Halo-effect-free white light phase imaging method and system based on deep learning
A deep learning and imaging system technology, applied in neural learning methods, phase influence characteristic measurement, scientific instruments, etc., can solve the time-consuming, inability to correct the deviation and dependence between the phase value and the accurate value, etc., to achieve good real-time, optical Halo effect elimination, high accuracy effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0041] A white light phase imaging method without halo effect based on deep learning, the steps are as follows figure 1 Shown:
[0042]Step 1, data preparation and processing; obtain 1000 1024×1024pixels 2μm standard polystyrene microsphere phase images through the built white light diffraction phase imaging system as input data; through regular iterative calculation combined with system hardware parameters, the corresponding The phase image without halo effect is used as the target data; the image is rotated 90° clockwise, 90° counterclockwise, vertically flipped and horizontally flipped, and the data set is enlarged by four times, and the data set is 7:2: 1 is divided into training set, verification set and test set, where the training set and verification set are used for model training, and the test set does not participate in training and is used for model performance testing;
[0043] Step 2, model training and export; build the convolutional layer, activation layer, po...
Embodiment 2
[0064] A halo-free white light phase imaging system based on deep learning, including:
[0065] White light diffraction phase imaging module, used to collect standard polystyrene microsphere phase images of living biological cells;
[0066] The data processing module obtains more than 1,000 phase images of 1024 1024pixels 2μm standard polystyrene microspheres through the white light diffraction phase imaging module as input data; regular iterative calculations are performed on the input data to obtain corresponding halo-free phase image images , as the target data; the image is subjected to data augmentation operation, thereby expanding the data set, and dividing it into training set, verification set and test set at 7:2:1;
[0067] The deep neural network model building module builds a deep neural network model based on the training set, adjusts the number of layers and parameters of the model during the training process, saves a variety of models, and uses the untrained test...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com