Coastal image classification and analysis based on convolutional neural networks and pattern recognition

Dr Bo Liu1,2, Bin Yang2, Professor Giovanni Coco1, Huina  Wang2, Sina Masoud-Ansari1, Professor Mark  Gahegan1

1Univ. Of Auckland, Auckland, New Zealand
2Beijing University of Technology, Beijing , China

The study of coastal processes is critical for the protection and development of beach amenities, infrastructure and properties. Many studies of beach evolution rely on data collected using remote sensing and show that beach evolution can be characterized by a finite number of “beach states”. However, due to practical constraints, long-term data displaying all beach states are rare. Also, when the dataset is available, the accuracy of the classification is not entirely objective since it depends on the operator. To address this problem, we have collected hourly coastal images and corresponding tidal data for more than 20 years (Nov. 1998-Aug. 2019). We classified the coastal images into 8 categories according to the classic beach state classification, defined as 1)reflective, 2)incident  scaled  bar, 3) non-rhythmic,  attached  bar, 4)attached rhythmic  bar, 5)offshore  rhythmic  bar, 6)non-rhythmic,  3-D  bar, 7)infragravity  scaled  2-D  bar, 8)dissipative. Classification models are usually based on convolutional neural networks. After image pre-processing with data enhancement, we have compared Densenet, Resnet, Resnext and improved Resnext models. The improved Resnext obtained the best and most stable classification with an accuracy of 94.58% and good generalization ability. The classification results of the whole dataset are transformed into time series data. FP-Growth and MDLats algorithms are used to find frequent patterns and motifs which represent the pattern of coastal morphology changes within a certain period of time. Combining the pattern of coastal morphology change and the corresponding tidal data, we also analysed the characteristics of beach morphology and the changes in morphological dynamics states.


Bo Liu is a visiting scholar at School of Computer Science, the University of Auckland, and an associate professor of School of Software Engineering, Beijing University of Technology. She received Ph.D. degree from the Department of Automation, Tsinghua University. She once worked in NEC Laboratory China as a researcher and at the University of Chicago and Argonne National Laboratory as a Research Professional. She joined Beijing University of Technology in 2015 as an associate professor. Her research interests include big data, data mining, machine learning, cloud computing, scientific workflow and Semantic Web. She has authored over 100 articles and inventions.

Recent Comments