CN203572772U - Cloud traditional Chinese medicine quality detection system based on multi-wavelength LED fluorescence spectrum - Google Patents

Abstract

The utility model relates to a cloud traditional Chinese medicine quality detection system based on multi-wavelength LED fluorescence spectrum. The cloud traditional Chinese medicine quality detection system comprises a detection terminal, a data collection and control system and a cloud server for processing data; the detection terminal is used for collecting fluorescence spectrum signals of traditional Chinese medicine; the data collection and control system is used for controlling the detection process; the cloud server is used for analyzing the principal components and quality factors of the traditional Chinese medicine and for generating traditional Chinese medicine quality information. The utility model is based on fluorescence spectrum and has the advantages of no damage, accuracy in detection, low cost and high detection efficiency.

Description

High in the clouds Quality Evaluation of Chinese Medicinal detection system based on multi-wavelength LED fluorescence spectrum

Technical field

The present invention relates to a kind of based on multi-wave length illuminating diode (Light Emitting Diode, LED) detection system that calculate in induced fluorescence spectrum and high in the clouds, the Quality Detection to different Chinese medicines be can realize, optical sensing technical field and laboratory instrument technical field belonged to.

Background technology

Chinese medicine is the peculiar medicine of Chinese Traditional Chinese Medicine, and it can be divided into Chinese patent drug and Chinese crude drug by processing technology, is the medicine and pharmacology wisdom crystallization that Chinese working people fights back the disease accumulated for thousands of years.Even in doctor trained in Western medicine present age prevailing, rural area and the city dweller of China still have very large dependence and demand to Chinese medicine.Data show, China has more than 80% city dweller to buy voluntarily Chinese patent drug.On the other hand, people also attempt Chinese medicine to introduce international market.

The traditional Chinese medicine research of China has also been obtained certain achievement, and Chinese Medicine Industry urgently further develops, but also has some problems.Wherein, the discriminating of quality of medicinal material and test are exactly a very important aspect.At present, people mainly rely on expert's evaluation, or by some most advanced and sophisticated instruments, the quality of Chinese crude drug or tcm product are detected.Ordinary populace is often difficult to distinguish the kind of Chinese medicine, true and false and grade.This has just limited the reasonable use of people to Chinese crude drug.Therefore, assessing quickly and efficiently Chinese medicine kind and grade, to guarantee that quality of medicinal material is reliable, drug safety, is an important problem.

In recent years, people have proposed some pick-up unit and methods to Quality Evaluation of Chinese Medicinal.As the micro-differential method that the patent No. is CN102621144A, can differentiate Chinese medicine realgar powder end.And for example patent CN101850070A adopts chromatography to carry out quality testing to Chinese medicine Tang blade.Also have scholar to propose fourier transform infrared spectroscopy and carry out the discriminating of the Fructus Amomi true and false.But these methods all exist defect, or need comparatively complicated preprocessing process, or need specific comparatively expensive instrument.In addition, these methods only detect for a certain specific Chinese medicine.

Summary of the invention

The present invention is directed to the deficiencies in the prior art, the present invention proposes a kind of Quality Evaluation of Chinese Medicinal detection system based on multi-wavelength LED induced fluorescence and cloud computing analytical technology.

The technical solution used in the present invention is:

The utility model is by sense terminals, Data Acquisition and Conversion System (DACS), and the Cloud Server three parts of deal with data composition.

Described sense terminals comprises multi-wavelength LED detection probe and optical filter wheel; Described multi-wavelength LED fluoroscopic examination probe for fixing six LED and a large core fiber, provides the dead end mouth of placing optical filter wheel simultaneously.Described optical filter wheel, for fixing the optical filter of six different-wavebands, is realized and being switched fast between optical filter, is used in conjunction with corresponding LED.

Described Data Acquisition and Conversion System (DACS) is by microprocessor, driving circuit, spectral detector, display module and Internet Transmission port composition; Microprocessor is by break-make and the luminous intensity of six LED of driving circuit control.The fluorescence signal of being collected by sense terminals, then shows by display module on the one hand to spectral detector by Optical Fiber Transmission after the pre-service of microprocessor, on the other hand by Internet Transmission port transmission to cloud server.

The beneficial effect that the present invention has is:

1. are harmless, the detections accurately based on fluorescence spectrum.

2. the cheapness of integrated LED, portable terminal have been adopted.

3. cloud server can be realized fast processing mass data, and energy real-time upgrading database is to expand the range of application of sense terminals.

Accompanying drawing explanation

Fig. 1 is high in the clouds Chinese medicine detection system data conveying flow figure.

Fig. 2 is multi-wavelength LED fluoroscopic examination probe.

Fig. 3 is the runner dish that can place six optical filters

Fig. 4 is traditional Chinese medicine quality detection system figure.

Fig. 5 is the normalization fluorescence spectrum of four kinds of coffee samples.

Fig. 6 (a), 6 (b), 6 (c), 6 (d) are the classification results of four kinds of coffee samples.

Fig. 7 is the testing result of nine grade Xihu Longjing Teas.

Embodiment

Below in conjunction with accompanying drawing, the present invention is further described:

With two concrete examples, the present invention is described further below, but do not limit protection scope of the present invention.

As shown in Figure 2, Figure 3 and Figure 4, when the concrete detection of carrying out sample, sample 4 is placed on the black aluminium sheet 3 without fluorescence signal.Microprocessor 9 is controlled driving circuit 7, and driving circuit 7 provides adjustable steady current for the LED on probe, thereby controls the bright dark and intensity of each LED.By runner dish 2, switch filter plate and coordinate different LED work.The exciting light 1 of LED outgoing is irradiated to after traditional Chinese medicine sample 4 to be measured, inspire the fluorescence signal 5 of traditional Chinese medicine sample, fluorescence signal 5 is by the optical filter on runner dish 2, after the exciting light (light that LED itself sends) that filtering intensity is larger, large core fiber 6 on probe, is transferred to spectrographic detection module 8.The fluorescence spectrum that spectrographic detection module 8 collects is sent to display module 10 through microprocessor 9, and by Network Interface Module 11, is sent to cloud server 15 simultaneously.The interface module 12 of Cloud Server will receive data-pushing to data management module 14 and carry out data storage, management, data are finally analyzed and are processed by cloud detection module 13, and the traditional Chinese medicine ingredients information obtaining after processing and quality factor are sent to sense terminals demonstration testing result by cloud interface module 12.

In conjunction with Fig. 1, think and provide take said apparatus as basic detection method step.

The classification of one: four kind of coffee samples of example

Coffee sample: on market, bought four kinds of different types of coffee beans, and grind into powder sample.They are respectively Brazil Coffee (Brazilian Coffee), Brazilian Coffee (Mandheling Coffee), Blue Mountain Coffee (Bluemoutain Coffee) and Jamaican Coffee (Jamaican Coffee).

1. gather LED(400nm) fluorescence spectrum of coffee samples under induction.

Fluorescence under LED induction, after the long pass filter of 450nm, gathers eight groups of spectroscopic datas for every kind of spectrometer for coffee samples (Ocean Optics USB2000), measures eight groups of background signals simultaneously.Remove background signal and obtain altogether 32 groups of spectroscopic datas, every group averages the normalization fluorescence spectrum figure that can obtain four kinds of coffees as shown in Figure 5 with smoothing processing by kind after being normalized respectively again.

2. pair fluorescence data carries out principal component analysis (PCA).

Fluorescence spectrum to record 32 groups of 300nm-1100nm carries out principal component analysis (PCA), and major component is pressed to the descending arrangement of variance contribution ratio, according to the requirement of contribution rate of accumulative total, gets a front T major component.Present case is got a front T=9 major component.

3. adopt leaving-one method to set up regression model.

Four kinds of coffee samples, measure eight times for every kind, total N=32 group spectroscopic data, and coffee kind is demarcated position m=1,2,3,4, represent respectively coffee sample Brazil (Brazilian), Man Telin (Mandheling), Lanshan County (Bluemoutain) and Jamaica (Jamaican) coffee.

Y _{k＝1,2...N,k≠i;1...M}＝Φ _{k＝1,2...N,k≠i;0...tr}θ _0...tr;1...M （1）

${\mathrm{\Φ}}_{k,j}=\left\{\begin{array}{cc}1& j=0;\\ U_{k,j}& j=\mathrm{1,2}...\mathrm{tr};\mathrm{}\end{array}\right.k=\mathrm{1,2}...N,k\≠i---\left(2\right)$

Y _{k=1,2...N, k ≠ i; 1...M}represent the preconditioning matrix of the given quality of coffee of expert; U _k,jrepresent j the contribution of major component to k sample; θ _{0...tr; 1...M}represent the linear coefficient matrix obtaining by equation of linear regression (1); N represents total spectrum sample given figure; I represents disallowable sample; M represents the number of variable in spectrum; T represents a front T major component.

The quality of coffee factor of the disallowable sample obtaining by linear regression model (LRM) can be expressed as:

${\hat{Y}}_{1...M}={\mathrm{\Φ}}_{i;0...\mathrm{tr}}{\mathrm{\θ}}_{0...\mathrm{tr};1...M}---\left(3\right)$

For present case, pre-service quality of coffee matrix:

$Y_{n,m}=\left\{\begin{array}{cc}1& n\∈m\\ 0& n\∉m\end{array}\right.n=\mathrm{1,2},...N---\left(4\right)$

Quote the quality that recurrence indexing parameter Q carrys out descriptive grade classification.

$Q_m=\frac{|{\mathrm{\μ}}_{{\hat{Y}}_{n\∈m}}-{\mathrm{\μ}}_{{\hat{Y}}_{n\∉m}}|}{\sqrt{{\mathrm{\σ}}_{{\hat{Y}}_{n\∈m}}^2-{\mathrm{\σ}}_{{\hat{Y}}_{n\∉m}}^2}}---\left(5\right)$

the consensus forecast kind value that expression obtains from regression algorithm for kind m.

represent

standard deviation.

4. interpretation of result.

Adopt the final classification results of Q value representation, as shown in Figure 6.Q value is larger, represents that this kind of coffee and other group coffee discriminations are higher.

5. multi-wavelength replenishment.

At single wavelength, classify in unconspicuous situation, can adopt multi-wavelength to supplement, finally selected best classification results, or classify in conjunction with certain two or some wavelength.The present case wavelength that places an order can obviously be distinguished four kinds of coffee samples.

The grade distinguishing of two: nine kinds of West Lake Dragon Well teas of example

Tealeaves sample: the Xihu Longjing Tea of nine grades is demarcated as 1,2,3,4,5,6,7,8,9 successively by differing from.

1. as the step 1 in example one and 2 is carried out spectra collection and principal component analysis (PCA).

Y _{k＝1,2...N,k≠i;1...M}＝Φ _{k＝1,2...N,k≠i;0...tr}θ _0...tr;1...M （1）

${\mathrm{\Φ}}_{k,j}=\left\{\begin{array}{cc}1& j=0;\\ U_{k,j}& j=\mathrm{1,2}...\mathrm{tr};\mathrm{}\end{array}\right.k=\mathrm{1,2}...N,k\≠i---\left(2\right)$

Y _{k=1,2...N, k ≠ i; 1...M}represent the preconditioning matrix of the given tea leaf quality of expert; U _k,jrepresent j the contribution of major component to k sample; θ _{0...tr; 1...M}represent the linear coefficient matrix obtaining by equation of linear regression (1); N represents total spectrum sample given figure; I represents disallowable sample; M represents the number of variable in spectrum; T represents a front T major component.

The tea leaf quality factor of the disallowable sample obtaining by linear regression model (LRM) can be expressed as:

${\hat{Y}}_{1...M}={\mathrm{\Φ}}_{i;0...\mathrm{tr}}{\mathrm{\θ}}_{0...\mathrm{tr};1...M}---\left(3\right)$

Preconditioning matrix can be expressed as:

$Y_{n_t}=n,n=\mathrm{1,2},...,9,t=\mathrm{1,2},...,5---\left(4\right)$

N represents tea grades, and t represents to measure number of times.

3. interpretation of result.

Grade distinguishing result as shown in Figure 7, linearity R=0.992.

Claims (1)

1. the high in the clouds Quality Evaluation of Chinese Medicinal detection system based on multi-wavelength LED fluorescence spectrum, by sense terminals, Data Acquisition and Conversion System (DACS), and the Cloud Server three parts of deal with data composition, is characterized in that:

Described sense terminals comprises multi-wavelength LED detection probe and optical filter wheel; Described multi-wavelength LED fluoroscopic examination probe for fixing six LED and a large core fiber, provides the dead end mouth of placing optical filter wheel simultaneously; Described optical filter wheel, for fixing the optical filter of six different-wavebands, is realized and being switched fast between optical filter, is used in conjunction with corresponding LED;

Described Data Acquisition and Conversion System (DACS) is by microprocessor, driving circuit, spectral detector, display module and Internet Transmission port composition; Microprocessor is by break-make and the luminous intensity of six LED of driving circuit control; The fluorescence signal of being collected by sense terminals, then shows by display module on the one hand to spectral detector by Optical Fiber Transmission after the pre-service of microprocessor, on the other hand by Internet Transmission port transmission to cloud server.

Images