In this work we extend the recently proposed Quantum Vision (QV) theory in deep learning for image classification by integrating it with the Xception architecture, forming a novel Heavy QV-Xception model. The QV theory, inspired by the particle–wave duality of quantum physics, treats objects as information waves rather than static images, enabling deep neural networks to capture richer representations. Building on this concept, the Heavy QV-Xception model leverages a robust QV block to transform conventional images into wave-function representations and processes them through the depthwise separable convolutional layers of Xception for enhanced feature extraction. This hybrid approach benefits from both the quantum-inspired information representation and the efficient, high-performance architecture of Xception.Traditional convolutional neural networks process spatial images as rigid arrays of local pixel intensities, which restricts their ability to model long-range correlations and dynamic context, and leaves them vulnerable to degradation when the input is affected by noise, occlusion, illumination variation or compression artefacts. To overcome this limitation, the proposed framework implements a robust QV transformation block at the frontend of the network. This block treats images not as static surfaces but as dynamic information waves, mapping standard spatial inputs into wave-function representations through amplitude and phase encoding. This quantum-inspired conversion enables the model to encode complex, non-local statistical features and structural variations that are typically missed during standard localised pixel processing.Once the image is converted into a wave-function representation, the features are processed through the core pipeline of the Xception framework. The backbone relies on depthwise separable convolutions, which factorise standard convolutions by separating spatial channel-wise mapping from cross-channel pointwise correlations, reducing computational cost by approximately an order of magnitude and thereby creating the headroom within which the cost of the front-end transformation is absorbed.The model was trained and evaluated on a six-class natural scene dataset comprising buildings, forest, glacier, mountain, sea and street categories. It achieved a training accuracy of 92.44 per cent, a validation accuracy of 93.57 per cent and a test accuracy of 93.16 per cent, correctly classifying 2804 of 3010 held-out images, with a weighted F1-score of 93.11 per cent. Analysis of the confusion matrix shows that classification errors are concentrated among visually related category pairs rather than distributed arbitrarily, and that the forest class achieved perfect recall. The trained model was deployed within a Flask web application providing authentication, image upload and real-time classification, demonstrating that the proposed architecture is both accurate and practically deployable.
In this work we extend the recently proposed Quantum Vision (QV) theory in deep learning for image classification by integrating it with the Xception architecture, forming a novel Heavy QV-Xception model. The QV theory, inspired by the particle–wave duality of quantum physics, treats objects as information waves rather than static images, enabling deep neural networks to capture richer representations. Building on this concept, the Heavy QV-Xception model leverages a robust QV block to transform conventional images into wave-function representations and processes them through the depthwise separable convolutional layers of Xception for enhanced feature extraction. This hybrid approach benefits from both the quantum-inspired information representation and the efficient, high-performance architecture of Xception.Traditional convolutional neural networks process spatial images as rigid arrays of local pixel intensities, which restricts their ability to model long-range correlations and dynamic context, and leaves them vulnerable to degradation when the input is affected by noise, occlusion, illumination variation or compression artefacts. To overcome this limitation, the proposed framework implements a robust QV transformation block at the frontend of the network. This block treats images not as static surfaces but as dynamic information waves, mapping standard spatial inputs into wave-function representations through amplitude and phase encoding. This quantum-inspired conversion enables the model to encode complex, non-local statistical features and structural variations that are typically missed during standard localised pixel processing.Once the image is converted into a wave-function representation, the features are processed through the core pipeline of the Xception framework. The backbone relies on depthwise separable convolutions, which factorise standard convolutions by separating spatial channel-wise mapping from cross-channel pointwise correlations, reducing computational cost by approximately an order of magnitude and thereby creating the headroom within which the cost of the front-end transformation is absorbed.The model was trained and evaluated on a six-class natural scene dataset comprising buildings, forest, glacier, mountain, sea and street categories. It achieved a training accuracy of 92.44 per cent, a validation accuracy of 93.57 per cent and a test accuracy of 93.16 per cent, correctly classifying 2804 of 3010 held-out images, with a weighted F1-score of 93.11 per cent. Analysis of the confusion matrix shows that classification errors are concentrated among visually related category pairs rather than distributed arbitrarily, and that the forest class achieved perfect recall. The trained model was deployed within a Flask web application providing authentication, image upload and real-time classification, demonstrating that the proposed architecture is both accurate and practically deployable.
In this work, we extend the recently proposed Quantum Vision (QV) theory in deep learning for object recognition by integrating it with the Xception architecture, forming a novel Heavy QV-Xception model.The QV theory, inspired by the particle-wave duality in quantum physics, treats objects as information waves rather than static images, enabling deep neural networks to capture richer representations.Building on this concept, our Heavy QV-Xception model leverages a robust QV block to transform conventional images into wave-function representations and processes them through the depthwise separable convolutional layers of Xception for enhanced feature extraction.This hybrid approach benefits from both the quantum-inspired information representation and the efficient, high-performance architecture of Xception.Extensive experiments on multiple benchmark datasets demonstrate that the Heavy QV-Xception model consistently outperforms standard Xception and other conventional CNNs, highlighting the effectiveness of combining QV theory with advanced deep learning architectures for improved object recognition accuracy. INTRODUCTIONObject recognition has become one of the most critical applications in the field of computer vision, with wide-ranging use cases in autonomous systems, medical imaging, surveillance, and human-computer interaction.Traditional deep learning models, particularly Convolutional Neural Networks (CNNs), have achieved remarkable progress in recognizing visual patterns.However, these models often rely solely on spatial feature extraction and fail to capture deeper contextual and representational information, which limits their effectiveness in handling complex and high-dimensional datasets.As object recognition tasks continue to grow in complexity, there is an increasing demand for models that combine efficiency with richer feature representations.Quantum-inspired approaches such as Quantum Vision (QV) have emerged to address this gap by introducing the concept of particle-wave duality from quantum physics into image recognition tasks.By treating objects as wave functions rather than fixed pixel intensities, QV enables models to capture more abstract and holistic representations of images.Although the integration of QV with CNNs has shown promise, these systems still face challenges, such as high computational costs, limited scalability, and inefficiency in extracting deeper hierarchical features.Hence, there is a need for architectures that can combine the representational strength of QV with modern, efficient deep learning frameworks.To overcome these challenges, this project introduces the Heavy QV-Xception model, which integrates a robust Quantum Vision block with the Xception architecture.The Xception model, based on depthwise separable convolutions, provides computational efficiency and high accuracy, while the Heavy QV block enriches image representations by transforming them into wave functions.This hybrid system leverages the advantages of both approaches, enabling scalable, efficient, and highly accurate object recognition.With extensive evaluation on benchmark datasets, the proposed model demonstrates superior performance compared to conventional CNNs and QV-CNN hybrids, marking a significant advancement in the field of deep learning-based object recognition. STIMULATINGThe scope of this project is to design and implement a Heavy QV-Xception model that integrates Quantum Vision (QV) theory with the Xception deep learning framework for advanced object recognition tasks.Unlike traditional CNN-based systems, which focus primarily on spatial pixel features, this model captures wave-function representations of objects, enabling a deeper and more holistic understanding of image data.By incorporating QV blocks, the model introduces quantum-inspired representational learning, which enhances the robustness of feature extraction across varying datasets and environmental conditions.This project is applicable to a wide range of real-world domains, such as autonomous driving, medical
As image datasets grow in scale and complexity, classical convolutional neural networks (CNNs) increasingly face limitations in computational efficiency, scalability, and generalisation in low-data regimes. Quantum machine learning (QML) offers a promising alternative, with quantum convolutional neural networks (QCNNs) exploiting quantum parallelism and entanglement for feature extraction. This paper introduces the Novel Quantum Convolutional Neural Network (No-QCNN), which differs from recent hybrid QCNN/QCQ-CNN architectures by employing an end-to-end quantum convolution-pooling pipeline without inserting an intermediate classical CNN block. No-QCNN is a hybrid quantum-classical model that employs a variational quantum classifier (VQC) optimised via the COBYLA algorithm and is designed for binary and multiclass classification of low-resolution colour images on near-term quantum devices. A key innovation is a problem-specific quantum feature map that pre-processes each image into a structured three-dimensional block-matrix representation, jointly encoding pixel colour (R, G, B) and spatial position before mapping this information into a hierarchical ZZFeatureMap. This encoding captures spatial-chromatic correlations at shallow circuit depth, making it compatible with noisy intermediate-scale quantum (NISQ) constraints. The model is implemented using IBM-Qiskit on a local quantum simulator and benchmarked against a classical CNN. For a six-class classification task on a 50-image dataset, No-QCNN achieves a validation accuracy of 82.05%, substantially outperforming the classical CNN's 40.00%, indicating improved generalisation and reduced overfitting in low-data, multiclass settings. Conversely, for a simpler binary classification task, the classical CNN achieves perfect validation accuracy of 100%, surpassing the No-QCNN's 89.7%. We further observe that No-QCNN performance declines with increasing dataset size and training time due to limited circuit expressibility, data-encoding overhead, and finite-sampling noise inherent to NISQ models. Overall, these results position No-QCNN as a complementary framework best suited to low-data, correlation-rich classification tasks, defining a realistic niche for quantum-enhanced artificial vision and quantum perception in the NISQ era.
The objective of this research is to develop a Hybrid Quantum-Classical Convolutional Neural Network (QC-CNN) framework that integrates parameterized quantum circuits (PQCs) within classical CNN architectures to enhance image classification performance.The proposed model leverages quantum principles such as superposition and entanglement for high-dimensional feature representation, achieving superior accuracy and reduced computational complexity.Experiments on MNIST and CIFAR-10 datasets demonstrated improved classification accuracy of 98.7% and 82.5%, respectively, surpassing traditional CNNs while reducing training time by 33% and parameters by 45%.Statistical analysis confirmed the significance of these improvements.Visualizations using t-SNE revealed enhanced class separability, and noise perturbation tests validated the model's robustness.The results highlight the hybrid QC-CNN's potential for efficient and scalable quantum-enhanced deep learning applications.Extended evaluations, including quantum-layer-depth analysis, robustness testing, and t-SNE visualization, empirically support the hybrid QC-CNN's effectiveness across varied scenarios.
Convolutional Neural Networks (CNNs) have exposed robust results for classifying images in data analytics. On the other hand, the effectiveness of these models is often limited by their substantial computational complexities, particularly when dealing with features represented in complex mathematical environments, multi class image classification, and especially when dealing with complex nonlinear relationships between features, which need to be addressed for transparent decision-making processes. This study developed a Quantum-Enhanced Convolutional Neural Network (Q-CNN) model, which incorporates traditional convolution-based feature extraction, Quantum Feature Encoding, and quantum processing. The developed model applies a quantum encoding mechanism to map classical features to quantum states for alternative feature representations and classification purposes. Experimental results show that the Q-CNN outperformed the standard CNN, reaching 97% training and validation accuracy compared to 94%, and lowering training and validation losses from 17% and 18% to 9% and 11%, respectively. Furthermore, precision, recall, and F1-score were increased from 90%, 88%, and 89% to 92%, 94%, and 93%, respectively. These findings demonstrate the potential effectiveness of quantum feature encoding for image classification tasks and provide a structured framework for enriched feature representation in convolution-based image processing models.
The integration of quantum computing principles with classical deep learning architectures has emerged as a promising direction for advancing image recognition systems beyond the limitations of purely classical convolutional neural networks (CNNs). This study proposes and analyzes a hybrid quantum–classical deep learning framework based on quanvolutional neural network (QNN) principles, where quantum circuits are embedded within convolutional feature extraction layers to enhance representation capability. The research investigates how quantum-enhanced filters, particularly quanvolutional layers, contribute to improved feature diversity, non-linear transformation capacity, and classification accuracy in image recognition tasks. Building upon established convolutional architectures and quantum machine learning theories, the proposed framework integrates quantum random circuit encoding with classical deep neural networks to evaluate performance gains in comparison to conventional CNN models. The study further incorporates theoretical insights from hybrid quantum-classical learning systems, emphasizing the role of parameterized quantum circuits in feature mapping and optimization stability. The methodology is grounded in NISQ-era quantum computing constraints and leverages hybrid optimization strategies to ensure computational feasibility. Experimental evaluation demonstrates that quanvolution-based models can provide improved robustness in feature extraction, particularly in high-dimensional image spaces, while maintaining competitive computational efficiency. However, scalability limitations and noise sensitivity in quantum circuits remain key challenges. Overall, this work contributes to the growing field of quantum-enhanced machine learning by providing a structured performance analysis of hybrid quanvolutional architectures and highlighting their potential role in next-generation intelligent visual recognition systems.
We present a hybrid deepfake detection model that combines classical and quantum techniques for robust feature extraction. Our pipeline integrates an Xception-based CNN with a ViT to capture spatial features. A custom Quantum Convolutional Layer, implemented using PennyLane, is introduced to leverage quantum properties for enhanced feature representation. To address class imbalance and improve generalization, we employ focal loss, mixup augmentation, and test-time augmentation. Evaluated on the Deepfake and real images dataset, our model achieves a ROC-AUC of 0.995, with 0.984 precision and 0.982 F1-score. These results demonstrate the potential of hybrid quantum-classical architectures in advancing deepfake detection and open pathways for applying quantum methods to broader computer vision challenges.