CN104849231B - A kind of method and device of plastic material ONLINE RECOGNITION - Google Patents
A kind of method and device of plastic material ONLINE RECOGNITION Download PDFInfo
- Publication number
- CN104849231B CN104849231B CN201510242919.6A CN201510242919A CN104849231B CN 104849231 B CN104849231 B CN 104849231B CN 201510242919 A CN201510242919 A CN 201510242919A CN 104849231 B CN104849231 B CN 104849231B
- Authority
- CN
- China
- Prior art keywords
- sample
- wavelength
- matrix
- classes
- waste
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of method and device of plastic material ONLINE RECOGNITION, this method sets up a kind of infrared spectrum identification model with versatility, according to the material species of identification raw material, set up identification model, the purpose that particular plastic material is recognized from waste or used plastics rubbish is reached, the environmental pollution that garbage processing procedure is caused is reduced.Meanwhile, by the ir data of the various materials of comprehensive analysis, characteristic information extraction reduces the spectroscopic data amount that identification process need to be measured, reduces data processing complexity, improve recognition speed.The device realizes that it possesses the ability of dynamic adjustment and extension measurement wavelength, pair can handle the material type of raw material without particular/special requirement, wide adaptability based on this method.
Description
Technical field
The present invention relates to Material Identification technical field, more particularly to a kind of method and device of plastic material ONLINE RECOGNITION.
Background technology
With the development and the raising of the level of consumption of China's plastics industry, plastic products are given birth in the application of China from industrial or agricultural
Produce clothing, food, lodging and transportion -- basic necessities of life ubiquitous.While huge material progress is brought, the generation of plastic refuse is also brought
A series of societies and environmental problem.Waste or used plastics in domestic waste are generally made up of multi items plastic hybrid, common class
Type includes：Polyethylene terephthalate (PET), polyethylene (PE), polypropylene (PP), polystyrene (PS), acrylonitrile-benzene
Ethylene-butadiene copolymer (ABS), polyamide (PA), polycarbonate resin (PC), polyvinyl chloride (PVC) etc..
Wherein, PVC industrialized productions more than 70 years, its abundant raw material is easy to get, and can be obtained after being mixed with each analog assistant
The form of obtaining is abundant, the product of different properties, with superior cost performance, the quilt in industrial and agricultural production, construction material, daily life
Extensive use, the environmental pressure brought is also particularly significant.Due to chloride height in PVC material, in pure PVC polymer resins, chlorine element is about
The 57% of gross mass is accounted for, to improve the heat endurance and mechanical property of material, addition metalline is needed in process of production (such as
Lead, barium, calcium, cadmium or organo-tin compound) stabilizer and the plasticizer based on phthalic acid ester, to PVC discarded objects
Improper processing can trigger a variety of environmental problems.
At present, the processing means of the common waste or used plastics of China include landfill disposal, burn and reclaim heat energy, at reclaiming
The approach such as reason.Under the conditions of landfill, stabilizer contained by PVC discarded objects and plasticizer decomposition can separate out heavy metallic salt and harmful gas
Body, is polluted to soil and water source.During using the means of burning processing waste or used plastics rubbish, not only contain a large amount of chlorine in release gas
Compound, causes air and water acidification, also greater risk generation high toxicity carcinogen dioxin.In addition, PVC materials are pyrolyzed
The hydrochloric acid that process is produced, to generating apparatus again also just larger harm.Because PVC discarded objects bring a series of pollution problem, identification
And the PVC material in separating waste, worn plastics is to realize one of garbage harmless, key technology of recycling treatment.
At present, the method for identification waste or used plastics material includes manual identified, electrostatic separation, gravity floatation, X-ray spectrum point
The means such as analysis, infrared spectrum analysis, wherein infrared spectrum analysis are relatively advanced, with extensive technology.Infra-red radiation and thing
When matter interacts, with the frequency multiplication of molecular vibrational frequency or combining the close wavelength location of frequency and can have stronger absorption,
Due to the difference of chemical analysis and functional group, variety classes plastics are substantially poor to existing in the Absorption Characteristics of near infrared range
It is different, by measuring transmission or reflection spectrum of the material near infrared region, plastic type can be identified.
When carrying out plastic material identification based on method of infrared spectrophotometry analysis, widely used technology is divided into two classes.The first kind
The two-dimension spectrum curve of technology measurement of species near infrared region, by the means such as principal component analysis and pattern-recognition with it is known
The indicatrix of material is contrasted, and judges measured matter composition；Homogenous material is only relied upon during such technology identification material to consolidate
Some spectral signatures, while various material can be identified for classification, but because the data volume of measurement and the processing of needs is big, be
System realization is more complicated, and recognition speed is slow, it is difficult to meet the requirement to recognition speed when handling waste or used plastics rubbish.Equations of The Second Kind technology
By measuring the radiation intensity of plastic sample different wave length position transmission or reflection near infrared range, the knowledge to material is realized
Not, still, such technology garbage raw material adaptability changeable to complicated components is poor, especially because comprising a variety of in plastic garbage
The plastic material of type, the identification to particular plastic material in plastic garbage raw material can not be realized using this scheme.
The content of the invention
It is an object of the invention to provide a kind of method and device of plastic material ONLINE RECOGNITION, answering for data processing is simplified
Miscellaneous degree, can fast and accurately realize the identification of particular plastic material.
The purpose of the present invention is achieved through the following technical solutions：
A kind of method of plastic material ONLINE RECOGNITION, this method includes：
Step A, the various material plastic products included from waste plastic raw material select more than one sample, and pressed respectively
Sample is divided into a classes and non-a classes according to plastic material a to be detected；Spectral measurement is carried out to each sample, and to the spectrum of acquisition
Measurement result data are pre-processed；
Step B, with certain sampling interval the pre-processed results of each sample are weighed within spectra collection scope
Sampling obtains one-dimensional spectroscopic data sequence, calculates the relative ratio of each data point in one-dimensional spectral sequence, obtains each sample
Dual wavelength relative ratio matrix, then calculate the deviation of a classes and non-a classes sample relative ratio matrix, obtain deviation matrix；
Step C, the sample for non-a classes, are compared using a default proportion threshold value and the deviation matrix, are obtained
Corresponding identity matrix；Further according to the identity matrix of each non-a classes sample, respectively by element carry out with computing and and computing, obtain
Obtain the overall identity matrix of sample and statistical matrix；N is chosen from overall recognition matrix and/or statistical matrix to identification wavelength and every
Wavelength ratio of the corresponding a classes sample of a pair of identification wavelength at the wavelengthI=1~N；
Step D, waste or used plastics raw material to be measured is placed in bispectrum section detector identification region and according to the knot in step C
Fruit sets detector operation wavelength, and measures waste or used plastics raw material diffusing reflection radiation to be measured in first pair of intensity recognized at wavelength
Ratio ki, further according to ratio kiWith ratioDifference and threshold value magnitude relationship, judge current waste or used plastics raw material to be measured as a
Class or non-a classes, so as to realize plastic material ONLINE RECOGNITION.
It is described that spectral measurement is carried out to each sample, and the results of spectral measurements data progress pretreatment of acquisition is included：
Spectral measurement is carried out with default spectral resolution and wave-length coverage to each sample, reflectance spectrum is gathered, obtains
Corresponding spectral absorption curve；
The smooth and standard normalizing of default points is carried out to the spectral absorption curve of each sample using spectral manipulation software
Change is handled；
Wherein, default spectral resolution is 1~3nm, and default wave-length coverage is 1100nm~2000nm, presets points
For 5~13 points.
It is described that the pre-processed results of each sample are adopted again with certain sampling interval within spectra collection scope
Sample obtains one-dimensional spectroscopic data sequence, calculates the relative ratio of each data point in one-dimensional spectral sequence, obtains each sample
Dual wavelength relative ratio matrix, then the deviation of a classes and non-a classes sample relative ratio matrix is calculated, obtain deviation matrix；Including：
The pre-processed results of each sample are designated as A (λ), and resampling result is designated as A'(k)；
The relative ratio matrix K (m, n) of the result of resampling represents each sample spectra curve at wavelength points m and n
Spectroscopic data ratio, is expressed as K (m, n)=A'(m)/A'(n)；Flat is asked by element to the relative ratio matrixes of all a classes samples
Average, its result is designated as Ka, the relative ratio matrix of each non-a classes sample is designated as KS；
To each non-a classes sample, its relative ratio matrix K is calculatedSWith a class sample relative ratio matrix KsaBetween element difference
Absolute value, be designated as the deviation matrix D of the sampleS(m, n), is expressed as：DS(m, n)=| KS-Ka|。
The sample for non-a classes, is compared using a default proportion threshold value and the deviation matrix, obtains phase
The identity matrix answered；Further according to the identity matrix of each non-a classes sample, respectively by element carry out with computing and and computing, obtain
Sample totality identity matrix and statistical matrix include：
For the sample of each non-a classes, a default proportion threshold value T and deviation matrix D are utilizedS(m, n) is compared, and is obtained
Obtain corresponding identity matrix MS(m, n), is expressed as：
For the identity matrix M of each non-a classes sampleS(m, n), respectively by element carry out with computing and and computing, obtain
Sample totality identity matrix P0(m, n) and statistical matrix Q0(m, n), is expressed as：
P0(m, n)=∪ MS(m,n)；
Q0(m, n)=∑ MS(m,n)。
The N that chosen from overall recognition matrix or statistical matrix is to identification wavelength and the corresponding a of every a pair of identification wavelength
Wavelength ratio of the class sample at the wavelengthIncluding：
If overall identity matrix P0(m, n) is null matrix, then from statistical matrix Q0Greatest member correspondence is searched in (m, n)
Index value (m, n), as the 1st pair identification wavelength, be designated asWithAnd record wavelength ratio of a classes sample at the wavelengthThen, removed from the sample in step A and meet MSThe sample of (m, n)=1, and recalculate the overall mark of remaining sample
Know matrix P1(m, n) and statistical matrix Q1(m, n), if overall identity matrix P1(m, n) is still null matrix, then repeatedly abovementioned steps,
Obtain the 2nd pair of identification wavelengthWithAnd corresponding ratioAnd continue to remove in the 2nd pair of selected identification wavelengthWithPlace meets MSThe sample of (m, n)=1, calculates remaining sample overall identity matrix P again2(m, n) and statistical matrix Q2(m,
n)；Repeat the above steps, until obtaining N-1 to identification wavelength；
If the overall identity matrix P that the n-th that repeats the above steps is calculatedN-1There are one or more nonzero elements,
A pair of wavelength are then selected from all nonzero elementsWithAs last to identification wavelength, and a class samples are recorded at this
Ratio at wavelength
It is described that waste or used plastics raw material to be measured is placed in bispectrum section detector identification region and according to the result in step C
Detector operation wavelength is set, and measures waste or used plastics raw material diffusing reflection radiation to be measured in first pair of strength ratio recognized at wavelength
Value ki, further according to ratio kiWith ratioDifference and threshold value magnitude relationship, judge current waste or used plastics raw material to be measured as a classes
Or non-a classes include：
Waste or used plastics raw material to be measured is placed in bispectrum section detector identification region and set according to the result in step C
The operation wavelength that detector is the 1st time isWithAnd waste or used plastics raw material diffusing reflection radiation to be measured is measured in first pair of identification ripple
The intensity rate k of strong point1, calculateIf d1＜ T, wherein, T is threshold value, then changes the work of detector the 2nd time
Wavelength isWithRepeat said process；
In above process, if there is di＞ T then judge the waste or used plastics raw material to be measured as non-a classes；If N is to identification ripple
In length, all diIt is satisfied by di＜ T, then the waste or used plastics raw material to be measured is a classes.
A kind of device of plastic material ONLINE RECOGNITION, the device is arranged in the conveying device of waste or used plastics raw material to be measured
Side, it includes：
Broadband IR source, pre-objective, beam splitter, two sets of identical optical processing structures, and it is foregoing for performing
The microprocessing unit of method；Wherein, the optical processing structure includes：Collimator objective, Tunable filters, the optically focused set gradually
Object lens and infrared sensor；
The infrared light radiation that broadband IR source is sent converges in waste or used plastics raw material surface to be measured, plastic sample diffusing reflection
Infra-red radiation collected by pre-objective, and converge on the beam splitter plane of incidence；Infra-red radiation is after beam splitter internal transmission
It is divided into two-way, respectively from two outgoing end face outgoing of beam splitter into optical processing structure；
Incident Tunable filters after collimator objective collimation in optical processing structure；Tunable filters working condition is by micro-
The control of processing unit, possesses function of the dynamic adjustment through spectral coverage centre wavelength, is filtered out from incident light radiation infrared
Arrowband monochromatic radiation；Filtered monochromatic radiation is received by collecting objective, is converged on the photosurface of infrared sensor, completes spoke
Intensity is penetrated to the conversion of electric signal；
Microprocessing unit is sequentially adjusted in passing through centre wavelength according to identification wavelength pair, control two-way Tunable filters, and
The output signal of two-way infrared detector is gathered, intensity and relative scale of the incident IR radiation in each Detection wavelength is calculated,
Complete the identification to waste or used plastics raw material to be measured.
The beam splitter is the single-core fiber beam splitting that two single multimode fibres use fused biconical taper technique to be processed into
Device, including an incidence channel and two exit channels, form y-type structure, and optical fiber fusion site is fixed with curing agent.
The Tunable filters are acousto-optic tunable filter or liquid crystal tunable harmonic wave device；
The center wavelength tuning scope of the Tunable filters is 1100nm~2000nm, and bandwidth is 5nm~30nm.
The infrared sensor includes：Infrared photodiode and amplifying circuit, photodiode is by incident IR radiation
Current signal is converted to, signal amplitude scope is voltage signal output after 0.1nA~1uA, amplified processing of circuit.
As seen from the above technical solution provided by the invention, the program sets up infrared based on Infrared Spectrum Technology
Spectral matching factor model, the spectroscopic data amount that compression identification process need to be gathered, so as to simplify the complexity of data processing, is realized to spy
Determine the fast and accurately identification of plastic material.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, being used required in being described below to embodiment
Accompanying drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for this
For the those of ordinary skill in field, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is a kind of flow chart of the method for plastic material ONLINE RECOGNITION that the embodiment of the present invention one is provided；
Fig. 2 is the flow chart for setting up identification model that the embodiment of the present invention one is provided；
Fig. 3 is the flow chart for the plastic material ONLINE RECOGNITION that the embodiment of the present invention one is provided；
Fig. 4 is the curve of spectrum schematic diagram of the PVC that the embodiment of the present invention one is provided and non-PVC materials sample；
Fig. 5 is the overall identity matrix schematic diagram that the embodiment of the present invention one is provided；
Fig. 6 is numeric distribution and the signal of maximum element index wavelength for the statistical matrix that the embodiment of the present invention one is provided
Figure；
Fig. 7 is the schematic diagram for various samples reflectance spectrum ratio at selected wavelength points that the embodiment of the present invention one is provided；
Fig. 8 is the schematic diagram to motion sample identification that the embodiment of the present invention two is provided；
Fig. 9 is a kind of schematic diagram of the device for plastic material ONLINE RECOGNITION that the embodiment of the present invention two is provided；
Figure 10 is the principle schematic for the beam splitter that the embodiment of the present invention two is provided；
Figure 11 is that the device for the plastic material ONLINE RECOGNITION that the embodiment of the present invention two is provided realizes that ONLINE RECOGNITION and sorting are useless
The schematic diagram of old plastic raw materials.
Embodiment
With reference to the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on this
The embodiment of invention, the every other implementation that those of ordinary skill in the art are obtained under the premise of creative work is not made
Example, belongs to protection scope of the present invention.
It is an object of the present invention to for plastic garbage recycling and the demand of harmless treatment, a kind of online knowledge of research
The method and related device of other waste plastics material, it is such as ethylene glycol terephthalate (PET), poly- from containing various common plastics
Ethene (PE), polyvinyl chloride (PVC), polypropylene (PP), polystyrene (PS), acrylonitrile-butadiene-styrene copolymer
(ABS), in the waste or used plastics raw material of polyamide (PA), polycarbonate resin (PC) etc., the sheet of specified plastic material is identified
Or block object.It is described in detail with reference to specific embodiment.
Embodiment one
Fig. 1 is a kind of flow chart of the method for plastic material ONLINE RECOGNITION that the embodiment of the present invention one is provided.Such as Fig. 1 institutes
Show, this method mainly comprises the following steps：
Step 11, set up identification model；As shown in Fig. 2 it mainly includes：
Step A, sampling, collection spectroscopic data and spectroscopic data pretreatment.
The various material plastic products included from waste plastic raw material select more than one sample respectively, and according to be checked
Sample is divided into a classes and non-a classes by the plastic material a of survey；Spectral measurement is carried out to each sample, and to the spectral measurement knot of acquisition
Fruit data are pre-processed；In real work, sample can also be classified again after pre-processing.
It is described that spectral measurement is carried out to each sample in the embodiment of the present invention, and to the results of spectral measurements data of acquisition
Carrying out pretreatment includes：Spectral measurement is carried out with default spectral resolution and wave-length coverage to each sample, reflected light is gathered
Spectrum, obtains corresponding spectral absorption curve；Preset is carried out to the spectral absorption curve of each sample using spectral manipulation software
Several smooth and standard normalizeds；Wherein, default spectral resolution is 1~3nm, and default wave-length coverage is 1100nm
~2000nm, it is 5~13 points to preset points.
Step B, spectroscopic data processing.
1) resampling is carried out to the pre-processed results of each sample with certain sampling interval within spectra collection scope
One-dimensional spectroscopic data sequence is obtained, the relative ratio of each data point in one-dimensional spectral sequence is calculated, the double of each sample are obtained
Wavelength relative ratio matrix, then the deviation of a classes and non-a classes sample relative ratio matrix is calculated, obtain deviation matrix.
Wherein, the pre-processed results of each sample are designated as A (λ), and resampling result is designated as A'(k), resampling at intervals of 2~
30nm, preferably 2nm.
The relative ratio matrix K (m, n) of the result of resampling represents each sample spectra curve at wavelength points m and n
Spectroscopic data ratio, is expressed as K (m, n)=A'(m)/A'(n), m, n ∈ k；Member is pressed to the relative ratio matrix of all a classes samples
Element is averaged, and its result is designated as Ka, the relative ratio matrix of each non-a classes sample is designated as KS。
To each non-a classes sample, its relative ratio matrix K is calculatedSWith a class sample relative ratio matrix KsaBetween element difference
Absolute value, be designated as the deviation matrix D of the sampleS(m, n), is expressed as：DS(m, n)=| KS-Ka|。
2) for the sample of non-a classes, it is compared using a default proportion threshold value and the deviation matrix, obtains corresponding
Identity matrix；
Specifically, for the sample of each non-a classes, a default threshold value T and deviation matrix D are utilizedS(m, n) is compared
Compared with the corresponding identity matrix M of acquisitionS(m, n), is expressed as：
Due to there is symmetry in the physical sense than value matrix, by the limitation to element index by identity matrix MS
(m, n) is processed as triangle battle array, to simplify follow-up calculate.The threshold value T spans are 0.1~1.0.Identity matrix intermediate value is 1
Element illustrate under conditions of the horizontal T of given threshold value, can distinguish the non-a using a pair of characteristic wavelengths corresponding with element index
Class sample and a class samples.
Step C, extraction characteristic wavelength.
According to the difference of unlike material sample spectra feature, extraction can characterize several characteristic wavelength points of material type, use
In the identification of plastic material.Wherein each two characteristic wavelength is constituted a pair, in the embodiment of the present invention according to aforementioned spectral data at
The result of reason, calculates the overall recognition matrix and statistical matrix of non-a classes sample to choose N to identification wavelength and every a pair of identification ripple
Wavelength ratio of the long corresponding a classes sample at the wavelengthI=1~N.Detailed process is as follows：
1) to the identity matrix of all non-a classes samples, respectively by element carry out with computing and and computing, obtain sample overall
Identity matrix P0(m, n) and statistical matrix Q0(m, n), is expressed as：
P0(m, n)=∪ MS(m,n)；
Q0(m, n)=∑ MS(m,n)。
If 2) overall identity matrix P0(m, n) is null matrix (i.e. P0(m, n)=0), then from statistical matrix Q0In (m, n)
The corresponding index value (m, n) of greatest member is searched, as the 1st pair of identification wavelength, is designated asWithAnd a class samples are recorded at this
Wavelength ratio at wavelengthThen, removed from the sample in step A and meet MSThe sample of (m, n)=1, by it is foregoing 1) in
Formula recalculates the overall identity matrix and statistical matrix of remaining sample, and P is designated as respectively1(m, n) and Q1(m, n), if overall mark
Know matrix P1(m, n) is still null matrix, then repeatedly abovementioned steps, obtains the 2nd pair of identification wavelengthWithAnd corresponding ratioAnd continue to remove in the 2nd pair of selected identification wavelengthWithPlace meets MSThe sample of (m, n)=1, to remaining sample again
It is secondary to calculate overall identity matrix P2(m, n) and statistical matrix Q2(m,n)；Repeat the above steps, until obtaining N-1 to identification wavelength
(i.e. until overall identity matrix PN-1There are one or more nonzero elements).
If 3) the overall identity matrix P that the n-th that repeats the above steps is calculatedN-1There are one or more non-zero entries
Element, then select a pair of wavelength from all nonzero elementsWithAs last to identification wavelength, and record a class samples
The ratio at the wavelength
Step 12, plastic material ONLINE RECOGNITION.As shown in figure 3, main process is as follows：
Waste or used plastics raw material to be measured is placed in bispectrum section detector identification region and set according to the result in step C
Detector operation wavelength, and waste or used plastics raw material diffusing reflection radiation to be measured is measured in first pair of intensity rate recognized at wavelength
ki, further according to ratio kiWith ratioDifference and threshold value magnitude relationship, judge current waste or used plastics raw material to be measured as a classes or
Non- a classes, so as to realize plastic material ONLINE RECOGNITION；Specifically：
Waste or used plastics raw material to be measured is placed in bispectrum section detector identification region and set according to the result in step C
The operation wavelength that detector is the 1st time isWithAnd waste or used plastics raw material diffusing reflection radiation to be measured is measured in first pair of identification ripple
The intensity rate k of strong point1, calculateIf d1＜ T, wherein, T is threshold value (consistent with aforesaid threshold values T), then change is visited
The operation wavelength for surveying device the 2nd time isWithRepeat said process；
In above process, if there is di＞ T then judge the waste or used plastics raw material to be measured as non-a classes；If N is to identification ripple
In length, all diIt is satisfied by di＜ T, then the waste or used plastics raw material to be measured is a classes.
In order to make it easy to understand, being described further with reference to two specific examples to the present invention.
Example one
(referred to recognizing and sorting PVC plastic in the waste or used plastics raw material comprising two kinds of materials of PVC and ABS in this example
PVC materials are determined for exemplified by foregoing plastic material a), implementation process is comprised the following steps：
1) 26 samples are collected from waste or used plastics raw material, using etalon spectrometers, to each sample with spectral resolution
3nm, wave-length coverage 1100nm~2000nm, measures reflection spectrum curve respectively, then smooth with 5 points of spectral manipulation software progress
And standard normalized, the curve of spectrum of each sample is obtained, and sample is classified according to PVC and non-PVC materials, such as scheme
Shown in 4, wherein using the curve of circles mark as PVC sample spectrum, the curve marked using square is ABS sample spectra.
2) to gather sample spectroscopic data with the resampling of 2nm intervals after, calculate PVC sample dual wavelength ratio matrix simultaneously
Average and obtain typical ratio matrix KPVC.Dual wavelength ratio matrix K is calculated to each ABS samplesS, and deviation is calculated respectively
Matrix：
DS(m, n)=| KS-KPVC|
Threshold value T=0.5 is taken to calculate sample identification matrix MS(m,n)：
For the identity matrix M of each non-a classes sampleS(m, n), respectively by element carry out with computing and and computing, obtain
Sample totality identity matrix P0(m, n) and statistical matrix Q0(m, n), is expressed as：
P0(m, n)=∪ MS(m,n)；
Q0(m, n)=∑ MS(m,n)。
Gained totality identity matrix is as shown in figure 5, wherein black picture element corresponds to P0(m, n) nonzeros, root
Determine that Detection wavelength is according to the index value of non-zero entry vegetarian refreshments in the A of regionWithAnd record a class samples
Product ratio at the wavelength.
Mode according still further to abovementioned steps 12 can then realize the ONLINE RECOGNITION of PVC materials.
Example two
From a variety of plastic materials in this example, PVC, ABS, PA, PE, PET, PP, PS, the waste or used plastics raw material of composition are included
The middle raw material for recognizing and sub-electing PVC materials, implementation process is comprised the following steps：
Choose 70 samples, measure spectrum data altogether from waste or used plastics raw material, and classified, spectral measurement and processing
Parameter is consistent with precedent, repeats no more.
Based on sample spectral data, selection threshold value T=0.25 calculates overall identity matrix P0(m, n) and statistical matrix Q0(m,
n).Overall identity matrix P in this example0(m, n) is full null matrix, then based on statistical matrix Q0(m, n) choosing then measures ripple for first pair
It is long.The numeric distribution and maximum element index wavelength of statistical matrix are shown in Fig. 6, select first to recognize wavelength pair accordingly：WithTypical ratioTo recognizing wavelength to can not correctly recognize using first
Sample, calculates overall identity matrix P again1(m, n), choosing second pair of identification wavelength according to matrix non-zero element index isWithTypical ratioAs shown in fig. 7, for various material samples in selected ripple
(Fig. 7 a are at long pointWithFig. 7 b areWithWhen) reflected light
Spectrum ratio, rectangle marked is PVC material sample ratios, and spider lable is non-PVC materials sample ratio, from data, is used
Above-mentioned two pairs of measurements wavelength can be identified from the mixed material comprising ABS, PA, PA66, PET, PS, PE, PP and PVC component
PVC materials.
The such scheme of the embodiment of the present invention mainly has the following advantages that：
1) a kind of infrared spectrum identification model with versatility is set up, according to the material species of identification raw material, sets up and knows
Other model, reaches the purpose that particular plastic material is recognized from waste or used plastics rubbish, reduces the environment that garbage processing procedure is caused
Pollution.
2) by the ir data of the various materials of comprehensive analysis, characteristic information extraction, reducing identification process needs measurement
Spectroscopic data amount, reduce data processing complexity, improve recognition speed.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment can
To be realized by software, the mode of necessary general hardware platform can also be added to realize by software.Understood based on such,
The technical scheme of above-described embodiment can be embodied in the form of software product, the software product can be stored in one it is non-easily
The property lost storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in, including some instructions are to cause a computer to set
Standby (can be personal computer, server, or network equipment etc.) performs the method described in each embodiment of the invention.
Embodiment two
The embodiment of the present invention provides a kind of device of plastic material ONLINE RECOGNITION, and it is former that the device is arranged on waste or used plastics to be measured
Above the conveying device of material, waste or used plastics raw material to be measured under the drive of conveying device by the detection zone of the device, it is different
Moment detector will change to the observation area of sample, as shown in Figure 8.Because the reflectivity of raw material surface diverse location is deposited
In difference, the bulk strength of raw material reflectance spectrum will change and change with observation area.The content for analyzing previous embodiment one can
Know, when recognizing raw material material by measuring the radiation intensity of multiple characteristic wavelength points, two in each pair characteristic wavelength need to be ensured
The measured zone of wavelength points is consistent, dual wavelength radiation intensity rate is correctly reflected the spectral signature of material.And due to identification side
Independently calculating that observation position between the ratio data of each wavelength pair, each group characteristic wavelength changes in method then will not be accurate to identification
True property produces influence.A kind of device of the plastic material ONLINE RECOGNITION provided in the embodiment of the present invention is that infrared double-wave length detects dress
Put, the paired extension based on the complete twin detector operation wavelength of tunable filter part and dynamic are adjusted, and use beam splitter
Incident radiation is divided into two-way by energy proportion, the influence to identification accuracy by sample motion is eliminated.Can by comprehensive utilization
The timesharing measurement capability of tunable filter part and the energy segmentation ability of beam splitter, reach the mixed species modeling to quickly moving
Material carries out the purpose of ONLINE RECOGNITION.
As shown in figure 9, the device mainly includes：
Broadband IR source 8, pre-objective 1, beam splitter 2, two sets of identical optical processing structures, and for execution before
The microprocessing unit 7 of the methods described of generation embodiment one；Wherein, the optical processing structure includes：The collimator objective set gradually
(3a, 3b), Tunable filters (4a, 4b), collecting objective (5a, 5b) and infrared sensor (6a, 6b)；
The infrared light radiation 9 that broadband IR source 8 is sent converges in the surface of waste or used plastics raw material 11 to be measured, and plastic sample overflows
The infra-red radiation 10 of reflection is collected by pre-objective 1, and is converged on the plane of incidence of beam splitter 2；Infra-red radiation is in beam splitter
It is divided into two-way after portion's transmission, respectively from two outgoing end face outgoing of beam splitter into optical processing structure；
Incident Tunable filters (4a, 4b) after collimator objective (3a, 3b) collimation in optical processing structure；It is tunable to filter
Piece (4a, 4b) working condition is possessed function of the dynamic adjustment through spectral coverage centre wavelength by the control of microprocessing unit 7, from entering
Infrared arrowband monochromatic radiation is filtered out in the light radiation penetrated；Filtered monochromatic radiation is received by collecting objective (5a, 5b), convergence
On infrared sensor (6a, 6b) photosurface, radiation intensity is completed to the conversion of electric signal；
Microprocessing unit 7 is sequentially adjusted in passing through center according to identification wavelength pair, control two-way Tunable filters (4a, 4b)
Wavelength, and two-way infrared detector (6a, 6b) output signal is gathered, incident IR radiation is calculated in the strong of each Detection wavelength
Degree and relative scale, complete the identification to waste or used plastics raw material to be measured.
In the embodiment of the present invention, as shown in Figure 10, the beam splitter 2 is two singles to the principle of the beam splitter 2
Multimode fibre uses the single-core fiber beam splitter that fused biconical taper technique is processed into, including an incidence channel and two outgoing to lead to
Road, forms y-type structure, optical fiber fusion site fixed with curing agent, it is ensured that three passage relative position relations are fixed, with eliminate because
External force effect causes the splitting ratio fluctuation that beam splitter structure change is produced, it is ensured that measurement is accurate.The single multimode fibre core diameter
For 0.6mm~1.0mm, preferably 1.0mm, numerical aperture is 0.39~0.6, preferably 0.48.Due to using single-core fiber, enter
Enter the infra-red radiation of fibre core via the multiple total reflection in transmitting procedure, it is random to enter in two exit channels, therefore by incidence end
Difference is assigned to two outgoing end faces into the equal equal proportion of infra-red radiation of beam splitter on face, so as to ensure that two-way is infrared
The infra-red radiation that detector is collected comes from sample surfaces the same area, improves the accuracy measured sample dual wavelength ratio, letter
The resetting difficulty of dual wavelength detector is changed.In addition, in the embodiment of the present invention, it is possible to use beam splitter prism or light splitting plain film conduct
Beam splitter member, as long as above-mentioned functions can be realized.
In the embodiment of the present invention, the center wavelength tuning scopes of the Tunable filters (4a, 4b) for 1100nm~
2000nm, passband width is 5nm~30nm, and the centre wavelength of optical filter free transmission range can be dynamic under the control of external electric signal
Adjustment, to filter out monochromatic component of the sample reflected radiation at specified wavelength.The Tunable filters can be that acousto-optic can
Tunable filter or liquid crystal tunable harmonic wave device；This two classes wave filter is respectively provided with without motion device, and the fast advantage of spectral coverage switch speed can
Completed in 10ms to the adjustment through spectral coverage, so that up to hundreds of time spectral scan speed.
In the embodiment of the present invention, the infrared sensor (6a, 6b) includes：Infrared photodiode and amplifying circuit, light
Incident IR radiation is converted to current signal by electric diode, and signal amplitude scope is 0.1nA~1uA, amplified processing of circuit
Voltage signal output afterwards.Amplifying circuit uses high-gain I-V conversion plans, and gain amplifier is 107-109V/A, response speed
It hurry up, export no-delay, realize the quick measurement to faint radiation signal.
Microprocessing unit 7 in the embodiment of the present invention can use industrial computer or Implementation of Embedded System, as long as can
The method for realizing step 12 described in previous embodiment one.In addition, function that the microprocessing unit 7 is realized is implemented
Mode has had a detailed description in embodiment one, therefore repeats no more herein.
In order to make it easy to understand, being described further with reference to two specific examples to the present invention.Following two
Individual example is corresponding with two examples in embodiment one, and the process of setting up of identification model refers in embodiment one two examples
Content, only describes the process being identified using the present apparatus herein.
Example one
Figure 11 is device (the infrared double-wave length detection dress based on plastic material ONLINE RECOGNITION provided in an embodiment of the present invention
Put) realize ONLINE RECOGNITION and sort the schematic diagram of waste or used plastics raw material.
Infrared double-wave length detection means is arranged on top, setting height(from bottom) 500mm in the middle part of conveyer belt.The preceding glove of detection means
The mirror angle of visual field is 3 °, and viewing plane is located at transmission belt surface.Broadband IR source (light source module) is 50W halogen tungsten lamp, light source
The field of view of pre-objective is illuminated in the broadband radiation sent after convergence.According to the knot for sampling and modeling to waste or used plastics raw material
Really (referring to the example one in embodiment one), the running parameter of infrared double-wave length detection means is set to by control computer：Inspection
Survey wavelength 1736nm, 1680nm, typical ratio kPVC=0.7, recognition threshold T=0.5.
Waste or used plastics raw material drops down onto down conveyor belt surface one by one by feeding mechanism control, and conveyer belt is transported with 3m/s speed
It is dynamic, drive raw material to pass through illumination region, its surface reflection is collected by detection means, by measuring two selected characteristic wavelength points
Radiation intensity, ratio calculated k, and according to selected typical ratio kPVCIdentify whether with threshold value T as PVC plastic.Sample
0.2≤k of dual wavelength ratio≤1.2 when, be identified as PVC materials, on the contrary then be identified as non-PVC materials (ABS), recognition result hair
Deliver to control computer.Control module and then the jet performing module for controlling conveying end of tape, discharge pressure-air, will reach and divide
The PVC raw materials of favored area are kicked down into PVC collecting boxs.
Example two
The structure used is similar with aforementioned exemplary one, will not be repeated here.According to waste or used plastics raw material sampling and
The result (example two of previous embodiment one) of modeling, the running parameter of double UV check device is set to T=0.25.
Raw material to be measured enters after measured zone, and dual wavelength detector is first to measure wavelength
Calculate spectral ratio k1, and and standard valueCompare, non-PVC materials are then identified as beyond threshold interval；Otherwise it is switched to survey
Measure wavelengthCalculate spectral ratio k2, and and standard valueCompare, know beyond threshold interval
Wei not non-PVC materials；Such as twice threshold judges to pass through, then sample is PVC materials.After the completion of identification, detection means will be recognized
As a result send to control module to complete the sorting operation to PVC.
Such scheme of the embodiment of the present invention also has the following advantages that in addition to having the advantages that described in previous embodiment one：
1) device possesses the ability of dynamic adjustment and extension measurement wavelength, pair can handle the material type of raw material without special
It is required that, wide adaptability.
2) using the segmentation incident radiation of single-core fiber beam splitter, dual-wavelength measurement is realized, it is ensured that two-way infrared sensor is seen
Examine that region is completely the same, improve the accuracy of spectral measurement.
3) infrared sensor uses infrared photodiode and high-gain I-V amplifying circuits, realizes to faint radiation signal
Real-time measurement, spectral measurement speed is fast, reaches the purpose of quick ONLINE RECOGNITION.
The foregoing is only a preferred embodiment of the present invention, but protection scope of the present invention be not limited thereto,
Any one skilled in the art is in the technical scope of present disclosure, the change or replacement that can be readily occurred in,
It should all be included within the scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of claims
Enclose and be defined.
Claims (10)
1. a kind of method of plastic material ONLINE RECOGNITION, it is characterised in that this method includes：
Step A, the various material plastic products included from waste plastic raw material select more than one sample respectively, and according to treating
Sample is divided into a classes and non-a classes by the plastic material a of detection；Spectral measurement is carried out to each sample, and to the spectral measurement of acquisition
Result data is pre-processed；
Step B, the pre-processed results progress resampling within spectra collection scope with certain sampling interval to each sample
One-dimensional spectroscopic data sequence is obtained, the relative ratio of each data point in one-dimensional spectral sequence is calculated, the double of each sample are obtained
Wavelength relative ratio matrix, then the deviation of a classes and non-a classes sample relative ratio matrix is calculated, obtain deviation matrix；
Step C, the sample for non-a classes, are compared using a default proportion threshold value and the deviation matrix, obtain corresponding
Identity matrix；Further according to the identity matrix of each non-a classes sample, respectively by element carry out with computing and and computing, obtain sample
This overall identity matrix and statistical matrix；N is chosen from overall identity matrix and/or statistical matrix to identification wavelength and every a pair
Recognize wavelength ratio of the corresponding a classes sample of wavelength at the wavelengthI=1~N；
Step D, waste or used plastics raw material to be measured is placed in bispectrum section detector identification region and set according to the result in step C
Detector operation wavelength is put, and measures waste or used plastics raw material diffusing reflection radiation to be measured in first pair of intensity rate recognized at wavelength
ki, further according to ratio kiWith ratioDifference and default proportion threshold value magnitude relationship, judge that current waste or used plastics to be measured are former
Expect for a classes or non-a classes, so as to realize plastic material ONLINE RECOGNITION.
2. according to the method described in claim 1, it is characterised in that described that spectral measurement is carried out to each sample, and to obtaining
Results of spectral measurements data carry out pretreatment include：
Spectral measurement is carried out with default spectral resolution and wave-length coverage to each sample, reflectance spectrum is gathered, obtains corresponding
Spectral absorption curve；
At the smooth and standard normalization for carrying out default points to the spectral absorption curve of each sample using spectral manipulation software
Reason；
Wherein, default spectral resolution is 1~3nm, and default wave-length coverage is 1100nm~2000nm, and it is 5 to preset points
~13 points.
3. according to the method described in claim 1, it is characterised in that it is described within spectra collection scope between certain sampling
Resampling is carried out every the pre-processed results to each sample and obtains one-dimensional spectroscopic data sequence, calculates each in one-dimensional spectral sequence
The relative ratio of data point, obtains the dual wavelength relative ratio matrix of each sample, then calculates a classes and compared with non-a classes sample
The deviation of value matrix, obtains deviation matrix；Including：
The pre-processed results of each sample are designated as A (λ), and resampling result is designated as A'(k)；
The relative ratio matrix K (m, n) of the result of resampling represents spectrum of each sample spectra curve at wavelength points m and n
Data ratio, is expressed as K (m, n)=A'(m)/A'(n)；The relative ratio matrix of all a classes samples is averaged by element,
Its result is designated as Ka, the relative ratio matrix of each non-a classes sample is designated as KS；
To each non-a classes sample, its relative ratio matrix K is calculatedSWith a class sample relative ratio matrix KsaBetween element difference it is exhausted
To value, the deviation matrix D of the sample is designated asS(m, n), is expressed as：DS(m, n)=| KS-Ka|。
4. according to the method described in claim 1, it is characterised in that the sample for non-a classes, utilize a default ratio
Threshold value is compared with the deviation matrix, obtains corresponding identity matrix；Further according to the identity matrix of each non-a classes sample,
Respectively by element carry out with computing and and computing, obtaining the overall identity matrix of sample and statistical matrix includes：
For the sample of each non-a classes, a default proportion threshold value T and deviation matrix D are utilizedS(m, n) is compared, and obtains phase
The identity matrix M answeredS(m, n), is expressed as：
For the identity matrix M of each non-a classes sampleS(m, n), respectively by element carry out with computing and and computing, obtain sample it is total
Body identity matrix P0(m, n) and statistical matrix Q0(m, n), is expressed as：
P0(m, n)=∪ MS(m,n)；
Q0(m, n)=∑ MS(m,n)。
5. according to the method described in claim 1, it is characterised in that described to choose N from overall identity matrix or statistical matrix
To the wavelength ratio of identification wavelength and the corresponding a classes sample of every a pair of identification wavelength at the wavelengthIncluding：
If overall identity matrix P0(m, n) is null matrix, then from statistical matrix Q0The corresponding rope of greatest member is searched in (m, n)
Draw value (m, n), as the 1st pair of identification wavelength, be designated asWithAnd record wavelength ratio of a classes sample at the wavelengthSo
Afterwards, removed from the sample in step A and meet MSThe sample of (m, n)=1, and recalculate the overall identity matrix of remaining sample
P1(m, n) and statistical matrix Q1(m, n), if overall identity matrix P1(m, n) is still null matrix, then repeatedly abovementioned steps, obtains the
2 pairs of identification wavelengthWithAnd corresponding ratioAnd continue to remove in the 2nd pair of selected identification wavelengthWithPlace meets MS
The sample of (m, n)=1, calculates remaining sample overall identity matrix P again2(m, n) and statistical matrix Q2(m,n)；Repeat above-mentioned
Step, until obtaining N-1 to identification wavelength；
If the overall identity matrix P that the n-th that repeats the above steps is calculatedN-1There are one or more nonzero elements, then from
A pair of wavelength are selected in all nonzero elementsWithAs last to identification wavelength, and a classes sample is recorded in the wavelength
Locate ratio
6. method according to claim 5, it is characterised in that described that waste or used plastics raw material to be measured is placed on bispectrum section spy
Survey in device identification region and detector operation wavelength is set according to the result in step C, and measure waste or used plastics raw material to be measured and overflow
Reflected radiation is in first couple of intensity rate k recognized at wavelengthi, further according to ratio kiWith ratioDifference and threshold value size
Relation, judges that current waste or used plastics raw material to be measured is that a classes or non-a classes include：
Waste or used plastics raw material to be measured is placed in bispectrum section detector identification region and set according to the result in step C and is detected
The operation wavelength that device is the 1st time isWithAnd waste or used plastics raw material diffusing reflection radiation to be measured is measured at first pair of identification wavelength
Intensity rate k1, calculateIf d1＜ T, wherein, T is default proportion threshold value, then changes detector the 2nd time
Operation wavelength isWithRepeat said process；
In above process, if there is di＞ T then judge the waste or used plastics raw material to be measured as non-a classes；If N is to recognizing in wavelength,
All diIt is satisfied by di＜ T, then the waste or used plastics raw material to be measured is a classes.
7. a kind of device of plastic material ONLINE RECOGNITION, it is characterised in that the device is arranged on the defeated of waste or used plastics raw material to be measured
Send above device, it includes：
Broadband IR source, pre-objective, beam splitter, two sets of identical optical processing structures, and for perform claim requirement
The microprocessing unit of any one of 1-6 methods describeds；Wherein, two sets of identical optical processing structures are bispectrum section detector, described
Optical processing structure includes：Collimator objective, Tunable filters, collecting objective and the infrared sensor set gradually；
The infrared light radiation that broadband IR source is sent converges in waste or used plastics raw material surface to be measured, and plastic sample is irreflexive red
External radiation is collected by pre-objective, and is converged on the beam splitter plane of incidence；Infra-red radiation is divided into after beam splitter internal transmission
Two-way, respectively from two outgoing end face outgoing of beam splitter into optical processing structure；
Incident Tunable filters after collimator objective collimation in optical processing structure；Tunable filters working condition is by microprocessor
The control of unit, possesses function of the dynamic adjustment through spectral coverage centre wavelength, infrared arrowband is filtered out from incident light radiation
Monochromatic radiation；Filtered monochromatic radiation is received by collecting objective, is converged on the photosurface of infrared sensor, completes radiation strong
Spend the conversion of electric signal；
Microprocessing unit is according to identification wavelength pair, and control two-way Tunable filters are sequentially adjusted in passing through centre wavelength, and gather
The output signal of two-way infrared detector, calculates intensity and relative scale of the incident IR radiation in each Detection wavelength, completes
Identification to waste or used plastics raw material to be measured.
8. device according to claim 7, it is characterised in that the beam splitter is two single multimode fibres using molten
Melt the single-core fiber beam splitter for drawing cone technique to be processed into, including an incidence channel and two exit channels, form Y type knots
Structure, optical fiber fusion site is fixed with curing agent.
9. device according to claim 7, it is characterised in that the Tunable filters are acousto-optic tunable filter or liquid
Brilliant adjustable harmonic device；
The center wavelength tuning scope of the Tunable filters is 1100nm~2000nm, and bandwidth is 5nm~30nm.
10. device according to claim 7, it is characterised in that the infrared sensor includes：Infrared photodiode and
Incident IR radiation is converted to current signal by amplifying circuit, photodiode, and signal amplitude scope is 0.1nA~1uA, through putting
Voltage signal output after big processing of circuit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510242919.6A CN104849231B (en) | 2015-05-13 | 2015-05-13 | A kind of method and device of plastic material ONLINE RECOGNITION |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510242919.6A CN104849231B (en) | 2015-05-13 | 2015-05-13 | A kind of method and device of plastic material ONLINE RECOGNITION |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104849231A CN104849231A (en) | 2015-08-19 |
| CN104849231B true CN104849231B (en) | 2017-11-07 |
Family
ID=53849039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510242919.6A Expired - Fee Related CN104849231B (en) | 2015-05-13 | 2015-05-13 | A kind of method and device of plastic material ONLINE RECOGNITION |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104849231B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105775502A (en) * | 2016-03-15 | 2016-07-20 | 上海理工大学 | Dustbin distinguishing objects |
| CN106872393B (en) * | 2017-04-14 | 2019-08-06 | 北京国科虹谱光电技术有限公司 | Target materials recognition methods and device in a kind of plastic refuse |
| EP3450029A1 (en) * | 2017-09-01 | 2019-03-06 | TOMRA Sorting GmbH | Classification method and apparatus |
| CN108181258A (en) * | 2017-12-19 | 2018-06-19 | 东风汽车集团有限公司 | The identification method of plastics trademarks |
| CN108335295B (en) * | 2018-02-08 | 2019-04-30 | 扬州宇天玻璃有限公司 | Brightness adaptive equalization formula windowsill in a kind of ward |
| CN110898890A (en) * | 2019-11-28 | 2020-03-24 | 广州广电计量检测股份有限公司 | Grinding and crushing device and grinding method |
| CN112265753A (en) * | 2020-09-02 | 2021-01-26 | 北京致感致联科技有限公司 | Intelligent residential district rubbish discernment classification system |
| CN117169141B (en) * | 2023-11-03 | 2024-02-23 | 中国电力科学研究院有限公司 | Cable insulation material type identification and performance deduction method and device thereof |
| CN117688466B (en) * | 2024-02-02 | 2024-04-30 | 佛山市绿能环保有限公司 | Garbage class grabbing, classifying and identifying method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0493239A1 (en) * | 1990-12-20 | 1992-07-01 | Microfilm Archimed Inc. | Machine for sorting plastic bottles as a function of their plastic composition in order to recycle the same |
| CN1463359A (en) * | 2001-03-27 | 2003-12-24 | 松下电器产业株式会社 | Method and appts. for identifying plastic |
| CN1524178A (en) * | 2001-11-28 | 2004-08-25 | ��ʽ�������»����������� | Plastic identifying method |
| CN1826517A (en) * | 2003-07-18 | 2006-08-30 | 松下电器产业株式会社 | Device and method of discrimating plastic |
| CN103480586A (en) * | 2013-10-08 | 2014-01-01 | 合肥美亚光电技术股份有限公司 | Dual infrared online plastic material sorting device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6610981B2 (en) * | 2000-04-27 | 2003-08-26 | National Recovery Technologies, Inc. | Method and apparatus for near-infrared sorting of recycled plastic waste |
| JP2002267599A (en) * | 2001-03-07 | 2002-09-18 | Mitsubishi Heavy Ind Ltd | Quality of material identification system for plastic and quality of material identification/assortment system for plastic |
-
2015
- 2015-05-13 CN CN201510242919.6A patent/CN104849231B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0493239A1 (en) * | 1990-12-20 | 1992-07-01 | Microfilm Archimed Inc. | Machine for sorting plastic bottles as a function of their plastic composition in order to recycle the same |
| CN1463359A (en) * | 2001-03-27 | 2003-12-24 | 松下电器产业株式会社 | Method and appts. for identifying plastic |
| CN1524178A (en) * | 2001-11-28 | 2004-08-25 | ��ʽ�������»����������� | Plastic identifying method |
| CN1826517A (en) * | 2003-07-18 | 2006-08-30 | 松下电器产业株式会社 | Device and method of discrimating plastic |
| CN103480586A (en) * | 2013-10-08 | 2014-01-01 | 合肥美亚光电技术股份有限公司 | Dual infrared online plastic material sorting device |
Non-Patent Citations (4)
| Title |
|---|
| 基于近红外光谱技术的常见塑料快速分类方法研究;郭慧玲 等;《现代科学仪器》;20121031(第5期);第28-31,44页 * |
| 基于近红外特征波长反射率比值法鉴别塑料;张毅民 等;《现代化工》;20140131;第34卷(第1期);第165-168 * |
| 红外光谱仪在鉴定塑料材质的应用;孟玲 等;《广东化工》;20131106;第40卷(第10期);第54-55,76页 * |
| 鉴别废旧塑料的先进近红外技术;欧育湘 等;《化工进展》;20040108;第22卷(第3期);第267-270页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104849231A (en) | 2015-08-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104849231B (en) | A kind of method and device of plastic material ONLINE RECOGNITION | |
| US6313423B1 (en) | Application of Raman spectroscopy to identification and sorting of post-consumer plastics for recycling | |
| Scott | A two-colour near-infrared sensor for sorting recycled plastic waste | |
| JP2012042248A (en) | Raman scattering signal acquisition device and plastic discriminating device, and raman scattering signal acquisition method and plastic discriminating method | |
| Ahmad | A new technology for automatic identification and sorting of plastics for recycling | |
| JP2019520551A5 (en) | ||
| CN101949686A (en) | Online nondestructive testing (NDT) method and device for comprehensive internal/external qualities of fruits | |
| CN106872393B (en) | Target materials recognition methods and device in a kind of plastic refuse | |
| CN1515895A (en) | Apparatus for monitoring stripping material edge | |
| CN106824825A (en) | Waste and old ore method for separating and device based on LIBS | |
| CN103376251A (en) | Portable Raman spectrum measuring device and system for quickly detecting article | |
| CN109115719A (en) | A kind of Citrus Huanglongbing pathogen Band fusion rapid detection method based on high light spectrum image-forming technology | |
| US6433338B1 (en) | Method and device for identification of a type of material in an object and utilization therefor | |
| CN104297206A (en) | Quick liquid milk brand identifying method | |
| JPH09509477A (en) | Fruit and vegetable inspection and / or grading apparatus, method and use thereof | |
| JP2011226821A (en) | Identification device and identification method | |
| CN203287313U (en) | Rapid pesticide residue detection device | |
| CN106778815A (en) | Wheat quality THz spectral classification methods based on DS evidence theories | |
| JP4932053B1 (en) | Raman scattering signal acquisition device, Raman scattering identification device, Raman scattering signal acquisition method, and Raman scattering identification method | |
| KR101298109B1 (en) | Apparatus for color discrimination in visible light band and plastic classification system using thereof | |
| JPH08300354A (en) | Plastic classifying apparatus | |
| CN102967557B (en) | Biological aerosol water-content determination based on near infrared spectroscopy and classification method thereof | |
| Cesetti et al. | Waste processing: new near infrared technologies for material identification and selection | |
| CN202661379U (en) | Laser-induced breakdown spectroscopy (LIBS) powdery material belt type detection device | |
| JP5143175B2 (en) | Identification method and identification device based on Raman scattering, and Raman scattering spectrum measurement method and measurement device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171107 Termination date: 20190513 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |