Mobile QR Code QR CODE

Journal of the Korea Concrete Institute

J Korea Inst. Struct. Maint. Insp.
  • Indexed by
  • Korea Citation Index (KCI)

Journal Search

XML PDF INFO REF
Title Performance and Repeated Training Stability of CNN-Based Concrete Damage Classification under Optimizer?Initial Learning Rate Combinations
Authors 김일순(Il Sun Kim) ; 양은익(Eun Ik Yang)
DOI https://doi.org/10.11112/jksmi.2026.30.4.166
Page pp.166-175
ISSN 2234-6937
Keywords 콘크리트 손상 분류; 합성곱 신경망; 초기 학습률; 최적화 알고리즘; 반복 학습 안정성 Concrete damage classification; Convolutional neural network; Initial learning rate; Optimizer; Repeated training stability
Abstract This study compared the classification performance and repeated training stability of convolutional neural network (CNN) models for concrete damage image classification under different optimizer and initial learning rate conditions. The target damage types consisted of three classes: crack, efflorescence, and rebar exposure. Four transfer learning-based CNN models, namely EfficientNet-B0, GoogLeNet, MobileNetV2, and ResNet-50, were applied. Adam, RMSProp, and SGDM were used as optimizers, and the initial learning rates were set to 0.001, 0.0003, and 0.0001. For each model?optimizer?learning rate combination, repeated training was performed by changing the random seed 10 times, resulting in a total of 360 training runs. The performance was evaluated using the Test Macro F1-score, standard deviation, minimum value, coefficient of variation, and computational time. The results showed that all CNN models achieved an average Test Macro F1-score higher than 0.94, while the differences in average performance among the models and optimizers were limited. However, performance degradation and repeated training variability were observed in specific model?optimizer?learning rate combinations. In terms of damage type, the rebar exposure class showed the highest average F1-score, whereas the efflorescence class showed relatively lower performance. Therefore, in CNN-based concrete damage classification, models and training conditions should be selected by considering not only average performance but also repeated training stability, minimum performance, class-wise performance, and computational efficiency.