Publisher

IAESCOREIAESCORE

Journal International Journal of Electrical and Computer Engineering (IJECE)International Journal of Electrical and Computer Engineering (IJECE) Abstract

The classification of geometric patterns, particularly in Islamic art, presents a compelling challenge for the field of computer vision due to its intricate symmetry and scale invariance. This study proposes an ensemble learning framework to classify geometric patterns, leveraging the novel quaternion cartesian fractional Hahn moments (QCFrHMs) as a robust feature extraction method. QCFrHMs integrate the fractional Hahn polynomial and quaternion algebra to provide compact, invariant descriptors for geometric patterns. Combined with Zernike Moments, this dual-feature approach ensures resilience against rotation, scaling, and noise variations. The extracted features were evaluated using support vector machines (SVM), random forest, and a soft-voting ensemble classifier. Experiments were conducted on a dataset comprising 1,204 geometric images categorized into two symmetry groups (p4m and p6m). Results demonstrated that the standalone models, achieving a classification accuracy of 82.15%. The ensemble classifier outperformed the integration of QCFrHMs significantly enhanced the systems robustness compared to traditional Zernike-only approaches, which aligns with findings in prior studies. This research contributes to the fields of image processing and pattern recognition by introducing an efficient feature extraction technique combined with ensemble learning for precise and scalable geometric pattern classification. The implications extend to art preservation, architectural analysis, and automated indexing of cultural heritage imagery.

Conclusion

This research successfully demonstrated the effectiveness of combining ensemble learning with advanced moment-based descriptors, specifically quaternion cartesian fractional Hahn moments (QCFrHMs) and Zernike moments, for classifying geometric patterns with p4m and p6m symmetries.The integration of both descriptor sets captured intricate geometric traits, while dimensionality reduction via PCA maintained a balance between complexity and accuracy.The voting classifier, composed of a random forest and an SVM, outperformed standalone models, highlighting the advantage of merging complementary algorithms.These findings open new pathways in shape analysis and promise wide-ranging applications in computer vision, from industrial quality control to the digital exploration of elaborate ornamental art.

Future Research

Future research should focus on addressing the computational expense associated with higher-order moment calculations, particularly for large datasets, and enhancing noise resilience through improved preprocessing techniques. Expanding the framework to encompass additional symmetry types and exploring real-time implementation via GPU acceleration and approximate sampling methods are also promising avenues. Furthermore, investigating the integration of deep learning frameworks with QCFrHMs for end-to-end classification pipelines, alongside exploring their applicability in 3D object analysis and generative modeling for geometric pattern synthesis, could yield even more robust and efficient classification methods. These advancements will not only refine the current state-of-the-art in geometric pattern classification but also broaden its applicability to diverse fields such as cultural heritage preservation, automated design, and industrial quality control, ultimately fostering a deeper understanding and appreciation of intricate visual patterns.

References

1. Exploring ensemble learning for classifying geometric patterns: insights from quaternion cartesian fractional... doi.org/10.11591/ijece.v15i5.pp4630-4641Exploring ensemble learning for classifying geometric patterns insights from quaternion cartesian fractional doi 10 11591 ijece v15i5 pp4630 4641
2. Image analysis via the general theory of moments*. analysis general theory moments josa history issues... doi.org/10.1364/josa.70.000920Image analysis via the general theory of moments analysis general theory moments josa history issues doi 10 1364 josa 70 000920
3. Mobile and Forensics. tiles neural network mobile forensics authors susanti aulia ramadayanti universitas... doi.org/10.12928/mf.v3i2.5643Mobile and Forensics tiles neural network mobile forensics authors susanti aulia ramadayanti universitas doi 10 12928 mf v3i2 5643

Metadata

File size	665.33 KB
Pages	12
DMCA	Report