Publication Title
Electronics
Document Type
Article
Abstract/Description
This paper develops a new multi-stage image distillation method that combines two well known techniques. In the first stage, our method creates a matrix from all training images. In the next stage, it adapts a modified principal component analysis (M-PCA) approach to transform the training matrix. In the third stage, Singular Value Decomposition (SVD) fur ther refines the training-image matrix through low-rank reconstruction and controlled row selection. In the fourth stage, rotation of small 2 ×2 matrix blocks on the entire left singular matrix is conducted. The upper m (user-selected number) rows of the reconstructed matrix are selected and transformed back to images, which we call distilled images. This dataset is significantly smaller yet retains the critical information needed for accurate classification. We validated the novelty and the advantages of the new method by applying the Base line, ResNet50V2, and ConvNetD4 CNNs and the public image databases Digit-MNIST, Fashion-MNIST, CIFAR-10, CIFAR-100, BloodMNIST, and Tiny ImageNet. Experimental results show that by distilling 50 images per class from CIFAR-10, CIFAR-100, and Tiny ImageNet, the proposed method achieves superior test accuracies of 75.59%, 56.27%, and 31.75%, respectively, when evaluated on ResNet and ConvNetD4.
Department
Mathematics
First Page
4219
DOI
https://doi.org/10.3390/electronics15184219
Volume
15
Date
Fall 9-16-2026
Citation Information
Shahnewaz, Tahsin; Chowdhury, Megdam Ahmed; and Sirakov, Nikolay Metodiev, "Non-Gradient Quaternion Training Matrix Modifications for Color-Image Distillation" (2026). Student Publications. 3.
https://lair.etamu.edu/cose-student-publications/3
