CN101696935A - Apple rigidity nondestructive testing method based on hyperspectral space scattering curve - Google Patents
Apple rigidity nondestructive testing method based on hyperspectral space scattering curve Download PDFInfo
- Publication number
- CN101696935A CN101696935A CN200910236908A CN200910236908A CN101696935A CN 101696935 A CN101696935 A CN 101696935A CN 200910236908 A CN200910236908 A CN 200910236908A CN 200910236908 A CN200910236908 A CN 200910236908A CN 101696935 A CN101696935 A CN 101696935A
- Authority
- CN
- China
- Prior art keywords
- apple
- rigidity
- scattering curve
- space scattering
- hyperspectral
- 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.)
- Pending
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses an apple rigidity nondestructive testing method based on hyperspectral space scattering curves, comprising the following steps: selecting samples; acquiring hyperspectral data; measuring a standard rigidity value; processing the hyperspectral data; and establishing and evaluating a rigidity prediction model. The invention acquires the hyperspectral space scattering curves carrying with apple internal tissue (directly related with the rigidity of apples) information to obtain the fitting parameters of the hyperspectral space scattering curves and establishes the rigidity prediction model by utilizing the fitting parameters, thereby realizing the testing of the rigidity of the apples by using the rigidity prediction model.
Description
Technical field
The present invention relates to the agricultural product technical field of nondestructive testing, relate in particular to a kind of apple hardness fast non-destructive detection method based on hyperspectral space scattering curve.
Background technology
Hardness is an important indicator of fruit internal attribute, and this index is along with such as climate condition, harvest time or degree of ripeness, and receive post-processing approach and actual storage mode etc. differently have very big difference.Especially for apple, hardness number is as an important indicator of its built-in attribute, and the research of its detection means is more and more received publicity.China's apple variety is very abundant, and according to statistics, in about 1000 kinds in the whole world, China accounts for 800 kinds.Though the apple production height of present China, owing to detect the backwardness of classification technique, the outlet rate of China's apple is still very low.In addition, the consumer is in selecting the process of apple, and not only to the size of apple, exterior qualities such as shape and color have requirement, and increasing consumer also pays attention to the inside quality of apple, pol for example, hardness and acidity etc.
Degree of ripeness all has important meaning to storage and shelf life, and the orchard worker judges the mainly dependence experience of plucking time of apple.And hardness is to judge an important indicator of apple degree of ripeness, also is a key factor of fruit quality classification simultaneously.Detection method to apple internal quality mainly diminishes at present, and only is applicable to that the sampling sample segment detects, and the check rate is very low.Near infrared technology is a comparative maturity at Dynamic Non-Destruction Measurement, but studies show that, utilizes near infrared technology unsatisfactory to predicting the outcome of apple hardness.This mainly is because the physical arrangement of apple hardness and apple (as: institutional framework in the apple, the size of cell) is relevant, and near infrared technology is main relevant with the absorption (being the content of certain chemical substance) of chemical substance.In addition, because the limited space of near infrared technology, can only the test section dot information, the spatial information of fruit can not be provided.
Summary of the invention
The objective of the invention is at the deficiencies in the prior art, providing a kind of can carry out the method that quick nondestructive detects to apple hardness, and method of the present invention also can be used to detect the fruit that other and apple have similar shape.
For achieving the above object, the invention provides a kind of apple rigidity nondestructive testing method based on hyperspectral space scattering curve, may further comprise the steps:
S1, choose sample: the apple of selecting sizableness, the no obvious scar in surface is as sample;
S2, the collection of high-spectral data: utilize the high spectrum image acquisition system that each apple is gathered the spectrum picture that the equidistant n of place in equator is ordered, with the averaged spectrum image of n width of cloth image high spectrum image as this apple;
Wherein, high spectral technique is the new technology that image technique and spectral technique are combined, image is made up of spectrum peacekeeping space dimension, be on the continuous light spectral coverage to same atural object with tens of even hundreds of wave band imaging formations one by the tactic view data of spectrum " cube ".And fruit can be by under the particular space in the difference of the inner structure of particular space and chemical constitution, the spectral value performance of specific wavelength.When light enters apple internal, behind the interior tissue interaction between component of apple, can reflect from the incidence surface of apple, the light that reflects carries the information about apple internal quality.The space diffuse reflectance curve of different tissues structure carries different information.Therefore using high spectral technique can detect apple otherness spatially.
S3, measure standard rigidity value: utilize sclerometer to record the hardness number that n is ordered in the position of each apple being gathered spectrum picture, the mean value of the hardness number that this n is ordered is as the standard value of each apple hardness;
S4 is to the processing of high-spectral data: sample is divided into correction group and checking group; And use the described high spectrum image that intercepts each apple perpendicular to the straight line of spectrum axle, and obtain the space scattering curve of each apple, and described space scattering curve is carried out match, obtain the fitting parameter of space scattering curve;
S5, the foundation of rigidity prediction model and evaluation: adopt the number of chemical metering method to utilize described fitting parameter to set up rigidity prediction model based on correction group, and based on the checking group described rigidity prediction model is verified, choose optimum rigidity prediction model according to the comparative result of predicted value and standard value.
Wherein, in described step S2 and S3, n can get 4.
Wherein, in described step S2, described high spectrum image acquisition system can comprise CCD camera, hyperspectral imager, has the light source feed system and the computing machine of feedback controller.Described light source feed system comprises halogen tungsten lamp.Described hyperspectral imager is selected the object lens of 25mm for use, and the time shutter is 100ms, and the length of scanning line is 50mm, and the peak of apple is 170mm apart from the distance of object lens, and for avoiding saturated, sweep trace is 2mm apart from the distance of the beam center of light source.
Wherein, in described step S4, can sample be divided into correction group and checking group according to 3: 1 ratio.
Wherein, in described step S4, the step of " described space scattering curve is carried out match, obtain the fitting parameter of space scattering curve " is specifically as follows: with Lorentzian described space scattering curve is carried out nonlinear fitting,
The prototype of described Lorentzian is as follows:
Wherein, a is the asymptotic value of space scattering curve, and b is the peak value of space scattering curve, and c is the half-wave bandwidth of space scattering curve, and x is the scattering distance; A, b, c, x are described fitting parameter.
Wherein, in described step S5, the step of " adopting the number of chemical metering method to utilize described fitting parameter to set up rigidity prediction model based on correction group " is specifically as follows: based on correction group adopt respectively partial least square method and progressively multiple linear regression method in the 524nm-1016nm wavelength band, utilize described fitting parameter to set up rigidity prediction model.
Compared with prior art, technical scheme of the present invention has following advantage: the present invention obtains the fitting parameter of space scattering curve by obtaining the hyperspectral space scattering curve that carries apple internal tissue (directly related with the hardness of apple) information, utilize this parameter to set up rigidity prediction model, thereby can utilize the detection of this model realization apple hardness.
Description of drawings
Fig. 1 is the method flow diagram of the embodiment of the invention;
Fig. 2 is the structural representation of the high spectrum image acquisition system of the embodiment of the invention;
Fig. 3 is the high spectrum image that utilizes the apple that the method for the embodiment of the invention gathers;
Fig. 4 utilizes the space scattering curve under single wavelength that the method for the embodiment of the invention obtains and the relation of matched curve;
Fig. 5 is a related coefficient of utilizing the method for the embodiment of the invention to obtain with the Lorentzian match in whole wavelength band;
Fig. 6 adopts the partial least square method prediction result in the method for the embodiment of the invention;
Fig. 7 adopts progressively multiple linear regression method prediction result in the method for the embodiment of the invention.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used to illustrate the present invention, but are not used for limiting the scope of the invention.
As shown in Figure 1, the method according to the embodiment of the invention may further comprise the steps:
S1, the choosing of apple specimen material
Can buy 40 of ' Yantai Fuji apple ' apples in market, when buying, be more or less the same, do not have the sample of obvious surface scars from visually selected shape, size.The apple that purchase is returned is placed in 3 ℃ of environment to be stored, and did once experiment every three days, did 20 apples every day.Each experiment is taken out apple evening before that day, is placed in the room temperature environment 12 hours, is doing experiment after making it reach room temperature, and apple need simply clean before experiment.
S2, the collection of high-spectral data
The structure of high spectrum image acquisition system as shown in Figure 2, comprise high performance CCD (ChargeCoupled Device, charge-coupled image sensor) digital camera 5 (Sencicam QE Germany) (spectral range of covering is 400-1100nm), phase machine controller 6, hyperspectral imager (spectrometer) 4 (ImSpector V10E, Spectral Imaging Ltd., Finland), collimating mirror 3, optical fiber 2, the light source feed system that has feedback controller (comprises halogen tungsten lamp, model: OrielInstruments, USA) 1 and computing machine 7.During collection, with apple be placed on hyperspectral imager 4 under, adjusting the apple peak is 170mm apart from the distance of object lens, avoids illumination to arrive the big scar place of apple surface.Each apple is gathered 4 width of cloth spectrum pictures at equidistant place, equator, with the averaged spectrum image of 4 width of cloth images original high spectrum image (as shown in Figure 3) as this apple.
S3 measures standard rigidity value
After spectrum data gathering is finished, in the spectra collection position of correspondence, according to the assay method of national standard (GB10651-89) regulation, adopt hardness of fruit meter (model is the GY-1 type) to measure the hardness of apple, the mean value of 4 point hardnesses is as the hardness number of an apple.
S4 is to the processing of high-spectral data
With all samples according to being divided into correction group and checking group at 3: 1.In Matlab software, use straight line intercepting high spectrum image perpendicular to the spectrum axle, obtain the space scattering curve of apple, use the Lorentzian of 4 parameters that space scattering curve is carried out nonlinear fitting (as shown in Figure 4), obtain 4 fitting parameters.The prototype of Lorentzian is as follows:
In the formula, a is the asymptotic value of space scattering curve, and b is the peak value of space scattering curve, and c is the half-wave bandwidth of space scattering curve, and x is the scattering distance; A, b, c, x are described fitting parameter.
Because the defective of light source self in the certain limit at whole wave band two ends, adopts the related coefficient of Lorentzian match lower.And in the 524-1016nm wavelength band, the related coefficient of Lorentz match reaches (as shown in Figure 5) more than 0.99.Therefore,, improve the accuracy of forecast model, can select the wave band modeling in the 524-1016nm scope in order to reduce noise.
S5, the foundation of rigidity prediction model and evaluation:
Adopt partial least square method and stoechiometric process such as multiple linear regression progressively, in the 524nm-1016nm wavelength band, utilize fitting parameter to set up the forecast model of hardness based on correction group.Find by the modeling result of each fitting parameter relatively, utilize the modeling result of peak value (b) modeling of space scattering curve better.The result who utilizes the modeling of offset minimum binary method is respectively shown in Fig. 6 (1) and Fig. 6 (2): as can be seen from the figure, and the correlation coefficient r of predicted value and standard value in the correction group
c=0.89, calibration standard difference SEC=0.71; The correlation coefficient r of predicted value and standard value is concentrated in checking
v=0.88, validation criteria difference SEV=0.88.The result who utilizes progressively multiple linear regression modeling respectively shown in Fig. 7 (1) and Fig. 7 (2), as can be seen, the correlation coefficient r of predicted value and standard value in the correction group
c=0.93, calibration standard difference SEC=0.54; The correlation coefficient r of predicted value and standard value is concentrated in checking
v=0.76, validation criteria difference SEV=1.67.This result shows, utilizes predicting the outcome of partial least square method modeling better.
As can be seen from the above embodiments, embodiments of the invention obtain the fitting parameter of space scattering curve by obtaining the hyperspectral space scattering curve that carries apple internal tissue (directly related with the hardness of apple) information, utilize this parameter to set up rigidity prediction model, thereby can utilize the detection of this model realization apple hardness.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the technology of the present invention principle; can also make some improvement and modification, these improve and modification also should be considered as protection scope of the present invention.
Claims (8)
1. apple rigidity nondestructive testing method based on hyperspectral space scattering curve may further comprise the steps:
S1, the choosing of sample: the apple of selecting sizableness, the no obvious scar in surface is as sample;
S2, the collection of high-spectral data: utilize the high spectrum image acquisition system that each apple is gathered the spectrum picture that the equidistant n of place in equator is ordered, with the averaged spectrum image of n width of cloth image high spectrum image as this apple;
S3, the mensuration of standard rigidity value: utilize sclerometer to record the hardness number that n is ordered in the position of each apple being gathered spectrum picture, the mean value of the hardness number that this n is ordered is as the standard value of each apple hardness;
S4 is to the processing of high-spectral data: sample is divided into correction group and checking group; And use the described high spectrum image that intercepts each apple perpendicular to the straight line of spectrum axle, and obtain the space scattering curve of each apple, and described space scattering curve is carried out match, obtain the fitting parameter of space scattering curve;
S5, the foundation of rigidity prediction model and evaluation: adopt the number of chemical metering method to utilize described fitting parameter to set up rigidity prediction model based on correction group, and based on the checking group described rigidity prediction model is verified, choose optimum rigidity prediction model according to the comparative result of predicted value and standard value.
2. the apple rigidity nondestructive testing method based on hyperspectral space scattering curve as claimed in claim 1 is characterized in that in described step S2 and S3, n gets 4.
3. the apple rigidity nondestructive testing method based on hyperspectral space scattering curve as claimed in claim 1, it is characterized in that, in described step S2, described high spectrum image acquisition system comprises CCD camera, hyperspectral imager, has the light source feed system and the computing machine of feedback controller.
4. the apple rigidity nondestructive testing method based on hyperspectral space scattering curve as claimed in claim 1 is characterized in that, in described step S4, according to 3: 1 ratios sample is divided into correction group and checking group.
5. the apple rigidity nondestructive testing method based on hyperspectral space scattering curve as claimed in claim 1, it is characterized in that, in described step S4, the step of " described space scattering curve being carried out match; obtain the fitting parameter of space scattering curve " is specially: with Lorentzian described space scattering curve is carried out nonlinear fitting
The prototype of described Lorentzian is as follows:
Wherein, a is the asymptotic value of space scattering curve, and b is the peak value of space scattering curve, and c is the half-wave bandwidth of space scattering curve, and x is the scattering distance; A, b, c, x are described fitting parameter.
6. the apple rigidity nondestructive testing method based on hyperspectral space scattering curve as claimed in claim 1, it is characterized in that, in described step S5, the step of " adopting the number of chemical metering method to utilize described fitting parameter to set up rigidity prediction model based on correction group " is specially: based on correction group adopt respectively partial least square method and progressively multiple linear regression method in the 524nm-1016nm wavelength band, utilize described fitting parameter to set up rigidity prediction model.
7. the apple rigidity nondestructive testing method based on hyperspectral space scattering curve as claimed in claim 3, it is characterized in that, described hyperspectral imager is selected the object lens of 25mm for use, time shutter is 100ms, the length of scanning line is 50mm, the peak of apple is 170mm apart from the distance of object lens, and for avoiding saturated, sweep trace is 2mm apart from the distance of the beam center of light source.
8. the apple rigidity nondestructive testing method based on hyperspectral space scattering curve as claimed in claim 3 is characterized in that, described light source feed system comprises halogen tungsten lamp.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910236908A CN101696935A (en) | 2009-10-27 | 2009-10-27 | Apple rigidity nondestructive testing method based on hyperspectral space scattering curve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910236908A CN101696935A (en) | 2009-10-27 | 2009-10-27 | Apple rigidity nondestructive testing method based on hyperspectral space scattering curve |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101696935A true CN101696935A (en) | 2010-04-21 |
Family
ID=42142051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910236908A Pending CN101696935A (en) | 2009-10-27 | 2009-10-27 | Apple rigidity nondestructive testing method based on hyperspectral space scattering curve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101696935A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102507457A (en) * | 2011-11-18 | 2012-06-20 | 江苏大学 | Device and method for rapidly and nondestructively detecting crop nutrient elements |
| CN103616335A (en) * | 2013-12-02 | 2014-03-05 | 江苏大学 | Method for rapidly identifying capacity of acid-forming bacteria of fermented food in producing acid |
| CN104034587A (en) * | 2014-06-17 | 2014-09-10 | 浙江大学 | Measurement method for firmness of spherical fruits and vegetables |
| CN104647351A (en) * | 2013-11-24 | 2015-05-27 | 邢玉明 | Parallel manipulator of spectral imager |
| CN105136593A (en) * | 2015-08-20 | 2015-12-09 | 江苏大学 | Nondestructive test method for microhardness of metal surface layer subjected to laser shock processing |
| CN106124474A (en) * | 2016-05-16 | 2016-11-16 | 中国农业大学 | Meat harmful chemicals remains lossless method for quick |
| CN107505325A (en) * | 2017-08-18 | 2017-12-22 | 西北农林科技大学 | The comprehensive quality detecting method of winter-jujube fruit |
| CN108662842A (en) * | 2017-03-27 | 2018-10-16 | 青岛海尔智能技术研发有限公司 | The detecting system and refrigerator of food in refrigerator |
| CN109115708A (en) * | 2018-09-29 | 2019-01-01 | 西北农林科技大学 | A kind of more quality integration nondestructive detection systems of apple internal and method |
| CN109406267A (en) * | 2018-04-28 | 2019-03-01 | 东北电力大学 | A kind of three-dimensional reconstruction method of apple tongue touching mechanical information |
-
2009
- 2009-10-27 CN CN200910236908A patent/CN101696935A/en active Pending
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102507457A (en) * | 2011-11-18 | 2012-06-20 | 江苏大学 | Device and method for rapidly and nondestructively detecting crop nutrient elements |
| CN104647351A (en) * | 2013-11-24 | 2015-05-27 | 邢玉明 | Parallel manipulator of spectral imager |
| CN103616335A (en) * | 2013-12-02 | 2014-03-05 | 江苏大学 | Method for rapidly identifying capacity of acid-forming bacteria of fermented food in producing acid |
| CN104034587A (en) * | 2014-06-17 | 2014-09-10 | 浙江大学 | Measurement method for firmness of spherical fruits and vegetables |
| CN104034587B (en) * | 2014-06-17 | 2016-03-30 | 浙江大学 | A kind of measuring method of spheral fruit consolidation |
| CN105136593A (en) * | 2015-08-20 | 2015-12-09 | 江苏大学 | Nondestructive test method for microhardness of metal surface layer subjected to laser shock processing |
| CN106124474A (en) * | 2016-05-16 | 2016-11-16 | 中国农业大学 | Meat harmful chemicals remains lossless method for quick |
| CN106124474B (en) * | 2016-05-16 | 2019-02-05 | 中国农业大学 | Meat harmful chemicals remain lossless rapid detection method |
| CN108662842A (en) * | 2017-03-27 | 2018-10-16 | 青岛海尔智能技术研发有限公司 | The detecting system and refrigerator of food in refrigerator |
| CN107505325A (en) * | 2017-08-18 | 2017-12-22 | 西北农林科技大学 | The comprehensive quality detecting method of winter-jujube fruit |
| CN107505325B (en) * | 2017-08-18 | 2023-04-25 | 西北农林科技大学 | Omnibearing quality detection method for winter jujube fruits |
| CN109406267A (en) * | 2018-04-28 | 2019-03-01 | 东北电力大学 | A kind of three-dimensional reconstruction method of apple tongue touching mechanical information |
| CN109115708A (en) * | 2018-09-29 | 2019-01-01 | 西北农林科技大学 | A kind of more quality integration nondestructive detection systems of apple internal and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101696935A (en) | Apple rigidity nondestructive testing method based on hyperspectral space scattering curve | |
| Boldrini et al. | Hyperspectral imaging: a review of best practice, performance and pitfalls for in-line and on-line applications | |
| Wu et al. | Advanced applications of hyperspectral imaging technology for food quality and safety analysis and assessment: A review—Part I: Fundamentals | |
| Cayuela | Vis/NIR soluble solids prediction in intact oranges (Citrus sinensis L.) cv. Valencia Late by reflectance | |
| Khodabakhshian et al. | Development of a multispectral imaging system for online quality assessment of pomegranate fruit | |
| CN105021617B (en) | Whole strain chlorophyll content of rice measurement apparatus and method based on high light spectrum image-forming | |
| CN107185850B (en) | Corn seed activity detection device based on hyperspectral imaging and electrical impedance | |
| CN105548070B (en) | A kind of apple soluble solid near infrared detection position compensation method and system | |
| CN108444924B (en) | Method for discriminating storage period of tea by applying hyperspectral image technology | |
| CN103674864A (en) | Fish water content distribution detection method based on hyperspectral imaging technology | |
| CN105021535B (en) | A kind of paddy content of fatty acid lossless detection method and system | |
| CN105181606B (en) | The method that cane sugar content is distributed in peanut is detected based on high light spectrum image-forming technology | |
| Huang et al. | Meta-analysis of the detection of plant pigment concentrations using hyperspectral remotely sensed data | |
| CN107064044B (en) | Method and device for rapidly detecting polyphenol content in tea extract | |
| Lu | Imaging spectroscopy for assessing internal quality of apple fruit | |
| CN107132197B (en) | Method and device for detecting total acid content of vinegar | |
| Qi et al. | Establishment of a comprehensive indicator to nondestructively analyze watermelon quality at different ripening stages | |
| CN109946246A (en) | A kind of detection method and device of apple soluble solid | |
| CN108627498A (en) | A kind of flour doping quantitative detecting method of multispectral data fusion | |
| Nielsen et al. | Supercontinuum light sources for hyperspectral subsurface laser scattering: Applications for food inspection | |
| CN103592258B (en) | Detection method of tea polyphenol contents in tea | |
| CN207703714U (en) | A kind of double probe transmission-types for solid particulate matter spectrum non-destructive testing are popped one's head in | |
| Itoh et al. | Measurement of nitrate concentration distribution in vegetables by near-infrared hyperspectral imaging | |
| CN114839151A (en) | Method and device for detecting hardness of multi-fruit acid | |
| CN105115909B (en) | The method that fat content profiles in peanut are detected based on high light spectrum image-forming technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100421 |