Yeum, C. M., Mohan, A., Dyke, S. J., Jahanshahi, M., Choi, J., Zhao, Z., Lenjani, A., & Ramirez, J. A. (2017). Image-based collection and measurements for construction pay items (Joint Transportation Research Program Publication No. FHWA/IN/JTRP-2017/10). West Lafayette, IN: Purdue University. https://doi.org/10.5703/1288284316510
Prior to each payment to contractors and suppliers, measurements are made to document the actual amount of pay items placed at the site. This manual process has substantial risk for personnel, and could be made more efficient and less prone to human errors. In this project a customized software tool package has been developed to address these concerns. The Pay Item Measurement (PIM) tool package is provided as two complementary tools, the Orthophoto Generation Tool and the Graphical Measurement Tool. PIM has been developed in close cooperation with the advisory committee and field engineers from INDOT, and is specifically designed to incorporate the typical actions that INDOT personnel follow. PIM is intended to generate orthophotos for measurements on a planar surface. User guidelines explain the process of how to collect suitable high-quality images, which is a critical step for successful orthophoto construction. INDOT will also use PIM to identify features and make annotations, and to readily compute distances, perimeters, and areas for documenting and recording. This customized tool package will be most useful, and accurate, when the user guidelines are followed. Several examples are included to demonstrate the characteristics of high-quality image sets that will be successful, and to also provide examples of sets that would fail. Step-by-step instructions are provided to demonstrate the correct use of the tool. An instructional video and sample digital image sets complement this report and tool package.
pay items, computer vision, structure-from-motion, orthophoto
Joint Transportation Research Program
Indiana Department of Transportation
West Lafayette, Indiana
Date of this Version