Matlab experiment ghostwriting, ghostwriting PLS-LDA modeling method, ghostwriting neural network algorithm
1. Contents and requirements of the subject research
Main research contents:
1. Research the modeling method of wood knot imaging technology based on visible-near-infrared hyperspectral;
2. Establish a detection model for different defects (live knots, dead knots, etc.) of sawn timber through experimental research;
3. Simplify the built model , to determine a method for detecting surface defects in sawn timber using spectroscopy.
Requirements:
1. Preparation of test materials and test instruments, determination of test methods;
2. Design of test plan and implementation of test;
3. Analysis and processing of test data, and drawing conclusions;
2. The main program of the study
uses hyperspectral The specific contents of the research and analysis of the wood samples by imaging technology are as follows:
(1). The test samples include Douglas fir, hemlock and spruce. All the test materials were cut into the same size and thickness, and the joints were marked on the veneer in sequence, and the non-knots were randomly marked near the joints of the plate.
(2). Use the near-infrared hyperspectral image acquisition system to collect the hyperspectral image of the sample board (the data has been collected). And perform image correction on hyperspectral images before image processing.
(3). After denoising the spectrum obtained by the experiment, select the characteristic wavelength of the newly obtained spectrum. The selection of characteristic wavelengths can use partial least squares-linear discriminant analysis (PLS-LDA), least squares support vector machine (LS-SVM) pattern recognition and other methods.
(4) Fit and analyze the processed data with neural network method.
(5). Simplify the model and compare and analyze the best hyperspectral imaging technology to detect the knots on the wood surface.
3. Project Research Expected Objectives To
study the hyperspectral detection methods for the surface defects (knots) of sawn timber of different species, establish the spectral detection model of various surface defects of sawn timber, and simplify the model.
Data description:
The given spectral data is low beam data NIR
HQ Citi
TS Hemlock
YS Spruce
All data are divided into calibration set (Cal..) and test set (Test)
good spectrum is 1, bad spectrum is -1
good The spectrum is 1 The bad spectrum is 2
The required sorting and simplification:
1. Denoise the spectral data obtained by the experiment
(Spectral preprocessing methods mainly include: standard normal variable transformation algorithm preprocessing (SNV), multivariate scattering correction (MSC) preprocessing, first derivative algorithm (FD) preprocessing, second derivative algorithm (SD) preprocessing, Wavelet denoising, etc.)
You can choose one of them.
2. When modeling with the PLS-LDA modeling method, the principal component is determined by the lowest point of the average classification error of cross-validation, and the corresponding model is established to obtain the accuracy of the wood calibration set and the test set. Analysis of the results of the established PLS-LDA discriminant model for spectral preprocessing between different tree species
3. When modeling with LS-SVM modeling method, establish the corresponding model to obtain the accuracy of wood calibration set and test set. Parameter selection using grid search strategy combined with leave-one-out cross-validation
4. Use the neural network method to fit and analyze the data processed in the first step
(1.doc document has all the implementation codes, document 1.doc has all except the neural network part)
http://www.daixie0.com/contents/18/1304.html
Our field of direction: window programming, numerical algorithm, AI, artificial intelligence, financial statistics, econometric analysis, big data, network programming, WEB programming, communication programming, game programming, multimedia linux, plug-in programming program, API, image processing, embedded/MCU database programming, console process and thread, network security, assembly language hardware Programming software design engineering standards and regulations. The ghostwriting and ghostwriting programming languages or tools include but are not limited to the following:
C/C++/C# ghostwriting
Java ghostwriting
IT ghostwriting
Python ghostwriting
Tutored programming assignments
Matlab ghostwriting
Haskell ghostwriting
Processing ghostwriting
Building a Linux environment
Rust ghostwriting
Data Structure Assginment
MIPS ghostwriting
Machine Learning homework ghostwriting
Oracle/SQL/PostgreSQL/Pig database ghostwriting/doing/coaching
web development, website development, website work
ASP.NET website development
Finance Insurance Statistics Statistics, Regression, Iteration
Prolog ghostwriting
Computer Computational method
Because professional, so trustworthy. If necessary, please add QQ: 99515681 or email: [email protected]
WeChat: codinghelp