19 Nov 2024

Deep Learning Approach for Accurate Segmentation of Sand Boils in Levee Systems

Abstract: Sand boils can contribute to the liquefaction of a portion of the levee, leading to levee failure. Accurately detecting and segmenting sand boils is crucial for effectively monitoring and maintaining levee systems. This paper presents SandBoilNet, a fully convolutional neural network with skip connections designed for accurate pixel-level classification or semantic segmentation of sand boils from images in levee systems. In this study, we explore the use of transfer learning for fast training and detecting sand boils through semantic segmentation. By utilizing a pretrained CNN model with ResNet50V2 architecture, our algorithm effectively leverages learned features for precise detection. We hypothesize that controlled feature extraction using a deeper pretrained CNN model can selectively generate the most relevant feature maps adapting to the domain, thereby improving performance. Experimental results demonstrate that SandBoilNet outperforms state-of-the-art semantic segmentation methods in accurately detecting sand boils, achieving a Balanced Accuracy (BA) of 85.52%, Macro F1-score (MaF1) of 73.12%, and an Intersection over Union (IoU) of 57.43% specifically for sand boils. This proposed approach represents a novel and effective solution for accurately detecting and segmenting sand boils from levee images toward automating the monitoring and maintenance of levee infrastructure.

12 May 2021

Automated Characterization of Ridge-Swale Patterns Along the Mississippi River

Abstract: The orientation of constructed levee embankments relative to alluvial swales is a useful measure for identifying regions susceptible to backward erosion piping (BEP). This research was conducted to create an automated, efficient process to classify patterns and orientations of swales within the Lower Mississippi Valley (LMV) to support levee risk assessments. Two machine learning algorithms are used to train the classification models: a convolutional neural network and a U-net. The resulting workflow can identify linear topographic features but is unable to reliably differentiate swales from other features, such as the levee structure and riverbanks. Further tuning of training data or manual identification of regions of interest could yield significantly better results. The workflow also provides an orientation to each linear feature to support subsequent analyses of position relative to levee alignments. While the individual models fall short of immediate applicability, the procedure provides a feasible, automated scheme to assist in swale classification and characterization within mature alluvial valley systems similar to LMV.