Abstract:Surface anomaly detection is a key function in automatic quality control systems for marble industries that produce marble-like tile materials. This study proposes automated quality control in detecting marble tile surface defects using DenseNet-201, a deep convolutional neural network (CNN) model. The performance of various activation functions, such as Rectified Linear Unit (ReLU), Swish, Mish, Gaussian Error Linear Unit (GELU), Activate or Not (ACON-C), Meta-ACON, Snake, Deep Interactive Click Extraction (DICE), and Leaky ReLU (LReLU) was evaluated on publicly available Marble Surface Anomaly Detection dataset from Kaggle over 50 training epochs. This dataset comprises......
Keywords: Anomaly detection, CNN, deep learning, Leaky ReLU, smart tiles factory
[1]
J. Yang, R. Xu, Z. Qi, And Y. Shi, “Visual Anomaly Detection For Images: A Systematic Survey,” Procedia Comput. Sci., Vol. 199, Pp. 471–478, 2022.
[2]
I. Jahan, M. O. Ali, M. H. Rahman, B. Chung And Y. M. Jang, "Vision Anomaly Detection Using Self-Gated Rectified Linear Unit," Proc. 2022 International Conference On Artificial Intelligence In Information And Communication (Icaiic), Jeju Island, Korea, Republic Of, 2022, Pp. 200-203.
[3]
A. T. Prihatno, I. B. K. Y. Utama, J. Y. Kim And Y. M. Jang, "Metal Defect Classification Using Deep Learning," Proc. 2021 Twelfth International Conference On Ubiquitous And Future Networks (Icufn), Jeju Island, Korea, Republic Of, 2021, Pp. 389-393.
[4]
A.-A. Tulbure, A.-A. Tulbure, And E.-H. Dulf, “A Review On Modern Defect Detection Models Using Dcnns - Deep Convolutional Neural Networks,” J. Adv. Res., Vol. 35, Pp. 33–48, 2022.
[5]
S. Y. Lee, B. A. Tama, S. J. Moon, And S. Lee, “Steel Surface Defect Diagnostics Using Deep Convolutional Neural Network And Class Activation Map,” Appl. Sci. (Basel), Vol. 9, No. 24, P. 5449, 2019
