基于量子支持向量机模型的AI生成图像识别研究

Research of recognition of AI-generated images based on quantum support vector machine model

摘要

随着量子技术的发展,量子机器学习在图像识别领域中有着巨大的应用潜力.聚焦于量子支持向量机在识别人工智能(artificial intelligence,AI)生成图像这一具体问题上的应用,在有限的量子比特条件下,探索主成分分析特征降维和超参数调优对量子支持向量机算法评价指标的影响.结果表明,与经典支持向量机相比,应用量子支持向量机明显提升了准确率、精确率、召回率和F1分数;量子支持向量机算法特征降维效果显著,并且对超参数的选择表现出令人满意的鲁棒性.

Abstract

With the development of quantum technology,quantum machine learning holds immense potential for applications in the field of image recognition.This paper focuses on the application of quantum support vector machines for the specific task of identifying AI-generated images.For a limited number of qubits,we explored the impact of principal component analysis for feature dimensionality reduction and hyperparameter tuning on the performance metrics of the quantum support vector machine algorithm.The results indicate that,compared to classical support vector machines,the application of quantum support vector machines significantly improved accuracy,precision,recall,and F1 scores.Furthermore,the feature dimensionality reduction for the quantum support vector machine algorithm proved to be highly effective,and the quantum algorithm demonstrated satisfactory robustness to the selection of hyperparameters.

关键词

量子机器学习 / 量子支持向量机 / 图像识别

Key words

quantum machine learning / quantum support vector machine / image recognition

参考文献

[1] 金帆, 张鹏翼. 社交媒体中用户对人工智能生成图片的识别与认知研究:识别准确度、依据与态度探析[J]. 情报理论与实践, 2025, 48(6):12-19.DOI:10.16353/j.cnki.1000-7490.2025.06.002.

[2] Jin Fan, Zhang Pengyi. Research on users' identification and cognition of AI-generated images in social media:an exploration of identification accuracy,basis and attitude[J]. Information Studies:Theory & Application, 2025, 48(6):12-19.

[3] Biamonte J, Wittek P, Pancotti N, et al. Quantum machine learning[J]. Nature, 2017, 549(7671):195-202.DOI:10.1038/nature23474.

[4] Schuld M, Killoran N. Quantum machine learning in feature Hilbert spaces[J]. Physical Review Letters, 2019, 122(4):040504.DOI:10.1103/physrevlett.122.040504.

[5] 沈培鑫, 蒋文杰, 李炜康, 等. 量子人工智能中的对抗学习[J]. 物理学报, 2021, 70(14):19-31.DOI:10.7498/aps.70.20210789.

[6] Shen Peixin, Jiang Wenjie, Li Weikang, et al. Adversarial learning in quantum artificial intelligence[J]. Acta Physica Sinica, 2021, 70(14):19-31.

[7] Lloyd S, Mohseni M, Rebentrost P. Quantum algorithms for supervised and unsupervised machine learning[PP/OL]. V2.(2013-11-04)[2025-09-03].https://arxiv.org/abs/1307.0411.

[8] Li Z K, Liu X M, Xu N Y, et al. Experimental realization of a quantum support vector machine[J]. Physical Review Letters, 2015, 114(14):140504.DOI:10.1103/physrevlett.114.140504.

[9] Rana A, Vaidya P, Gupta G. A comparative study of quantum support vector machine algorithm for handwritten recognition with support vector machine algorithm[J]. Materials Today:Proceedings, 2022, 56:2025-2030.DOI:10.1016/j.matpr.2021.11.350.

[10] Wang Y L, Cao L Y. Quantum phase transition detection via quantum support vector machine[J]. Quantum Science and Technology, 2025, 10(1):015043.DOI:10.1088/2058-9565/ad985f.

[11] Miroszewski A, Mielczarek J, Czelusta G, et al. Detecting clouds in multispectral satellite images using quantum-kernel support vector machines[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16:7601-7613.DOI:10.1109/JSTARS.2023.3304122.

[12] Rodriguez-Grasa P, Farzan-Rodriguez R, Novelli G, et al. Satellite image classification with neural quantum kernels[J]. Machine Learning:Science and Technology, 2025, 6(1):015043.DOI:10.1088/2632-2153/ada86c.

[13] Nielsen M A, Chuang I L. Quantum computation and quantum information[M]. Cambridge: Cambridge University Press,2000:1-676.

[14] 纪豪森, 何英秋, 余明星, 等. 基于CLJAYA策略量子遗传算法的应急物资配送中心选址研究[J]. 中国科学:物理学力学天文学, 2025, 55(4):57-67.DOI:10.1360/SSPMA-2024-0350.

[15] Ji Haosen, He Yingqiu, Yu Mingxing, et al. Research on the location selection of emergency supply distribution centers using the CLJAYA strategy-based quantum genetic algorithm[J]. Scientia Sinica(Physica,Mechanica & Astronomica), 2025, 55(4):57-66.

基金资助

河北省高等教育教学改革研究与实践项目(2021GJJG482)

河北省中央引导地方科技发展资金(246Z0902G)

河北省高层次人才基金(C20221068)