Prediction of Asphalt Mixture Resistance Using Neural Network via Laboratorial X-ray Images
Fereidoon Moghadas Nejad1, Ahmad Mehrabi1,2, and Hamzeh Zakeri1,2
1.Dept. of Civil and Environment Engineering, Amirkabir Univ. of Technology, Tehran, Iran
2.Department of Civil and Environmental Engineering, Islamic Azad University, Science and Research Branch, Tehran, Iran
2.Department of Civil and Environmental Engineering, Islamic Azad University, Science and Research Branch, Tehran, Iran
Abstract—This paper presents a relation between Marshall Resistance of asphalt samples and their visual and physical features. Laboratorial samples with known mixture scheme are prepared in laboratory and are digitized through a CT-scan system. The physical features of asphalt, captured by scanning the asphalt samples, include air void percentage, bitumen percentage, aggregate percentage, arrangement and the direction of aggregate .Then some analytically model using visual characteristics and artificial intelligence methods are proposed. Finally the results from attained multi-pages image processing are related to laboratory parameters. The proposed model provides closed-form solution for predicting the Marshall resistance of asphalt mixtures.
Index Terms—artificial neural network; C.T scan images; asphalt concrete; marshall stability
Cite: Fereidoon Moghadas Nejad, Ahmad Mehrabi, and Hamzeh Zakeri, "Prediction of Asphalt Mixture Resistance Using Neural Network via Laboratorial X-ray Images," Journal of Industrial and Intelligent Information, Vol. 3, No. 1, pp. 48-53, March 2015. doi: 10.12720/jiii.3.1.48-53
Index Terms—artificial neural network; C.T scan images; asphalt concrete; marshall stability
Cite: Fereidoon Moghadas Nejad, Ahmad Mehrabi, and Hamzeh Zakeri, "Prediction of Asphalt Mixture Resistance Using Neural Network via Laboratorial X-ray Images," Journal of Industrial and Intelligent Information, Vol. 3, No. 1, pp. 48-53, March 2015. doi: 10.12720/jiii.3.1.48-53