Dynamic Adaptive Random Projection with Stacking Pseudoinverse Learning for Image Classification
Open Access
- 13 November 2026
- journal article
- research article
- Published by MDPI AG in Mathematics
- Vol. 14 (22) , 4172
Abstract
Random projection is a vital dimensionality-reduction technique for high-dimensional data, whose performance heavily relies on the distribution design of its projection matrix. Conventional approaches generate projection matrices with predefined fixed distributions and thus fail to adapt to the complex distributions inherent in image data. This deficiency frequently causes feature loss or computational redundancy, which degrades overall model performance. To tackle this problem, this paper proposes a dynamic adaptive random projection with stacking pseudoinverse learning (DARPSP) for image classification. The core of the proposed method lies in building a fully dynamic adaptive random-projection mechanism. First, a data-feature perception module dynamically captures the underlying data distribution. Second, a parameterized dynamic distribution-control network predicts distribution parameters for the projection matrix to realize adaptive adjustment. A fusion strategy based on feature-importance evaluation is then adopted to refine the combination of multi-distribution projection outputs. Finally, the above components are integrated into a single-hidden-layer stacking pseudoinverse learning architecture to further boost model performance. The experimental results on the MNIST, COIL-20, and Salinas public datasets show that the proposed method outperforms the comparison methods on accuracy, precision, recall, and F1.Keywords
Funding Information
- Scientific Research Funding Project for Outstanding Doctors (Post-doctors) Working in Shanxi Province, China (20025LJ020)
- Taiyuan Institute of Technology Scientific Research Initial Funding, China (2024KJ032)
- Shanxi Provincial Key Research and Development Project (202202150401005)
- Science and Technology Innovation Projects of Colleges and Universities in Shanxi Province in 2023 (2023L352)
This publication has 39 references indexed in Scilit:
- Limited Sample Remote Sensing Image Classification Method Fused with the Optimized EfficientNetV2 and Capsule NetworkArabian Journal for Science and Engineering, 2026
- Advanced Breast Cancer Detection Using Attention-Guided MIL (Multi-Instance Learning) Optimized by Gray Wolf Optimization and Graph Convolution Aggregation ApproachArabian Journal for Science and Engineering, 2026
- AM-RP Stacking PILers: Random projection stacking pseudoinverse learning algorithm based on attention mechanismThe Visual Computer, 2023
- EnNeRFACE: improving the generalization of face reenactment with adaptive ensemble neural radiance fieldsThe Visual Computer, 2022
- An improved parameter learning methodology for RVFL based on pseudoinverse learnersNeural Computing and Applications, 2022
- DensePILAE: a feature reuse pseudoinverse learning algorithm for deep stacked autoencoderComplex & Intelligent Systems, 2021
- Tuning Database-Friendly Random Projection Matrices for Improved Distance Preservation on Specific DataApplied Intelligence, 2021
- A Robust Automated Machine Learning System with Pseudoinverse LearningCognitive Computation, 2021
- Spatial-prior generalized fuzziness extreme learning machine autoencoder-based active learning for hyperspectral image classificationOptik, 2020
- Segmented and non-segmented stacked denoising autoencoder for hyperspectral band reductionOptik, 2018