Remote Sensing for Hazard Assessment

 

Summary of activities


Remote sensing is a technology to sense phenomena without contact.  Broadly speaking, it includes two branches, i.e. photogrammetry and remote sensing.  The former is more concerned with sensing the geometric information of phenomena and the more concerned with physical (thematic) information.  The computer tomography (CT) technology, widely used in hospitals, is a kind of photogrammetric product.

In photogrammetry, 3-D surfaces are reconstructed from a pair of so-called stereo images, which were taken of the same object but from two slightly different angles.  It makes use of the optical geometry of two images.  On the other hand, in remote sensing, a number of sensors are used to take images.  Each sensor records the energy of electromagnetic waves of limited wavelength (or frequency).  The waves would be radiated from or reflected by phenomena.  Remote sensing makes use of the reflectance of different waves by phenomena

Remote sensing has wide applications in various areas such as mapping, water, hydrology, groundwater, coastal zone, marine, ice/snow, land use/cover, ecology/environmental studies, urbanisation, hazardous waste, marketing, soil/geology, exploration, mapping, transportation, structure/geomorphology, flood control, site suitability, vegetation, agriculture, forestry, rangeland management, weather/climate, archaeology, ecosystem health management.

The main purpose of using remote sensing is to acquire information of phenomena for monitoring, assessment, management and/or planning.  It has be used for monitoring of and for assessment of the damages caused by various types of hazards,  including collapse of an engineering construction (buildings, bridges, etc), slope-sliding and landslides (Figure 1), geological (e.g. earthquakes and volcano, etc), and oceanic and atmospheric hazards (e.g. El Nino events and red tides), fire outbreaks, and so on.

Damage by land slide

 

The remote sensing group of the Department of Land Surveying and Geo-Informatics, the Faculty of Construction and Land Use, is a strong team in this area.  Photogrammetry is used to monitor the deformation of structures and slope (Figure 2), and computation of slope/land slide volumes.

 

Figure 2  Photogrammetry for modelling slope and its deformation

 

Aerial photographs and high resolution satellite images can be used to detect both small and large landslide features (Figure 3).  Indirect methods of monitoring involve integration of remotely sensed observations with GIS data on landslide activity, landslide phases, slope angle, material, proximity to drainage channels and reservoirs, type and amount of vegetation cover, and land use.  The data are then combined using multi-criteria  evaluation techniques.

 

Figure 3. IKONOS panchromatic image November 2000, showing avalanche tracks (A) on slope and debris accumulation (D) at foot of slope, near to housing development and road at Quarry Bay

 

Satellite images to assess the damage bush fire (Figure 3). By using multi-temporal satellite data the areas burnt every year can be compared with other prediction variables such as altitude, slope, nearness to settlements, road access and grave sites, and the areas most susceptible to burning in a particular year can be flagged for extra precautionary measures to be taken. The exclusion of fire is the only way Hong Kong's natural flora and fauna can be restored as well as the problems of soil erosion, sedimentation and landslides be mitigated.


 

Figure4. LANDSAT Thematic Mapper colour composite, bands 2, 4 and 6 with band 6 (thermal band) displayed as red and band 4 (visible infra-red) displayed as green. Red areas thus represent hot spots and correspond to areas of grassland which have been burnt during the recent dry season.

 

There are also a lot of related activities such as reconstruction of 3D buildings from satellite images, remote sensing for water quality analysis, remote sensing for urban heat analysis, remote sensing for pollution analysis.

Recently, the group has purchased a terrestrial laser scanning system.  This systems can be used for monitoring structure damage and deformation, and slope deformation (Figures 5 and 6).

 

Figure 5   Terrestrial laser scanning system for slope monitoring

 

Figure 6  Terrestrial laser scanning system for structure monitoring

Representative Projects

  • Terrestrial laser scanning system for structure monitoring and modelling;
  • Structure monitoring with close range photogrammetry;
  • Reconstruction of urban buildings from satellite images;
  • Remote sensing and GIS for mapping of avalanche and slope failure hazard areas, and soil erosion;
  • Fire hazard mapping and monitoring in Hong Kong's country parks using visible and thermal infra-red satellite imagery;
  • Remote sensing for urban heat and pollution analysis;
  • Remote sensing for water quality analysis;

Major Facilities

  • 2 digital photogrammetric workstations
  • 1 terrestrial laser scanning system
  • A number of CCD cameras
  • A number of PC-based digital photogrammetric systems
  • A number of remote sensing image processing systems
  • A number modelling systems for 3D visualisation

Selective Publication

  • Huang, G. Z. and King, Bruce, 1998.  A methodology for oblique close range photogrammetry.  Bulletin of Surveying and Mapping, 1998(9): 17-19. (in Chinese).
  • Li, Zhilin, Yuan, X.X., and Lam, K., 2002.  Effect of JPEG compression on the Accuracy of Photogrammetric point determination. Photogrammetric Engineering and Remote Sensing, (in press).
  • LI, Zhilin, XU, Zhu, CHEN Minyi and DING, Xiaoli, 2001. Robust Surface Matching for Automated Detection of Local Deformations Using Least Median of Squares Estimator.  Photogrammetric Engineering and Remote Sensing, 67(11): 1283-1292.
  • Mok, Eemond., King, Bruce., Zhu, Qing., Li, Zhilin. and Lam, Kent., 1998.  Digital Close Range PhotogrammetryicData Acquisition System with Combined GPS Geo-reference and Its Calibration. International Archives for Photogrammetry and Remote Sensing, Vol.32(5):, 253-257.
  • Nichol, J.E., 1993. Remote Sensing of water quality in the Singapore-Johor-Riau growth triangle, Remote Sensing of Environment,  U.S.A., 43: 139-148.
  • Nichol, J.E., 1993. Remote Sensing of tropical blackwater rivers: a method for environmental water quality assessment.  Applied Geography, U.K. (1993), 13:153-168.
  • Yip, S.W., 2000.  Computer modelling of the Han Tomb at Lei Cheng Uk by photogrammetric survey, Proceedings of the 3rd Across-the-strait Geomatics Conference, Hong Kong, 11-14 December 2000. 210-218.

Contact Information of the contact person

Dr Z.L. Li, Associate Professor
Department of Land Surveying and Geo-Informatics
The Hong Kong Polytechnic University
Kowloon, Hong Kong

Fax:     + 852 - 2330 2994
Phone:   + 852 - 2766 5960
E-mail:  lszlli@polyu.edu.hk

:: Back to Research Areas

 

Research Centre for Urban Hazards Mitigation