Evaluating the Effectiveness of Hybrid Quantum-Classical Convolutional Neural Networks for Image Classification in Multiple Color Spaces

• URL: http://arxiv.org/abs/2406.02229v2
• Date: Tue, 19 Nov 2024 03:53:43 GMT
• Title: Evaluating the Effectiveness of Hybrid Quantum-Classical Convolutional Neural Networks for Image Classification in Multiple Color Spaces
• Authors: Kwok-Ho Ng, Tingting Song,
• Abstract summary: We propose a hybrid quantum-classical convolutional neural network (HQCCNN) model to analyze performance in four different color space images. For some superclasses, the model performs in the Lab, YCrCb, and HSV color spaces as well as or better than RGB.
• Score: 1.565361244756411
• License:
• Abstract: As the complexity and scale of image processing tasks continue to expand, quantum convolutional neural networks (QCNNs) have demonstrated the capability to improve image processing performance. These networks can accelerate processing speed and enhance classification accuracy while reducing the number of model parameters. However, these studies restrict images to the RGB color space, and the effectiveness of QCNNs in other color spaces still needs to be explored. In this work, we propose a hybrid quantum-classical convolutional neural network (HQCCNN) model to analyze performance in four different color space images: RGB, Lab, YCrCb, and HSV. Using several multi-qubit entangled gates, HQCCNN is constructed as a parameterized quantum circuit model to evaluate classification performance across these color spaces. Furthermore, we verify the effectiveness of HQCCNN on the CIFAR-100 dataset. The experimental results show that it achieves significantly greater accuracy in binary classification tasks within the RGB color space. For some superclasses, the model performs in the Lab, YCrCb, and HSV color spaces as well as or better than RGB. This serves as an essential reference for QCNN in processing various color space image data for computer vision applications.

Related papers

• A Hybrid Spiking-Convolutional Neural Network Approach for Advancing Machine Learning Models [6.528272856589831]
  We propose a novel hybrid Spiking-Convolutional Neural Network (SC-NN) model and test on using image inpainting tasks. Our approach uses the unique capabilities of SNNs, such as event-based computation and temporal processing, along with the strong representation learning abilities of CNNs. The model is trained on a custom dataset specifically designed for image inpainting, where missing regions are created using masks.
  arXiv  Detail & Related papers  (2024-07-11T20:50:33Z)
• Color Equivariant Convolutional Networks [50.655443383582124]
  CNNs struggle if there is data imbalance between color variations introduced by accidental recording conditions. We propose Color Equivariant Convolutions ( CEConvs), a novel deep learning building block that enables shape feature sharing across the color spectrum. We demonstrate the benefits of CEConvs in terms of downstream performance to various tasks and improved robustness to color changes, including train-test distribution shifts.
  arXiv  Detail & Related papers  (2023-10-30T09:18:49Z)
• Hybrid Quantum Neural Network Structures for Image Multi-classification [0.0]
  Two recent image classification methods have emerged, one employs PCA dimensionality reduction and angle encoding, the other integrates QNNs into CNNs to boost performance. This study explores these algorithms' performance in multi-class image classification and proposes an optimized hybrid quantum neural network suitable for the current environment.
  arXiv  Detail & Related papers  (2023-08-30T12:48:05Z)
• Name Your Colour For the Task: Artificially Discover Colour Naming via Colour Quantisation Transformer [62.75343115345667]
  We propose a novel colour quantisation transformer, CQFormer, that quantises colour space while maintaining machine recognition on the quantised images. We observe the consistent evolution pattern between our artificial colour system and basic colour terms across human languages. Our colour quantisation method also offers an efficient quantisation method that effectively compresses the image storage.
  arXiv  Detail & Related papers  (2022-12-07T03:39:18Z)
• Learning to Structure an Image with Few Colors and Beyond [59.34619548026885]
  We propose a color quantization network, ColorCNN, which learns to structure an image in limited color spaces by minimizing the classification loss. We introduce ColorCNN+, which supports multiple color space size configurations, and addresses the previous issues of poor recognition accuracy and undesirable visual fidelity under large color spaces. For potential applications, we show that ColorCNNs can be used as image compression methods for network recognition.
  arXiv  Detail & Related papers  (2022-08-17T17:59:15Z)
• Universal Deep Network for Steganalysis of Color Image based on Channel Representation [41.86330101334733]
  We design a universal color image steganalysis network (called UCNet) in spatial and JPEG domains. The proposed method includes preprocessing, convolutional, and classification modules. We conduct extensive experiments on ALASKA II to demonstrate that the proposed method can achieve state-of-the-art results.
  arXiv  Detail & Related papers  (2021-11-24T02:22:13Z)
• RGB Image Classification with Quantum Convolutional Ansaetze [18.379304679643436]
  We propose two types of quantum circuit ansaetze to simulate convolution operations on RGB images. To the best of our knowledge, this is the first work of a quantum convolutional circuit to deal with RGB images effectively. We also investigate the relationship between the size of quantum circuit ansatz and the learnability of the hybrid quantum-classical convolutional neural network.
  arXiv  Detail & Related papers  (2021-07-23T09:38:59Z)
• CNNs for JPEGs: A Study in Computational Cost [49.97673761305336]
  Convolutional neural networks (CNNs) have achieved astonishing advances over the past decade. CNNs are capable of learning robust representations of the data directly from the RGB pixels. Deep learning methods capable of learning directly from the compressed domain have been gaining attention in recent years.
  arXiv  Detail & Related papers  (2020-12-26T15:00:10Z)
• When CNNs Meet Random RNNs: Towards Multi-Level Analysis for RGB-D Object and Scene Recognition [10.796613905980609]
  We propose a novel framework that extracts discriminative feature representations from multi-modal RGB-D images for object and scene recognition tasks. To cope with the high dimensionality of CNN activations, a random weighted pooling scheme has been proposed. Experiments verify that fully randomized structure in RNN stage encodes CNN activations to discriminative solid features successfully.
  arXiv  Detail & Related papers  (2020-04-26T10:58:27Z)
• Learning to Structure an Image with Few Colors [59.34619548026885]
  We propose a color quantization network, ColorCNN, which learns to structure the images from the classification loss in an end-to-end manner. With only a 1-bit color space (i.e., two colors), the proposed network achieves 82.1% top-1 accuracy on the CIFAR10 dataset. For applications, when encoded with PNG, the proposed color quantization shows superiority over other image compression methods in the extremely low bit-rate regime.
  arXiv  Detail & Related papers  (2020-03-17T17:56:15Z)
• Widening and Squeezing: Towards Accurate and Efficient QNNs [125.172220129257]
  Quantization neural networks (QNNs) are very attractive to the industry because their extremely cheap calculation and storage overhead, but their performance is still worse than that of networks with full-precision parameters. Most of existing methods aim to enhance performance of QNNs especially binary neural networks by exploiting more effective training techniques. We address this problem by projecting features in original full-precision networks to high-dimensional quantization features.
  arXiv  Detail & Related papers  (2020-02-03T04:11:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.