Aggregate Weight Prediction Based on Two-dimensional Image Feature Extraction. Issue 1 (February 2021)

• Record Type: Journal Article
• Authors: Pei, Lili; Yu, Ting; Yuan, Haochen; Li, Wei; Li, Yuxuan; Hao, Xueli
• Is Part Of: IOP conference series. Volume 668:Issue 1(2021)
• Publication Date: 2021-02
• Subjects: Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/668/1/012069
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: The weight of aggregates of different particle sizes is vital for the online monitoring of the gradation of asphalt mixtures. In order to complete the online intelligent detection of asphalt aggregate gradation, an automatic coarse aggregate weight prediction algorithm based on the feature extraction of aggregate two-dimensional images is proposed. First, use OpenCV to extract the two-dimensional morphological features of coarse aggregate and use high precision electronic gram scale to obtain the actual weight of the aggregate. Then analyze the correlation between these characteristics and aggregate weight. Finally, the weight of coarse aggregate particles can be accurately predicted by establishing a BPNN (Back-Propagation Neural Network) model. The results show that the weight prediction accuracy of coarse aggregate can be achieved 89.49%. The manual weighing process is reduced, which greatly improves the efficiency of online intelligent detection.