Atom Cloud Detection Using a Deep Neural Network

Abstract

We use a deep neural network to detect and place region-of-interest boxes around ultracold atom clouds in absorption and fluorescence images---with the ability to identify and bound multiple clouds within a single image. The neural network also outputs segmentation masks that identify the size, shape and orientation of each cloud from which we extract the clouds' Gaussian parameters. This allows 2D Gaussian fits to be reliably seeded thereby enabling fully automatic image processing.

References in corpus (11)

• FaceNet: A Unified Embedding for Face Recognition and Clustering
• scikit-image: Image processing in Python
• Machine learning phases of matter
• Bose-Einstein condensation of atoms in a uniform potential
• Machine Learning Phases of Strongly Correlated Fermions
• Matter-wave diffraction from a quasicrystalline optical lattice
• Narrow-line magneto-optical trap for erbium
• Hermite-Gaussian Mode Detection via Convolution Neural Networks
• Machine learning assisted readout of trapped-ion qubits
• Single-exposure absorption imaging of ultracold atoms using deep learning
• Deep Learning-Assisted Classification of Site-Resolved Quantum Gas Microscope Images

Cited by in corpus (3)

• Efficient sorting of orbital-angular-momentum states with large topological charges and their unknown superpositions via machine learning
• Vortex detection in atomic Bose-Einstein condensates using neural networks trained on synthetic images
• Measuring Laser Beams with a Neural Network