We propose a data-adaptive probabilistic intensity remapping framework for structure-preserving transformation of grayscale images. The suggested method formulates intensity transformation as a continuous, data-driven remapping process, in contrast to traditional histogram-based techniques that rely on hard thresholding and generate piecewise-constant mappings. The image statistics yield representative intensity values, and Gaussian-based weighting methods probabilistically allocate each pixel to several components. Smooth transitions while preserving structural features are achieved by computing the output intensity as an expectation over these components. A smooth transition from soft probabilistic remapping to hard assignment is made possible by the introduction of a nonlinear sharpening parameter $γ$ to regulate the degree of localization. This offers clear control over the trade-off between intensity discrimination and smoothing. Furthermore, the resolution of the remapping function is determined by the number of components $k$. When compared to thresholding-based methods, experimental results on standard benchmark images show that the suggested method achieves better structural fidelity and controlled information reduction as measured by PSNR, SSIM, and entropy. Overall, by allowing continuous, probabilistic intensity modifications, the framework provides a robust and efficient substitute for discrete thresholding.
We propose a data-adaptive probabilistic intensity remapping framework for structure-preserving transformation of grayscale images. The suggested method formulates intensity transformation as a continuous, data-driven remapping process, in contrast to traditional histogram-based techniques that rely on hard thresholding and generate piecewise-constant mappings. The image statistics yield representative intensity values, and Gaussian-based weighting methods probabilistically allocate each pixel to several components. Smooth transitions while preserving structural features are achieved by computing the output intensity as an expectation over these components. A smooth transition from soft probabilistic remapping to hard assignment is made possible by the introduction of a nonlinear sharpening parameter γ to regulate the degree of localization. This offers clear control over the trade-off between intensity discrimination and smoothing. Furthermore, the resolution of the remapping function is determined by the number of components k. When compared to thresholding-based methods, experimental results on standard benchmark images show that the suggested method achieves better structural fidelity and controlled information reduction as measured by PSNR, SSIM, and entropy. Overall, by allowing continuous, probabilistic intensity modifications, the framework provides a robust and efficient substitute for discrete thresholding.
Quantum image representation (QIR) is the basic idea behind quantum image processing. It explains how a normal image is converted into quantum states so that it can be processed using quantum computers. The commonly used models for QIR are Flexible Representation of Quantum Images (FRQIs) and Novel Enhanced Quantum Representation (NEQR). Though these approaches highlight the potential of quantum-based image encoding, the limitation of practical applicability on Noisy Intermediate-Scale Quantum (NISQ) devices exists. In this paper, we propose an intensity-preserving quantum image representation (IP-QIR) scheme that aims to maintain accurate grayscale intensity information while significantly reducing quantum resource usage. The proposed method employs a controlled rotation-based encoding strategy, where pixel intensities are embedded into the measurement probability of a single intensity qubit, and spatial information is represented using position qubits. To further enhance feasibility on near-term quantum hardware, the framework operates on small image patches instead of full-resolution images, thereby reducing circuit depth and overall complexity. The performance of the proposed IP-QIR approach is evaluated through IBM Qiskit simulations on three types of grayscale images: synthetic image patches, synthetic aperture radar (SAR) images, and medical tuberculosis (TB) chest X-ray images. Experimental results demonstrate that IP-QIR achieves better intensity preservation than FRQIs and NEQR, with fidelity values reaching up to 84.12% for both SAR and medical datasets. In addition, IP-QIR represents a 4×4 image patch using only five qubits, which significantly reduces the qubit requirement when compared to NEQR, while still preserving high reconstruction accuracy.
Abstract Color images (i.e. RGB) contain rich visual information essential for object recognition, spatial context analysis, and decision-making across fields such as medical imaging, remote sensing, and computer vision. Processing such data efficiently requires significant computational resources, which can be addressed by the parallelism and speed of quantum computing. Representing RGB image data within quantum states enables efficient and scalable processing compared to classical methods. In recent years, various quantum image representation techniques have been developed based on probabilistic and basis state encoding. Probabilistic encoding methods are efficient for representing quantum binary images but face challenges in accurately reconstructing RGB images. This is due to the randomness introduced by measurement. To overcome this limitation, several basis state encoding approaches have been proposed. These methods map classical pixel values directly to quantum basis states. Thus enabling accurate grayscale and RGB image representation. However, they often require a larger number of qubits and complex gate operations, leading to higher circuit preparation costs. So, there exists substantial potential for quantum probabilistic RGB image representation using randomized measurements. Hence, our objective is to develop a probabilistic quantum image representation, capable of representing RGB images using fewer qubits. This method aims to reconstruct the classical image with high accuracy using randomized measurements, thereby addressing the major limitations of probabilistic methods. To balance image quality with resource efficiency, we propose an Improved Probabilistic Quantum Image Representation (IPQIR) method based on equal superposition states. This approach utilizes a quantum circuit with Hadamard Gate to encode classical pixel values into equal superposition states. The original image is then reconstructed by performing multiple quantum measurements and applying expectation value estimation to average the outcomes. To align with existing research in quantum image representation, quantum state tomography and different applications are utilized as standard evaluation methods to assess the effectiveness of the proposed algorithm. The tomography achieved a fidelity of 97.51%. In addition, various parametric and non-parametric statistical tests were performed to verify the accuracy of the reconstructed classical image. The high Pearson (0.9983) and Spearman (0.9959) correlation coefficients indicate a nearly perfect linear and monotonic relationship between the original and reconstructed images.
Aiming to implement image segmentation precisely and efficiently, we exploit new ways to encode images and achieve the optimal thresholding on quantum state space. Firstly, the state vector and density matrix are adopted for the representation of pixel intensities and their probability distribution, respectively. Then, the method based on global quantum entropy maximization (GQEM) is proposed, which has an equivalent object function to Otsu's, but gives a more explicit physical interpretation of image thresholding in the language of quantum mechanics. To reduce the time consumption for searching for optimal thresholds, the method of quantum lossy-encoding-based entropy maximization (QLEEM) is presented, in which the eigenvalues of density matrices can give direct clues for thresholding, and then, the process of optimal searching can be avoided. Meanwhile, the QLEEM algorithm achieves two additional effects: (1) the upper bound of the thresholding level can be implicitly determined according to the eigenvalues; and (2) the proposed approaches ensure that the local information in images is retained as much as possible, and simultaneously, the inter-class separability is maximized in the segmented images. Both of them contribute to the structural characteristics of images, which the human visual system is highly adapted to extract. Experimental results show that the proposed methods are able to achieve a competitive quality of thresholding and the fastest computation speed compared with the state-of-the-art methods.
Medical image encoding has several challenges. File sizes are large. Processing is slow. Fine details like edges, textures, and soft tissue patterns can be lost. Traditional methods use matrix operations that are powerful but often hard to scale and interpret. These limitations highlight opportunities to improve image encoding with better interpretability. To explore new opportunities for scalable image encoding, we introduce a quantum inspired model built on a dictionary structure. In this model, quantum states are stored as key value pairs for ensuring interpretability and scalability. We do not directly implement quantum gates. we provide dictionary based alternatives that emulate gate behavior using key value logic. We simulate the effect of quantum gates on these structures. For example, a Hadamard like rule may help capture sharp or high contrast areas. Phase like rules (such asSandT) may respond to smooth or soft changes. Control rules, likeCXandCCX, can connect different parts of the image using logic. It may reduce the number of steps needed for image operations. This makes it useful for small quantum image datasets. It also makes the steps more explainable. The process is also traceable through the dictionary logic. We tested this method on grayscale MRI image patches. It performs well in tasks such as image compression. It also supports pattern recognition and secure image sharing. These results suggest its suitability for medical image applications. The model balances technical performance and explainability. This makes it helpful for developers building quantum tools. It also supports researchers and students in medical image analysis.
This paper presents a hybrid quantum-classical framework for grayscale image compression and decompression, leveraging the strengths of quantum computing and deep learning. The compression pipeline integrates a Variational Quantum Daubechies Wavelet Transform (V-QDWT) and a trainable Quantum Convolutional Neural Network (QCNN) optimized end-to-end to achieve efficient, image-adaptive multi-resolution analysis and entanglement-based feature reduction. Input images are encoded using the Normal Arbitrary Superposition State (NASS) representation, enabling compact and scalable quantum storage. For decompression, we implement inverse QCNN and V-QDWT circuits to reconstruct coarse image features natively, followed by a classical Super-Resolution Generative Adversarial Network (SRGAN) to enhance perceptual quality. Experimental evaluations on benchmark grayscale datasets demonstrate the efficacy of our hybrid approach. By jointly training the quantum layers, the base quantum pipeline closely rivals classical JPEG2000 standards. Subsequent SRGAN refinement substantially pushes the boundaries of the reconstruction, achieving superior structural fidelity (PSNR: 30.0667 dB, SSIM: 0.8744, Histogram Correlation: 0.9244). Histogram analysis and qualitative comparisons further validate the restoration of fine textures and intensity distributions. Our findings highlight the potential of combining variational quantum compression with classical deep learning to enable efficient, scalable, and perceptually-aware image processing in quantum-enhanced computing environments.
The fusion of quantum computing with image processing has significantly improved both computational efficiency and fidelity in amplitude-encoded image reconstruction. While prior investigations have delved into advancements in quantum-to-classical data decoding, there remains a dearth of research addressing hybrid quantum-classical architectures and strategies to enhance noise resilience in quantum image processing. This study evaluates the implementation of three quantum transforms: Quantum Fourier Transform (QFT), Quantum Wavelet Transform (QWT), and Quantum Principal Component Analysis (QPCA), focusing on their efficacy in reconstructing images subjected to Gaussian and salt-and-pepper noise. Reconstruction quality is quantified through metrics such as error rates, similarity indices, and uncertainty quantification metrics, supported by a sensitivity analysis to ascertain robustness under varying noise conditions. The results demonstrate that QWT consistently achieves the highest Structural Similarity Index Measure (SSIM), while QFT often yields negative histogram similarity values, indicating significant alterations in color distribution. For instance, when analyzing one image, QWT exhibited a moderate decrease in Peak Signal-to-Noise Ratio (PSNR) from 37.83 to 37.71 with noise introduced, whereas QFT's PSNR experienced a more pronounced decline from 38.70 to 38.48. The study elucidates an inverse relationship between Mean Squared Error (MSE) and PSNR, underscoring the adverse effects of noise on image quality. Notably, QPCA was observed to produce the lowest Euclidean distance, signifying a strong retention of image integrity. These findings underscore the importance of judiciously selecting quantum methodologies tailored to specific application requirements, particularly in critical domains such as medical imaging, where precision and accuracy are paramount.