Related Experiment Video
Updated: Aug 7, 2025
Evaluation and Manipulation of Neural Activity Using Two-Photon Holographic Microscopy
Published on: September 16, 2022
Multi-frequency acoustic hologram generation with a physics-enhanced deep neural network.
Qin Lin1, Rujun Zhang2, Feiyan Cai2
1School of Information Engineering, Guangdong Medical University, Dongguan 523808, China; Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China.
A new deep neural network method, PhysNet_MFAH, designs multi-frequency acoustic holograms by integrating physical models. This physics-enhanced multi-frequency acoustic hologram (PhysNet_MFAH) approach rapidly generates high-quality holograms for diverse acoustic fields.
Area of Science:
- Acoustics
- Deep Learning
- Computational Physics
Background:
- Acoustic holograms enable precise control of sound fields.
- Designing multi-frequency acoustic holograms presents significant computational challenges.
- Existing methods often struggle with accuracy, speed, or multi-frequency capabilities.
Purpose of the Study:
- To introduce a novel deep neural network method for designing multi-frequency acoustic holograms.
- To enhance the accuracy and speed of acoustic hologram generation.
- To enable holographic rendering of complex acoustic fields at multiple frequencies.
Main Methods:
- Development of a physics-enhanced multi-frequency acoustic hologram deep neural network (PhysNet_MFAH).
- Integration of multiple physical models of acoustic wave propagation into the neural network architecture.
- Training the network with frequency-specific target acoustic patterns.
Main Results:
- PhysNet_MFAH automatically, accurately, and rapidly generates high-quality multi-frequency acoustic holograms.
- The method achieves superior reconstructed acoustic intensity fields compared to IASA and DS methods.
- Fast computational speed is maintained while improving hologram quality.
Conclusions:
- The proposed PhysNet_MFAH method offers a significant advancement in multi-frequency acoustic hologram design.
- It provides a robust and efficient tool for creating complex acoustic fields.
- Potential applications include dynamic particle manipulation and volumetric displays.
