Abstract
The semantic segmentation for irregularly and not uniformly disposed patterns becomes even more difficult when the occurrence of categories is imbalanced within the images. One example is represented by heavily corroded artefacts in archaeological digs. The current study therefore proposes a weighted loss function within a deep learning architecture for semantic segmentation of corrosion compounds from microscopy images of archaeological objects, where the values for the class weights are generated via genetic algorithms. The fitness evaluation of individuals is the estimation that a surrogate of the deep learner gives concerning the segmentation accuracy. The obtained class weight values are compared to a random search through the space of potential configurations and another automated means to compute them, in terms of resulting model accuracy.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Arun, C., Lakshmi, C.: Genetic algorithm-based oversampling approach to prune the class imbalance issue in software defect prediction. Soft Comput. 26(23), 12915–12931 (2022)
Bacanin, N., Stoean, R., Zivkovic, M., Petrovic, A., Rashid, T.A., Bezdan, T.: Performance of a novel chaotic firefly algorithm with enhanced exploration for tackling global optimization problems: application for dropout regularization. Mathematics 9(21), 1 (2021)
Bartz-Beielstein, T., Chandrasekaran, S., Rehbach, F.: Case Study III: Tuning of Deep Neural Networks, pp. 235–269. Springer Nature Singapore, Singapore (2023)
Bi, J., Zhang, C.: An empirical comparison on state-of-the-art multi-class imbalance learning algorithms and a new diversified ensemble learning scheme. KBS 158, 81–93 (2018)
Dutta, A., Zisserman, A.: The VIA annotation software for images, audio and video. In: 27th ACM International Conference on Multimedia, p. 4 (2019)
Forkan, A.R.M., et al.: Corrdetector: a framework for structural corrosion detection from drone images using ensemble deep learning. Expert Syst. Appl. 193, 116461 (2022)
Gad, A.F.: PyGAD: An intuitive genetic algorithm python library. https://arxiv.org/abs/2106.06158 (2021)
Katsamenis, I., Protopapadakis, E., Doulamis, A., Doulamis, N., Voulodimos, A.: Pixel-level corrosion detection on metal constructions by fusion of deep learning semantic and contour segmentation. In: Bebis, G., et al. (eds.) ISVC 2020. LNCS, vol. 12509, pp. 160–169. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-64556-4_13
King, G., Zeng, L.: Logistic regression in rare events data. Polit. Anal. 9, 137–163 (2001)
Kumar, P., Batra, S., Raman, B.: Deep neural network hyper-parameter tuning through twofold genetic approach. Soft Comput. 25(13), 8747–8771 (2021). https://doi.org/10.1007/s00500-021-05770-w
Pedregosa, F., et al.: Scikit-learn: machine learning in python. JMLR 12, 2825–2830 (2011)
Postavaru, S., Stoean, R., Stoean, C., Caparros, G.J.: Adaptation of deep convolutional neural networks for cancer grading from histopathological images. In: Rojas, I., Joya, G., Catala, A. (eds.) IWANN 2017. LNCS, vol. 10306, pp. 38–49. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-59147-6_4
Rehbach, F., Zaefferer, M., Naujoks, B., Bartz-Beielstein, T.: Expected improvement versus predicted value in surrogate-based optimization. In: The 2020 Genetic and Evolutionary Computation Conference, pp. 868–876 (2020)
Roshan, S.E., Asadi, S.: Improvement of bagging performance for classification of imbalanced datasets using evolutionary multi-objective optimization. Eng. Appl. Artif. Intell. 87, 103319 (2020)
Samide, A., Stoean, C., Stoean, R.: Surface study of inhibitor films formed by polyvinyl alcohol and silver nanoparticles on stainless steel in hydrochloric acid solution using convolutional neural networks. Appl. Surf. Sci. 475, 1–5 (2019)
Samide, A., Stoean, R., Stoean, C., Tutunaru, B., Grecu, R.: Investigation of polymer coatings formed by polyvinyl alcohol and silver nanoparticles on copper surface in acid medium by means of deep convolutional neural networks. Coatings 9, 105 (2019)
Stoean, C., Stoean, R., Sandita, A., Ciobanu, D., Mesina, C., Gruia, C.L.: SVM-based cancer grading from histopathological images using morphological and topological features of glands and nuclei. In: De Pietro, G., Gallo, L., Howlett, R.J., Jain, L.C. (eds.) Intelligent Interactive Multimedia Systems and Services 2016. SIST, vol. 55, pp. 145–155. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-39345-2_13
Stoean, R.: Analysis on the potential of an EA-surrogate modelling tandem for deep learning parametrization: an example for cancer classification from medical images. Neural Comput. Appl. 32, 313–322 (2020)
Stoean, R., Bacanin, N., Stoean, C., Ionescu, L., Atencia, M., Joya, G.: Computational framework for the evaluation of the composition and degradation state of metal heritage assets by deep learning. J. Cult. Heritage (under review) (2023)
Tanha, J., Abdi, Y., Samadi, N., Razzaghi, N., Asadpour, M.: Boosting methods for multi-class imbalanced data classification: an experimental review. J. Big Data 7(1), 1–47 (2020). https://doi.org/10.1186/s40537-020-00349-y
Zhang, M., Li, H., Pan, S., Lyu, J., Ling, S., Su, S.: Convolutional neural networks-based lung nodule classification: a surrogate-assisted evolutionary algorithm for hyperparameter optimization. IEEE Trans. Evol. Comput. 25(5), 869–882 (2021)
Acknowledgement
This work was supported by a grant of the Romanian Ministry of Research and Innovation, CCCDI – UEFISCDI, project number 178PCE/2021, PN-III-P4-ID-PCE-2020-0788, Object PErception and Reconstruction with deep neural Architectures (OPERA), within PNCDI III.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Stoean, R., Báez, P.G., Araujo, C.P.S., Bacanin, N., Atencia, M., Stoean, C. (2023). Automatic Control of Class Weights in the Semantic Segmentation of Corrosion Compounds on Archaeological Artefacts. In: Rojas, I., Joya, G., Catala, A. (eds) Advances in Computational Intelligence. IWANN 2023. Lecture Notes in Computer Science, vol 14135. Springer, Cham. https://doi.org/10.1007/978-3-031-43078-7_38
Download citation
DOI: https://doi.org/10.1007/978-3-031-43078-7_38
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-43077-0
Online ISBN: 978-3-031-43078-7
eBook Packages: Computer ScienceComputer Science (R0)