Deep learning-based key-block classification framework for discontinuous rock slopes

Zhu, Honghu
Azarafza, Mohammad
Akgün, Haluk
© 2022 Institute of Rock and Soil Mechanics, Chinese Academy of SciencesThe key-blocks are the main reason accounting for structural failure in discontinuous rock slopes, and automated identification of these block types is critical for evaluating the stability conditions. This paper presents a classification framework to categorize rock blocks based on the principles of block theory. The deep convolutional neural network (CNN) procedure was utilized to analyze a total of 1240 high-resolution images from 130 slope masses at the South Pars Special Zone, Assalouyeh, Southwest Iran. Based on Goodman's theory, a recognition system has been implemented to classify three types of rock blocks, namely, key blocks, trapped blocks, and stable blocks. The proposed prediction model has been validated with the loss function, root mean square error (RMSE), and mean square error (MSE). As a justification of the model, the support vector machine (SVM), random forest (RF), Gaussian naïve Bayes (GNB), multilayer perceptron (MLP), Bernoulli naïve Bayes (BNB), and decision tree (DT) classifiers have been used to evaluate the accuracy, precision, recall, F1-score, and confusion matrix. Accuracy and precision of the proposed model are 0.95 and 0.93, respectively, in comparison with SVM (accuracy = 0.85, precision = 0.85), RF (accuracy = 0.71, precision = 0.71), GNB (accuracy = 0.75, precision = 0.65), MLP (accuracy = 0.88, precision = 0.9), BNB (accuracy = 0.75, precision = 0.69), and DT (accuracy = 0.85, precision = 0.76). In addition, the proposed model reduced the loss function to less than 0.3 and the RMSE and MSE to less than 0.2, which demonstrated a low error rate during processing.
Journal of Rock Mechanics and Geotechnical Engineering


Probabilistic-numerical modeling of stability of a rock slope in Amasya-Turkey
Gheibie, Sohrab; Akgün, Aykut; Department of Mining Engineering (2011)
Rock slope stability is considered as one of the most important fields in rock engineering. Developments of computation facilities and increase in application of sophisticated mathematical concepts in engineering problems have also affected the methods of slope stability analysis. In recent years, the numerical modeling methods have extensively applied instead of limit equilibrium methods. Also, the probabilistic methods are considered in rock slope designs to quantify the uncertainties of input effecting v...
Comparative assessment of the efficiency of seismic isolation for seismic retrofitting of highway bridges in regions of low-to-moderate seismicity
Karalar, Memduh; Dicleli, Murat (null; 2019-07-06)
In this paper, the economical and structural efficiency of friction pendulum bearings (FPB) for retrofitting typical seismically vulnerable bridges in the State of Illinois is studied. For this purpose, a bridge was selected by the Illinois Department of Transportation (IDOT) to represent typical seismically vulnerable bridges commonly used in the State of Illinois. A comprehensive structural model of the bridge was first constructed for seismic analysis. An iterative multi-mode response spectrum (MMRS) ana...
Numerical comparison of retaining system behavior for a deep excavation case with and without ground improvement
Maghsoudloo, Arash; Ahmadinaghadeh, Reza; Toker, Nabi Kartal (2012-11-02)
In the construction of deep excavations in urban areas, the safety of adjacent ground and structures becomes major concern for engineers. In soft clays, the main reason for occurrence of large deflections of soil support systems in excavations is instability of the excavation base. This paper will focus on analyzing and comparing design of an excavation with and without jet grout improvement applied to the excavation base by employing the finite-element code PLAXIS. A well-documented case study is analyzed,...
Numerical modeling of cold-formed steel sheathed shear walls under lateral loading
Topçuoğlugil , Yağmur; Baran, Eray; Department of Civil Engineering (2019)
Cold-formed steel (CFS) structural systems have been used increasingly in seismically active regions. Shear walls consisting of CFS framing and sheathing panels constitute one type of lateral force resisting system for such CFS framed buildings. The purpose of this study is to develop an efficient numerical model to predict the lateral load response of shear walls that are framed with CFS profiles and sheathed with oriented strand board (OSB) panels. Numerical modeling of shear walls was conducted in OpenSe...
Metaheuristic based backcalculation of rock mass parameters around tunnels
Gedik, Görkem; Pekcan, Onur; Department of Civil Engineering (2018)
Due to uncertainities in the ground conditions and the complexity of soil-structure interactions, the determination of accurate ground parameters, which are not only used in tunnel construction but in the design of all underground structures, have a great significance in having structures that are cost-efficient. Backcalculation methods which rely not only on laborotory and field tests but also on field monitoring and field data provide real structure conditions and therefore it is gaining popularity in geo...
Citation Formats
H. Zhu, M. Azarafza, and H. Akgün, “Deep learning-based key-block classification framework for discontinuous rock slopes,” Journal of Rock Mechanics and Geotechnical Engineering, vol. 14, no. 4, pp. 1131–1139, 2022, Accessed: 00, 2022. [Online]. Available: