Multiscale contiguous block pixel entangler for image recognition on hybrid quantum-classical computing system

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published April 7th, 2026

Detected April 7th, 2026

Summary

The USPTO granted patent US12596951B2 to IONQ, Inc. covering a quantum convolutional neural network (QCNN) implementation method for image recognition using a hybrid quantum-classical computing system. The patent describes encoding pixel data onto a quantum processor using column and row qubits, with a contiguous block pixel entangler for feature detection. The 20-claim patent protects innovations in quantum-classical hybrid image processing.

View original document View source feed page

What changed

The USPTO issued patent grant US12596951B2 to IONQ, Inc. for a method of implementing quantum convolutional neural networks in hybrid quantum-classical computing systems. The patent covers encoding input image pixel data onto a quantum processor, applying convolutional layer operations with a contiguous block pixel entangler to detect features, performing pooling operations via one-qubit operations, and measuring output qubit states. The invention spans multiple CPC classifications including G06N (quantum computing/neural networks) and G06V (image recognition).

For technology companies and researchers developing quantum machine learning applications, this patent establishes intellectual property protection for QCNN-based image recognition methods. Competitors in quantum computing, AI hardware, and image recognition technology should review this patent when designing quantum-classical hybrid systems to avoid potential infringement. The patent's broad coverage of contiguous block entanglement patterns in quantum image processing may affect research directions and product development strategies in the quantum AI space.

What to do next

Monitor for updates

Source document (simplified)

← USPTO Patent Grants

Multiscale contiguous block pixel entangler for image recognition on hybrid quantum-classical computing system

Grant US12596951B2 Kind: B2 Apr 07, 2026

Assignee

IONQ, INC

Inventors

Ananth Prakash Kaushik, Sonika Johri, Jason John Iaconis, Soon Cheol Park, Hanlae Jo

Abstract

A method of performing implementing a quantum convolutional neural network (QCNN) in a hybrid quantum-classical computing system includes performing a data load operation, a set of a convolutional layer operation and a pooling operation, a measurement operation. The data load operation includes encoding pixel data of an input image onto a quantum processor using column qubits and row qubits. The convolutional layer operation includes a contiguous block pixel entangler that entangles a column qubit and a row qubit, depending on a pattern of a feature to detect in the input image. The pooling layer operation includes applying a series of one-qubit operations to the column qubits and the row qubits. The measurement operation includes measuring a state of an output qubit among the column qubits and the row qubits.

CPC Classifications

G06N 10/40 G06N 3/045 G06N 3/08 G06N 3/084 G06N 10/20 G06N 10/70 G06N 10/60 G06N 10/80 G06N 3/0464 G06N 3/09 G06N 10/00 G06N 3/04 G06N 3/048 G06N 3/088 G06V 10/82 G06V 10/955 B82Y 10/00