Abstract
In the task of ancient architecture image classification, due to the unconstrained random initialization of the weight matrix, the existing random projection branches cannot effectively focus on the key feature dimensions such as roof contour and bucket arch structure. This not only weakens the core basis for classification but also impairs the ability to distinguish between similar building variants. Meanwhile, the training mode of inputting all samples at one time is easy to cause the model to over-fit the architectural style of specific regions and weaken the cross-regional generalization performance. To address these issues, we propose an ancient architecture image classification with progressive stacking pseudoinverse learning (AAPSP). It comprises two core modules: the key features stacking pseudoinverse learning of ancient architecture (KFSP) and the progressive optimization learning model of ancient architecture samples (POL). Specifically, KFSP initializes the weight matrix to a specific pattern that conforms to the distribution of architectural features (such as Gaussian distribution focusing on the roof contour, uniform distribution capturing decorative patterns). By combining the attention mechanism (AM), a higher weight is assigned to the base learner of key component recognition to optimize the feature space modeling. POL employs a dynamic sample screening strategy to preferentially select sample iterative optimization models with rare features to reduce redundancy and enhance generalization ability. Experiments on six types of Chinese ancient architecture datasets show that AAPSP performs well in accuracy, precision, recall rate and F1 score.
Similar content being viewed by others
Data availability
The datasets used during the current study are available from the corresponding author on reasonable request.
References
Wang, C. et al. Assessment of multiscale variability and influencing factors of humidity environment in cave temples of Wudang Mountain. J. Cult. Herit. 74, 66–79 (2025).
Jia, S. et al. Conservation and management of Chinese classical royal garden heritages based on 3D digitalization-A case study of Jianxin courtyard in Jingyi garden in Fragrant Hills. J. Cult. Herit. 58, 102–111 (2022).
Liao, B., Zuo, H., Yu, Y. & Li, Y. GraphMriNet: A few-shot brain tumor MRI image classification model based on Prewitt operator and graph isomorphic network. Complex. Intell. Syst. 10(5), 6917–6930 (2024).
Yang, B. et al. Transformer-based multiple instance learning network with 2D positional encoding for histopathology image classification. Complex. Intell. Syst. 11(5), 1–17 (2025).
Ottoni, A. L. C. et al. A deep learning approach for cultural heritage building classification using transfer learning and data augmentation. J. Cult. Herit. 74, 214–224 (2025).
Ramirez-Arellano, A. et al. Deng entropy and LSTM neural network to classify built cultural heritage with severe damage. J. Cult. Herit. 73, 286–294 (2025).
He, L., Wei, S., Zhou, H. & Hu, Q. Defending deep learning models: A hybrid algorithm employing de-noising and coordinate-disruption techniques against adversarial attacks. Complex. Intell. Syst. 11(10), 1–17 (2025).
Yang, J. et al. An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning. Complex. Intell. Syst. 11(8), 1–17 (2025).
Artopoulos, G. et al. An artificial neural network framework for classifying the style of Cypriot hybrid examples of built heritage in 3D. J. Cult. Herit. 63, 135–147 (2023).
Werbos, P. et al. Applications of advances in nonlinear sensitivity analysis. System Modeling and Optimization, Proceedings of the 10th IFIP Conference: 3-12 (1981).
Sun, Y. et al. An exact supervised learning for a three-Layer supervised neural network. Proceedings of the International Conference on Neural Information Processing ICONIP’95 (1995).
Guo, P. et al Pseudoinverse learning algorithm for feedforward neural networks. Advances in neural networks and applications, Athens, Greece: World Scientific and Engineering Society Press 321–326 (2001).
Mezina, A. et al. EnsArtNet: Ensemble neural network architecture for identifying art styles from paintings. J. Cult. Herit. 72, 71–80 (2025).
Guo, P. et al. A case study on stacked generalization with software reliability growth modeling data. Neural information processing-18th international conference iconip 2001 (2001).
Guo, P. et al. A pseudoinverse learning algorithm for feedforward neural networks with stacked generalization applications to software reliability growth data. Neurocomputing 56(1), 101–121 (2004).
Li, S. et al. A hierarchical model with pseudoinverse learning algorithm optimazation for pulsar candidate selection. 2018 IEEE Congress on Evolutionary Computation (CEC) (2018).
Feng, S. et al. An ensemble model for error modeling with pseudoinverse learning algorithm. 2019 IEEE International Conference on Systems, Man and Cybernetics (SMC) (2019).
Affentranger, F. et al. Random projections of regular simplices. Discrete Comput. Geom. 7(3), 219–226 (1992).
Omidiran, D. et al. High-dimensional variable selection with sparse random projections: measurement sparsity and statistical efficiency. J. Mach. Learn. Res. 11, 2361–2386 (2010).
Eftekhari, A. et al. Two-dimensional random projection. Signal Process. 91(7), 1589–1603 (2015).
Young, J. et al. An application of random projection to parameter estimation in partial differential equations. SIAM J. Sci. Comput. 34(4), A2344–A2365 (2012).
Nguyen, T. T. Weighted multiple classifier framework based on random projection. Inf. Sci. 490, 36–58 (2019).
Lin, J. et al. Convergences of regularized algorithms and stochastic gradient methods with random projections. J. Mach. Learn. Res. 21(20), 1–44 (2020).
Zhang, B. et al. Neural machine translation with deep attention. IEEE Trans. Pattern Anal. Mach. Intell. https://doi.org/10.1109/TPAMI.2018.2876404 (2020).
Li, X. et al. Vision-to-Language tasks based on attributes and attention mechanism. IEEE Trans. Cybern. 99, 1–14 (2016).
Li, C. et al. ASIF-Net: attention steered interweave fusion network for RGB-D salient object detection. IEEE Trans. Cybern. 99, 1–13 (2020).
Zhao, Z. et al. End-to-end keyword search system based on attention mechanism and energy scorer for low resource languages. Neural Netw. 139, 326–334 (2021).
Ahmad, M. et al. Fuzziness-based active learning framework to enhance hyperspectral image classification performance for discriminative and generative classifiers. Plos One 13(1), e0188996 (2017).
Ahmad, M. et al. Spatial-prior generalized fuzziness extreme learning machine autoencoder-based active learning for hyperspectral image classification. Optik 206, 163712 (2020).
Pasolli, E. et al. SVM active learning approach for image classification using spatial information. IEEE Trans. Geosci. Remote Sens. 52(4), 2217–2233 (2014).
Li, J. et al. Spectral-Spatial classification of hyperspectral data using loopy belief propagation and active learning. IEEE Trans. Geosci. Remote Sens. 51(2), 844–856 (2015).
Shi, Q. et al. Spatial coherence-based batch-mode active learning for remote sensing image classification. IEEE Trans. Image Process. 24(7), 2037–2050 (2015).
Li, J. et al. Hyperspectral image segmentation using a new Bayesian approach with active learning. IEEE Trans. Geosci. Remote Sens. 49(10), 3947–3960 (2011).
Lewis, D.D. et al. A sequential algorithm for training text classifiers: Corrigendum and additional data.Acm sigir forum. 29 (2), 13–19 (1995).
Di, W. et al. Active learning via multi-view and local proximity co-regularization for hyperspectral image classification. IEEE J. Sel. Top. Signal Process. 5(3), 618–628 (2011).
Hinke. et al. Query by committee. Proceedings of the fifth annual workshop on Computational learning theory. 287–294 (1991).
Rajan, S., Ghosh, J. & Crawford, M. M. An active learning approach to hyperspectral data classification. IEEE Trans. Geosci. Remote Sens. https://doi.org/10.1109/TGRS.2007.910220 (2008).
Ahmad, M. et al. Patial prior fuzziness pool-based interactive classification of hyperspectral images. Remote Sens. 11(9), 1136 (2019).
Pasolli, E. et al. Improving active learning methods using spatial information. Geoscience & Remote Sensing Symposium (2011).
Liu, A., Jun, G. & Ghosh, J. et al. Active learning of hyperspectral data with spatially dependent label acquisition costs. Geoscience & Remote Sensing Symposium (2009).
Tuia, et al. A survey of active learning algorithms for supervised remote sensing image classification. IEEE Journal of Selected Topics in Signal Processing 5(3), 606–617 (2011).
Haines, T.S.F. et al. Active learning using dirichlet processes for rare class discovery and classification. Ieice Technical Report Microwaves (2011).
Inglada, M.M. et al. Lazy yet efficient land-cover map generation for HR optical images. 2010 IEEE International Geoscience and Remote Sensing Symposium (2010).
Persello, D. et al. Batch-mode active-learning methods for the interactive classification of remote sensing images. IEEE Trans. Geosci. Remote Sens. 49(3), 1014–1031 (2011).
Munoz-Mari, J. et al. Semisupervised classification of remote sensing images with active queries. IEEE Trans. Geosci. Remote Sens. 50(10), 3751–3763 (2012).
He, L. et al. Recent advances on spectral-spatial hyperspectral image classification: An overview and new guidelines. IEEE Trans. Geosci. Remote Sens. 56(3), 1579–1597 (2018).
Datasets for 3D reconstruction. National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences [Online]. Available: http://vision.ia.ac.cn/data (2018).
Wei, G. Chinese Ancient Architecture Image Database. V1. Science Data Bank. https://doi.org/10.57760/sciencedb.ai.00008 (2024).
Funding
This work was supported by the Scientific Research Funding Project for Outstanding Doctors (Post-doctors) Working in Shanxi Province, China (Grant No. 20025LJ020), the Taiyuan Institute of Technology Scientific Research Initial Funding, China (Grant No.2024KJ032), the Key Research and Development Plan of Shanxi Province (Grant No. 202202150401005), and the Special Research Project of Science and Technology Strategy in Shanxi Province (Grant No. 202404030401104).
Author information
Authors and Affiliations
Contributions
All authors contributed to the current work. Conceptualization: C.Z., Z.S. Methodology: C.Z., S.X., Z.S. Formal analysis: C.Z., S.X., Z.Z., W.P. Example analysis: C.Z., S.X., W.P. Writing-original draft: C.Z. Writing-review: S.X., Z.S. Supervision: Z.S., Z.Z. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Cai, Z., Sun, X., Zhang, S. et al. Ancient architecture image classification with progressive stacking pseudoinverse learning. Sci Rep (2026). https://doi.org/10.1038/s41598-026-44876-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-44876-9
