Two PhD students were engaged on the project, under the supervision of Project Leaders, Prof Simon Jones and Dr Andrew Haywood.
Dr Lola Suarez was engaged on the project as a research fellow. Dr Will Woodgate (graduated) and Phil Wilkes were engaged as the PhD students on the project.
Additional projects which were added later to the Automated Spatial Information Generation will continue to 2016: