Brilakis, I and Soibelman, L (2005) Comparison of manual and user guided methodologies for the classification and retrieval of construction site images. In: the ASCE Construction Research Congress, 5-4-2005 to 7-4-2005, San Diego, CA, United States of America pp. 1-10..Full text not available from this repository.
The amount of original imaging information produced yearly during the last decade has experienced a tremendous growth in all industries due to the technological breakthroughs in digital imaging and electronic storage capabilities. This trend is affecting the construction industry as well, where digital cameras and image databases are gradually replacing traditional photography. Owners demand complete site photograph logs and engineers store thousands of images for each project to use in a number of construction management tasks like monitoring an activity's progress and keeping evidence of the "as built" in case any disputes arise. So far, retrieval methodologies are done manually with the user being responsible for imaging classification according to specific rules that serve a limited number of construction management tasks. New methods that, with the guidance of the user, can automatically classify and retrieve construction site images are being developed and promise to remove the heavy burden of manually indexing images. In this paper, both the existing methods and a novel image retrieval method developed by the authors for the classification and retrieval of construction site images are described and compared. Specifically a number of examples are deployed in order to present their advantages and limitations. The results from this comparison demonstrates that the content based image retrieval method developed by the authors can reduce the overall time spent for the classification and retrieval of construction images while providing the user with the flexibility to retrieve images according different classification schemes.
|Item Type:||Conference or Workshop Item (UNSPECIFIED)|
|Depositing User:||Cron Job|
|Date Deposited:||24 Aug 2012 06:10|
|Last Modified:||04 Feb 2013 01:54|
Actions (login required)