Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Mar 25, 2026

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
12:49

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

13.5K

Ancient architecture image classification with progressive stacking pseudoinverse learning.

Zhenjiao Cai1, Xuejian Sun2, Sulan Zhang3

  • 1Department of Computer Engineering, Taiyuan Institute of Technology, No. 31, Xinlan Road, Taiyuan, 030008, China. caizj224@163.com.

Scientific Reports
|March 24, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A Both-End Blocked Peroxidase-Mimicking DNAzyme for Low-Background Chemiluminescent Sensing of miRNA.

ACS sensors·2017
Same author

A Common Clinical Dilemma: Management of Persistent hrHPV Infection.

Trends in cancer·2017
Same author

Diastereoselective synthesis of novel heterocyclic scaffolds through tandem Petasis 3-component/intramolecular Diels-Alder and ROM-RCM reactions.

Chemical communications (Cambridge, England)·2017
Same author

Heterogeneity in Estimates of the Impact of Influenza on Population Mortality: A Systematic Review.

American journal of epidemiology·2017
Same author

Digestion of isolated legume cells in a stomach-duodenum model: three mechanisms limit starch and protein hydrolysis.

Food & function·2017
Same author

Comparison of efficacy of treatments for early syphilis: A systematic review and network meta-analysis of randomized controlled trials and observational studies.

PloS one·2017

This study introduces a new method for classifying ancient architecture images by improving feature focus and generalization. The progressive stacking pseudoinverse learning (AAPSP) model enhances accuracy in distinguishing architectural styles.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Computer Vision

Background:

  • Existing methods for ancient architecture image classification struggle with unconstrained weight initialization, hindering focus on key features like roof contours and bucket arches.
  • Current training approaches can lead to overfitting on regional styles, weakening cross-regional generalization capabilities.

Purpose of the Study:

  • To propose an improved ancient architecture image classification method using progressive stacking pseudoinverse learning (AAPSP).
  • To enhance the model's ability to focus on critical architectural features and improve cross-regional generalization.

Main Methods:

  • The proposed AAPSP method incorporates two modules: Key Features Stacking Pseudoinverse Learning (KFSP) and Progressive Optimization Learning (POL).
Keywords:
Ancient architecture image classificationKey featuresSample progressive optimizationStacking pseudoinverse learning

More Related Videos

Photorealistic Learned Landscapes for Augmented Reality
06:54

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

854

Related Experiment Videos

Last Updated: Mar 25, 2026

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
12:49

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

13.5K
Photorealistic Learned Landscapes for Augmented Reality
06:54

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

854
  • KFSP initializes weight matrices based on architectural feature distributions (e.g., Gaussian, uniform) and uses an attention mechanism (AM) to prioritize key component recognition.
  • POL utilizes a dynamic sample screening strategy to select rare features for iterative optimization, reducing redundancy and improving generalization.

    • Main Results:

    • Experiments on six Chinese ancient architecture datasets demonstrated the effectiveness of AAPSP.
    • The AAPSP model achieved superior performance in accuracy, precision, recall rate, and F1 score compared to existing methods.

    Conclusions:

    • The AAPSP method effectively addresses limitations in ancient architecture image classification by optimizing feature extraction and generalization.
    • The proposed approach shows significant potential for improving the classification of diverse architectural styles across different regions.