20230122985. SINGLE-FRAME FRINGE PATTERN ANALYSIS METHOD BASED ON MULTI-SCALE GENERATIVE ADVERSARIAL NETWORK simplified abstract (NANJING UNIVERSITY OF SCIENCE AND TECHNOLOGY)
Contents
SINGLE-FRAME FRINGE PATTERN ANALYSIS METHOD BASED ON MULTI-SCALE GENERATIVE ADVERSARIAL NETWORK
Organization Name
NANJING UNIVERSITY OF SCIENCE AND TECHNOLOGY
Inventor(s)
SINGLE-FRAME FRINGE PATTERN ANALYSIS METHOD BASED ON MULTI-SCALE GENERATIVE ADVERSARIAL NETWORK - A simplified explanation of the abstract
This abstract first appeared for US patent application 20230122985 titled 'SINGLE-FRAME FRINGE PATTERN ANALYSIS METHOD BASED ON MULTI-SCALE GENERATIVE ADVERSARIAL NETWORK
Simplified Explanation
The invention is a method for analyzing fringe patterns using a multi-scale generative adversarial network. The network is trained using training data to generate the sine term, cosine term, and modulation image of a given fringe pattern. The phase is then computed using the arctangent function. The method eliminates the need for manual parameter tuning and provides an efficient and accurate phase calculation method for moving objects.
- A multi-scale generative adversarial network is constructed and trained using training data.
- The network takes a single fringe pattern as input and generates the sine term, cosine term, and modulation image.
- The phase is computed using the arctangent function.
- The method eliminates the need for manual parameter tuning during calculation.
- The method provides an efficient and high-precision phase calculation method for moving objects.
Potential Applications
- Optical metrology
- Interferometry
- Surface profilometry
- 3D shape measurement
Problems Solved
- Manual parameter tuning during calculation
- Inefficient and inaccurate phase calculation for moving objects
Benefits
- Eliminates the need for manual parameter tuning
- Provides an efficient and high-precision phase calculation method
- Suitable for analyzing fringe patterns of moving objects
Original Abstract Submitted
the invention discloses a single-frame fringe pattern analysis method based on multi-scale generative adversarial network. a multi-scale generative adversarial neural network model is constructed and a comprehensive loss function is applied. next, training data are collected to train the multi-scale generative adversarial network. during the prediction, a fringe pattern is fed into the trained multi-scale network where the generator outputs the sine term, cosine term, and the modulation image of the input pattern. finally, the arctangent function is applied to compute the phase. when the network is trained, the parameters of the network do not need to manually tune during the calculation. since the input of the neural network is only a single fringe pattern, the invention provides an efficient and high-precision phase calculation method for moving objects.
