Quantum image compression with autoencoders based on parameterized quantum circuits

The analysis and processing of digital images play a vital role in information processing. However, the pixel-based operations on images often lead to significant complexity as the image data grow rapidly. Encoding images into a quantum system and leveraging the principles of superposition and entan...

Full description

Saved in:
Bibliographic Details
Published in:Quantum information processing Vol. 23; no. 2
Main Authors: Wang, Hengyan, Tan, Jing, Huang, Yixiao, Zheng, Wenqiang
Format: Journal Article
Language:English
Published: New York Springer US 30.01.2024
Subjects:
Data Structures and Information Theory
Mathematical Physics
Physics
Physics and Astronomy
Quantum Computing
Quantum Information Technology
Quantum Physics
Spintronics
Quantum image compression
Parameterized quantum circuit
Quantum autoencoder
ISSN:1573-1332, 1573-1332
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The analysis and processing of digital images play a vital role in information processing. However, the pixel-based operations on images often lead to significant complexity as the image data grow rapidly. Encoding images into a quantum system and leveraging the principles of superposition and entanglement offer a chance to alleviate the challenges. A further improvement in efficiency is promising by combining quantum image processing with machine learning algorithms. Here a quantum autoencoder is trained to compress the image data into a lower-dimensional space using a hybrid quantum-classical control approach. The optimization of the parameterized quantum circuit involves the measurement of simple observables, alleviating the computational burden associated with the calculation of cost functions and gradients. We applied our quantum autoencoder to compress the MNIST handwritten digit dataset. The results exhibit the feasibility and effectiveness of the quantum compression approach. This work highlights the potential application of quantum neural networks in achieving high-efficiency quantum image processing.
ISSN:1573-1332
1573-1332
DOI:10.1007/s11128-023-04243-3