A method for improving the recognition rate of edible oil adulteration spectrum detection
A spectrum detection and recognition rate technology, applied in the field of pattern recognition, can solve the problem of not being able to optimally distinguish different types of samples, and achieve the effect of improving the recognition rate
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0060] Embodiment 1: as Figure 1-5 As shown, a method to improve the detection and recognition rate of edible oil adulteration spectrum is to establish their own feature spaces for various samples, use the feature vectors of each feature space to reconstruct the original test samples, and calculate the reconstruction error, and reconstruct the sample with the smallest error as the category of the original test sample.
[0061] The concrete steps of described method are as follows:
[0062] Step1, vector matrix: randomly select M from the first type of samples 1 sample vectors as training samples, convert each vector into a column vector, and arrange them into N×M 1 matrix B of
[0063] in
[0064] Among them, N represents the number of data of a sample vector;
[0065] Step2, find the average vector: M 1 Average vector of sample vectors for:
[0066]
[0067] Step3. Construct the covariance matrix: each sample vector x j with the mean vector The difference ve...
Embodiment 2
[0079] Embodiment 2: as Figure 1-5 As shown, a method to improve the detection and recognition rate of edible oil adulteration spectrum is to establish their own feature spaces for various samples, use the feature vectors of each feature space to reconstruct the original test samples, and calculate the reconstruction error, and reconstruct the sample with the smallest error as the category of the original test sample.
Embodiment 3
[0081] Obtain 100 sets of spectral data of 4 types of experimental oil products through ultraviolet spectrometer, and the images of various ultraviolet-visible spectra are as attached Figures 2 to 5 As shown in , a group (that is, a sample vector) is randomly selected from 400 sets of spectral data as the original test sample (the original test sample selected at this time belongs to pure sesame oil).
[0082] Step1, vector matrix: randomly select 40 sample vectors (ie 40 groups) from 100 groups of pure sesame oil samples as training samples, convert each vector into a column vector, and arrange them into a matrix B of 1500×40;
[0083] where B=(x 1 , x 2 , ..., x 40 );
[0084] Step2, find the average vector: the average vector of 40 samples for:
[0085]
[0086] Step3. Construct the covariance matrix: each vector x j the difference vector y from the mean vector j for:
[0087]
[0088] Then the covariance matrix C of the training sample is:
[0089] C=AA T...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More