TY - GEN
T1 - Design of pothole detection system based on digital image correlation using Kinect sensor
AU - Anggoro, W.
AU - Nasution, A.
AU - Rosohadi, I. R.
N1 - Publisher Copyright:
© 2019 SPIE.
PY - 2019
Y1 - 2019
N2 - Pothole detection system based on digital image correlation had been made using Kinect sensor that could detect and quantified pothole dimensions such as depth, area, volume and position. Pothole detection was start with normalization process on depth image from the pothole that was used as a quantifying object. Then thresholding was done on the normalized image in order to separate between pothole and its background. Dimension quantification was done by making the work model from the sensor characterization result. Next, validation system process was done by testing the artificial pothole with the depth variation of 2 cm, 4 cm, 6 cm, 8 cm, 10 cm and 12 cm, every depth variation had each given diameter of 5 cm, 10 cm, 15 cm, 20 cm, and 25 cm. Based on the results of the validation, the system has an error value of less than 7% on the hole diameter range of 5-25cm with a depth range of 6-12 cm. The depth resolution of the system is 0.43 cm while the system spatial resolution depends on the distance between sensor and the quantifying object, for sensor system with distance span of 120 cm would have spatial resolution on horizontal and vertical axis of 0.21 cm and 0.22 cm respectively.
AB - Pothole detection system based on digital image correlation had been made using Kinect sensor that could detect and quantified pothole dimensions such as depth, area, volume and position. Pothole detection was start with normalization process on depth image from the pothole that was used as a quantifying object. Then thresholding was done on the normalized image in order to separate between pothole and its background. Dimension quantification was done by making the work model from the sensor characterization result. Next, validation system process was done by testing the artificial pothole with the depth variation of 2 cm, 4 cm, 6 cm, 8 cm, 10 cm and 12 cm, every depth variation had each given diameter of 5 cm, 10 cm, 15 cm, 20 cm, and 25 cm. Based on the results of the validation, the system has an error value of less than 7% on the hole diameter range of 5-25cm with a depth range of 6-12 cm. The depth resolution of the system is 0.43 cm while the system spatial resolution depends on the distance between sensor and the quantifying object, for sensor system with distance span of 120 cm would have spatial resolution on horizontal and vertical axis of 0.21 cm and 0.22 cm respectively.
KW - Detection and Quantification of Road Damage
KW - Digital image correlation
KW - Microsoft Kinect
KW - Potholes
UR - http://www.scopus.com/inward/record.url?scp=85065618897&partnerID=8YFLogxK
U2 - 10.1117/12.2504739
DO - 10.1117/12.2504739
M3 - Conference contribution
AN - SCOPUS:85065618897
T3 - Proceedings of SPIE - The International Society for Optical Engineering
BT - Third International Seminar on Photonics, Optics, and Its Applications, ISPhOA 2018
A2 - Hatta, Agus Muhammad
A2 - Nasution, Aulia
PB - SPIE
T2 - 3rd International Seminar on Photonics, Optics, and Its Applications, ISPhOA 2018
Y2 - 1 August 2018 through 2 August 2018
ER -