高光譜影像與一般彩色影像不同，其每個像素皆包含數百個連續光譜波段，構成獨特的「光譜指紋」，可用以辨識農作物、礦物、建築、土壤等不同地物類別。此技術已廣泛應用於農業監測、環境變遷分析與地質勘查等領域，能有效提升地表觀測的準確度。然而，高光譜影像通常具有極高維度、樣本稀少，且存在光譜與空間特徵整合等挑戰，使得分類任務具有挑戰性。現有方法主要依賴機器學習與深度學習架構進行特徵提取與分類。雖然這些架構在多數情境下表現穩定，但在面對高維複雜資料時，往往會遭遇表徵能力不足與過擬合問題。近年來，隨著量子計算的興起，量子神經網路被視為一種具潛力的全新架構。量子神經網路結合量子疊加與量子糾纏等特性，展現出處理高維度與複雜資料的潛力，逐漸受到關注。然而，目前應用量子神經網路於高光譜分類的研究仍屬初期，僅限於使用模擬器處理少量樣本，且大多未建立完整分類流程，亦缺乏分類圖層級的分析。有鑑於單一量子神經網路模型在實務中的侷限，部分研究轉向發展與經典神經網路（如卷積神經網路）結合的混合式架構，且現有混合模型多針對自然影像設計，缺乏對高光譜分類任務進行架構設計。為解決上述問題，本研究提出一種混合式模型，整合量子神經網路分支至卷積神經網路架構中，設計一套兼顧局部與全局關聯的高光譜影像分類架構。模型以卷積神經網路為主體，負責提取局部空間–光譜特徵；量子神經網路分支則由參數化量子電路構成，其量子性質有助於捕捉資料中非經典的特徵關聯，在處理高維度資料時具潛在輔助價值，進而提升分類效能。最終融合兩者所提取的特徵進行分類。本研究於三組公開資料集（Indian Pines、Pavia University 與 Salinas）上進行實驗，並與八種主流方法（SVM、RF、1D-CNN、2D-CNN、3DCNN、SSRN、Cubic-CNN、HybridSN）比較。實驗結果顯示，本研究模型在整體準確率（OA）、平均準確率（AA）與Kappa 係數等指標上皆優於對照方法，並於分類圖層級展現更平滑且穩定的預測表現。綜上所述，本研究驗證量子神經網路與卷積神經網路之融合架構於高光譜分類任務中的可行性與優越性，並為後續量子人工智慧技術於遙測影像領域的應用提供實證基礎。

Hyperspectral imaging differs from conventional RGB images in that each pixel contains hundreds of contiguous spectral bands, forming a unique “spectral fingerprint”that enables the identification of land cover types such as crops, minerals, buildings, and soil. This technology has been widely applied in fields including agricultural monitoring, environmental change analysis, and geological exploration, significantly enhancing the accuracy of Earth observation. However, HSI classification remains a challenging task due to the high dimensionality of the data, limited labeled samples, and the difficulty of integrating both spectral and spatial features. Existing methods primarily rely on machine learning and deep learning architectures, such as convolutional neural network and recurrent neural network, for feature extraction and classification. Although these models demonstrate stable performance in many scenarios, they often face representational limitations and overfitting issues when dealing with high-dimensional and complex data. In recent years, with the rise of quantum computing, the quantum neural network has emerged as a promising architecture. By leveraging quantum properties such as superposition and entanglement, quantum neural network shows great potential for modeling high-dimensional and complex data, and has gradually gained attention. Nevertheless, the application of quantum neural network to HSI classification is still in its early stage. Most studies rely on quantum simulators and use only small datasets, often lacking a complete classification pipeline or classification maplevel analysis. Due to the limitations of standalone quantum neural network, some researchers have proposed hybrid models that combine quantum neural network with classical neural networks like convolutional neural network. However, most existing hybrid models are designed for natural image tasks and are not tailored to the characteristics of hyperspectral data. To address these issues, this study proposes a hybrid model that integrates a quantum neural network branch into a CNN-based framework, designed to capture both local features and global relationships. The convolutional neural network serves as the main feature extractor for local spatial–spectral patterns, while the quantum neural network branch, built with parameterized quantum circuits, captures non-classical feature correlations. Features from both branches are fused for final classification. Experiments were conducted on three public datasets—Indian Pines, Pavia University, and Salinas—and compared with eight mainstream methods (SVM, RF, 1D-CNN, 2D-CNN, 3D-CNN, SSRN, Cubic-CNN, and HybridSN). Results show that our model achieves superior performance in terms of overall accuracy (OA), average accuracy (AA), and the Kappa coefficient, and provides smoother and more stable classification maps. In summary, this study verifies the feasibility and effectiveness of integrating quantum neural network and convolutional neural network for HSI classification and provides empirical evidence to support the future application of quantum artificial intelligence in the field of remote sensing.

[1] G. Hughes. On the mean accuracy of statistical pattern recognizers. IEEE Transactions on Information Theory, 1968.
[2] A. Plaza et al. Recent advances in techniques for hyperspectral image processing. Remote Sensing of Environment, 2009.
[3] J. A. Richards and X. Jia. Remote Sensing Digital Image Analysis. Springer, 2006.
[4] L. Bruzzone and D. Fernandez-Prieto. A technique for the selection of kernel function parameters in rbf neural networks for classification of remote-sensing images. IEEE Transactions on Geoscience and Remote Sensing, 2000.
[5] Y. Tarabalka et al. Classification of hyperspectral data using extended morphological profiles. IEEE Geoscience and Remote Sensing Letters, 2009.
[6] G. Camps-Valls and L. Bruzzone. Kernel-based methods for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing, 43(6):1351–1362, Jun 2005.
[7] L. Breiman. Random forests. Machine Learning, 45:5–32, 2001.
[8] P. Gislason, J. Benediktsson, and J. Sveinsson. Random forests for land cover classification. Pattern Recognition Letters, 27(4):294–300, 2006.
[9] Y. Chen, H. Jiang, C. Li, X. Jia, and P. Ghamisi. Deep feature extraction and classification of hyperspectral images based on convolutional neural networks. IEEE Transactions on Geoscience and Remote Sensing, 54(10):6232–6251, Oct 2016.
[10] W. Zhong, J. Li, and Z. Luo. Spectral–spatial residual network for hyperspectral image classification: A 3-d deep learning framework. IEEE Transactions on Geoscience and Remote Sensing, 57(2):810–823, Feb 2019.
[11] L. Sun and X. Song. Cubic convolutional neural network for hyperspectral image classification. In Artificial Intelligence and Security (ICAIS 2020), volume 1254 of CCIS, pages 234–245. Springer, 2020.
[12] A. Roy, S. Krishna, S. Dubey, and B. Chaudhuri. Hybridsn: Exploring 3d–2d cnn feature hierarchy for hyperspectral image classification. IEEE Geoscience and Remote Sensing Letters, 17(2):277–281, Feb 2020.
[13] Y. Chen, Z. Lin, X. Zhao, G. Wang, and Y. Gu. Deep learning-based classification of hyperspectral data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016.
[14] J. Li, J. Bioucas-Dias, and A. Plaza. Spectral–spatial classification of hyperspectral data using loopy belief propagation and active learning. IEEE Transactions on Geoscience and Remote Sensing, 51(2):844–856, Feb 2013.
[15] Y. Chen, X. Zhao, and X. Jia. Spectral–spatial classification of hyperspectral data based on deep belief network. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8(6):2381–2392, Jun 2015.
[16] M. A. Nielsen and I. L. Chuang. Quantum computation and quantum information. Cambridge University Press, 2010.
[17] A. Montanaro. Quantum algorithms: An overview. npj Quantum Information, 2(1):15023, 2016.
[18] J. Preskill. Quantum computing in the nisq era and beyond. Quantum, 2:79, 2018.
[19] M. Schuld, I. Sinayskiy, and F. Petruccione. An introduction to quantum machine learning. Contemporary Physics, 56(2):172–185, 2015.
[20] M. Schuld, I. Sinayskiy, and F. Petruccione. The quest for a quantum neural network. Quantum Information Processing, 13(11):2567–2586, 2014.
[21] F. Tacchino, C. Macchiavello, D. Gerace, and D. Bajoni. An artificial neuron implemented on an actual quantum processor. npj Quantum Information, 5(1):26, 2019.
[22] E. Farhi and H. Neven. Classification with quantum neural networks on near term processors, 2018. arXiv:1802.06002.
[23] A. Abbas, D. Sutter, C. Zoufal, A. Lucchi, A. Figalli, and S. Woerner. The power of quantum neural networks. Nature Computational Science, 1(6):403–409, 2021.
[24] J. R. McClean, S. Boixo, V. N. Smelyanskiy, R. Babbush, and H. Neven. Barren plateaus in quantum neural network training landscapes. Nature Communications, 9:4812, 2018.
[25] M. Benedetti, E. Lloyd, S. Sack, and M. Fiorentini. Parameterized quantum circuits as machine learning models. Quantum Science and Technology, 4(4):043001, 2019.
[26] M. Shi, H. Situ, and C. Zhang. Hybrid quantum neural network structures for image multi-classification, 2023. arXiv:2308.16005.
[27] M. F. Shahriyar and G. Tanbhir. Advancements and challenges in quantum machine learning for medical image classification: A comprehensive review, 2025. ResearchGate.
[28] Y. Li, H. Zhao, and X. Lu. Spectral–spatial classification of hyperspectral imagery with 3d convolutional neural network. Remote Sensing, 9(1):67, 2017.
[29] Z. He and Y. Li. Dimensionality reduction techniques for hyperspectral image analysis: A review. IEEE Geoscience and Remote Sensing Magazine, 4(3):63–86, 2016.
[30] A. Vaswani, N. Shazeer, N. Parmar, et al. Attention is all you need. In Advances in Neural Information Processing Systems, volume 30, 2017.