Detection of streamflow change to prevent residents from potential disasters

Research output: Contribution to conferencePaperpeer-review

Abstract

Remote sensing technologies have been applied to numerous fields and already benefited those in improving efficiency and recognition accuracy. Change of streamflow can be devastating to residents living along riversides. Still it is difficult to detect such deathly change in large area scale by traditional field surveys. The research aims at provision of detecting streamflow change using a mathematic method based on remote sensing technologies. The scope is on remote sensing stream images taken during the invasion by Typhoon Morakot, bringing one of the major catastrophes to Taiwan. An algorithm was adopted to deal with streamflow recognition on RGB images. The algorithm is capable of yielding change measured by meters and illustrates how main streams shift. An practical example for the Baolai area surrounded by Laonong stream shows that the streamflow moved 88 meters (36 pixels) toward the downtown area during the invasion by Typhoon Morakot. In addition, the results highlight all dangerous zones (scouring zones) for residents to prevent further potential damage.

Original languageEnglish
Pages102-109
Number of pages8
StatePublished - 2013
Event30th International Symposium on Automation and Robotics in Construction and Mining, ISARC 2013, Held in Conjunction with the 23rd World Mining Congress - Montreal, QC, Canada
Duration: 11 Aug 201315 Aug 2013

Conference

Conference30th International Symposium on Automation and Robotics in Construction and Mining, ISARC 2013, Held in Conjunction with the 23rd World Mining Congress
Country/TerritoryCanada
CityMontreal, QC
Period11/08/1315/08/13

Keywords

  • Automatic detection
  • Disaster prevention
  • Pattern recognition
  • Remote sensing
  • RGB image

Fingerprint

Dive into the research topics of 'Detection of streamflow change to prevent residents from potential disasters'. Together they form a unique fingerprint.

Cite this