PCIer: Pavement Condition Evaluation Using Aerial Imagery and Deep Learning

This paper aims to explore and evaluate aerial imagery deep learning technology in pavement condition evaluation. A convolutional neural network (CNN) model, named PCIer, was designed process images produce index (PCI) estimations, which are classified into four scales of Good (PCI ≥ 70), Fair (50 ≤ PCI < Poor (25 50), Very 25). In the experiment, datasets were retrieved from published report by City Sacramento, CA. Following datasets, authors also collected corresponding image containing 100 for each grade Google Earth. An 80% proportion used PCIer model training, remaining testing. Comparisons showed using a 128-channel heatmap layer proposed saving with best validation accuracy would yield performance, testing 0.97, weighted average precision, recall, F1-score 0.98, respectively. Moreover, future research recommendations provided discussion improving effectiveness evaluation via learning.