CN107860722A - A kind of honeydew melon inside quality online test method and system - Google Patents
A kind of honeydew melon inside quality online test method and system Download PDFInfo
- Publication number
- CN107860722A CN107860722A CN201711041446.9A CN201711041446A CN107860722A CN 107860722 A CN107860722 A CN 107860722A CN 201711041446 A CN201711041446 A CN 201711041446A CN 107860722 A CN107860722 A CN 107860722A
- Authority
- CN
- China
- Prior art keywords
- spectrum
- melon
- pulp
- honeydew melon
- honeydew
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/255—Details, e.g. use of specially adapted sources, lighting or optical systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
Landscapes
- Physics & Mathematics (AREA)
- Biochemistry (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of honeydew melon inside quality online test method and system, methods described to include：Obtain the diffusing transmission spectrum and image of honeydew melon sample；Water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in pulp spectrum and whole melon spectrum is extracted respectively；Establish spectrum transformation model；Extract the color characteristic of image；Calculate the volume and fruit shape index of honeydew melon sample；According to the water stain weld defect control spectral information in pulp spectrum, soluble solid content sensitivity spectrum information, color characteristic, volume and fruit shape index in pulp spectrum, honeydew melon inside quality grade on-line checking model is established；Tested honeydew melon is detected using detection model, obtains the inside quality grade of tested honeydew melon.The present invention is eliminated influence of the non-edible tissue to testing result, can be greatly improved accuracy of detection using pulp as detection object, and spectral information is combined with surface and detected, and can further improve accuracy of detection.
Description
Technical field
The present invention relates to Quality Detection technical field, more particularly to a kind of honeydew melon inside quality online test method and is
System.
Background technology
Honeydew melon is the even fruit of a kind of thick pericarp, big fruit shape, streaking, is the important industrial crops in the Northwest.It is long
Since phase, maturity and quality division are all to use traditional artificial process or destructive sampling Detection in honeydew melon sales process,
Time and effort consuming, subjective factor influence big, detect and are classified extensive, cause the good and the bad to mix, reduce the market competitiveness of the fruit
And selling price.Solving above mentioned problem existing for honeydew melon can not be only merely by cultivation new varieties, improvement planting conditions and storage
Arc manipulation part, it is often more important that, focus on the Quality Detection technology in commercialization.
At present, generally use carries out Non-Destructive Testing, existing inspection based on spectrum and Digital image technology to the quality of honeydew melon
Survey method is not considered the influence of pericarp and fruit caving Nei Fei edible tissues to testing result, caused using whole melon as detection object
Accuracy of detection is not high.
The content of the invention
Based on this, it is necessary to provide a kind of honeydew melon inside quality online test method that can improve accuracy of detection and be
System.
To achieve the above object, the invention provides following scheme：
A kind of honeydew melon inside quality online test method, including：
Obtain honeydew melon sample diffusing transmission spectrum and honeydew melon sample image, the diffusing transmission spectrum include whole melon spectrum and
Pulp spectrum, the pulp spectrum are that whole melon removes the spectrum unless being obtained after edible tissue；
Extract the water stain weld defect control spectral information and soluble solid content sensitivity spectrum letter in the pulp spectrum
Breath；
Extract the water stain weld defect control spectral information and soluble solid content sensitivity spectrum letter in the whole melon spectrum
Breath；
Believe according to the water stain weld defect control spectral information in the pulp spectrum and soluble solid content sensitivity spectrum
Water stain weld defect control spectral information and soluble solid content sensitivity spectrum information in breath and the whole melon spectrum, establish
Spectrum transformation model, the spectrum transformation model are used to whole melon spectrum being converted to pulp spectrum；
The color characteristic of the image of the honeydew melon sample is extracted, the color characteristic includes being used to react the honeydew melon sample
The water stain defect color characteristic of water content and the soluble solid content face of the sugar content for reacting the honeydew melon sample
Color characteristic；
According to the image of the honeydew melon sample, the volume and fruit shape index of the honeydew melon sample are calculated；
According to the water stain weld defect control spectral information in the pulp spectrum, the soluble solid in the pulp spectrum
Content sensitivity spectrum information, the water stain defect color characteristic, the soluble solid content color characteristic, the volume and
The fruit shape index, establish honeydew melon inside quality grade on-line checking model；
Tested honeydew melon is detected using the honeydew melon inside quality grade on-line checking model, obtains tested honeydew melon
Inside quality grade.
Optionally, after the diffusing transmission spectrum of the acquisition honeydew melon sample and the image of honeydew melon sample, in addition to：To institute
The image of the diffusing transmission spectrum and the honeydew melon sample of stating honeydew melon sample is pre-processed respectively；
The diffusing transmission spectrum to the honeydew melon sample pre-processes, including noise elimination, background deduction, Norris
Filtering, multiplicative scatter correction, wavelet transformation and data normalization；
The image to the honeydew melon sample pre-processes, including morphology denoising, adaptive threshold fuzziness and right
Described image is marked.
Optionally, water stain the weld defect control spectral information and soluble solid content in the extraction pulp spectrum
Sensitivity spectrum information, is specifically included：
Obtain in the honeydew melon sample in the pulp spectrum of the honeydew melon of water content within a preset range and the honeydew melon sample
Pulp spectrum of the water content beyond the honeydew melon of the preset range；
Contrast in the honeydew melon sample in the pulp spectrum of the honeydew melon of water content within a preset range and the honeydew melon sample
Water content exceeds the pulp spectrum of the honeydew melon of the preset range, according to water content in the honeydew melon sample within a preset range
Difference of the water content between the pulp spectrum of the honeydew melon of the preset range in the pulp spectrum of honeydew melon and the honeydew melon sample
Different feature, determine the water stain weld defect control wave band in the pulp spectrum；
After first differential processing, least square fitting processing, multiplicative scatter correction and filtering are carried out to the pulp spectrum,
Determine soluble solid content sensitive band in the pulp spectrum；
According to soluble solid content in the water stain weld defect control wave band in the pulp spectrum and the pulp spectrum
Sensitive band, extract water stain weld defect control spectral information and soluble solid content sensitivity spectrum letter in the pulp spectrum
Breath；
Water stain weld defect control spectral information and soluble solid content sensitivity light in the extraction whole melon spectrum
Spectrum information, specifically include：
Obtain the pulp spectrum of honeydew melon of the water content in the preset range and the honeydew melon sample in the honeydew melon sample
Whole melon spectrum of the water content beyond the honeydew melon of the preset range in product；
Contrast in the honeydew melon sample in the whole melon spectrum of the honeydew melon of water content within a preset range and the honeydew melon sample
Water content exceeds the whole melon spectrum of the honeydew melon of the preset range, according to water content in the honeydew melon sample within a preset range
Difference of the water content between the whole melon spectrum of the honeydew melon of the preset range in the whole melon spectrum and the honeydew melon sample of honeydew melon
Different feature, determine the water stain weld defect control wave band in the whole melon spectrum；
First differential processing, least square fitting processing, multiplicative scatter correction and filtering are carried out to the whole melon spectrum
Afterwards, soluble solid content sensitive band in the whole melon spectrum is determined；
According to soluble solid content in the water stain weld defect control wave band in the whole melon spectrum and the whole melon spectrum
Sensitive band, extract water stain weld defect control spectral information and soluble solid content sensitivity spectrum letter in the whole melon spectrum
Breath.
Optionally, it is described to establish spectrum transformation model, specifically include：
By water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in the pulp spectrum
It is respectively converted into matrix form；
By water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in the whole melon spectrum
It is respectively converted into matrix form；
Matrix form and the institute of the water stain weld defect control spectral information in the whole melon spectrum are determined using spectra methods
The difference characteristic between the matrix form of the water stain weld defect control spectral information in pulp spectrum is stated, is established in the whole melon spectrum
Water stain weld defect control spectral information and the pulp spectrum in water stain weld defect control spectral information between transformation model；
The matrix of the soluble solid content sensitivity spectrum information in the whole melon spectrum is determined using spectra methods
Difference characteristic between the matrix form of soluble solid content sensitivity spectrum information in form and the pulp spectrum, builds
Found the soluble solid content sensitivity spectrum information in the whole melon spectrum and the soluble solid of the pulp spectrum contains
Measure the transformation model between sensitivity spectrum information.
Optionally, it is described to establish honeydew melon inside quality grade on-line checking model, specifically include：
According to the water stain weld defect control spectral information in the pulp spectrum, the water stain defect color characteristic, the body
Product and the fruit shape index, establish the water stain defect Nondestructive Testing Model of honeydew melon；
According to the soluble solid content sensitivity spectrum information in the pulp spectrum, the soluble solid content
Color characteristic, the volume and the fruit shape index, establish honeydew melon soluble solid content Nondestructive Testing Model；
According to the water stain defect Nondestructive Testing Model of the honeydew melon and the honeydew melon soluble solid content Non-Destructive Testing mould
Type, establish honeydew melon inside quality grade on-line checking model.
Present invention also offers a kind of honeydew melon inside quality on-line detecting system, including：
Data obtaining module, for obtaining the diffusing transmission spectrum of honeydew melon sample and the image of honeydew melon sample, the diffusing transmission
Spectrum includes whole melon spectrum and pulp spectrum, and the pulp spectrum is that whole melon removes the spectrum unless being obtained after edible tissue；
Pulp spectrum sensitive information extraction modules, for extracting the water stain weld defect control spectral information in the pulp spectrum
With soluble solid content sensitivity spectrum information；
Whole melon spectrum sensitive information extraction modules, for extracting the water stain weld defect control spectral information in the whole melon spectrum
With soluble solid content sensitivity spectrum information；
Transformation model establishes module, for according to the water stain weld defect control spectral information and solubility in the pulp spectrum
Water stain weld defect control spectral information and soluble solid in solid content sensitivity spectrum information and the whole melon spectrum
Content sensitivity spectrum information, establishes spectrum transformation model, and the spectrum transformation model is used to whole melon spectrum being converted to pulp light
Spectrum；
Color feature extracted module, the color characteristic of the image for extracting the honeydew melon sample, the color characteristic bag
Include the water content for reacting the honeydew melon sample water stain defect color characteristic and for react the honeydew melon sample containing sugar
The soluble solid content color characteristic of amount；
Volume and fruit shape index computing module, for the image according to the honeydew melon sample, calculate the honeydew melon sample
Volume and fruit shape index；
Detection model establishes module, for according to the water stain weld defect control spectral information in the pulp spectrum, the fruit
Soluble solid content sensitivity spectrum information, the water stain defect color characteristic in meat spectrum, the soluble solid
Content color characteristic, the volume and the fruit shape index, establish honeydew melon inside quality grade on-line checking model；
Detection module, for being detected using the honeydew melon inside quality grade on-line checking model to tested honeydew melon,
Obtain the inside quality grade of tested honeydew melon.
Optionally, the system also includes pretreatment module, and the pretreatment module specifically includes：
Pretreated spectra unit, it is described to the honey for being pre-processed to the diffusing transmission spectrum of the honeydew melon sample
The diffusing transmission spectrum of melon sample is pre-processed, including noise elimination, background deduction, Norris filtering, multiplicative scatter correction, small
Wave conversion and data normalization；
Image pre-processing unit, it is described to the honeydew melon sample for being pre-processed to the image of the honeydew melon sample
Image pre-processed, including morphology denoising, adaptive threshold fuzziness and described image is marked.
Optionally, the pulp spectrum sensitive information extraction modules, are specifically included：
Pulp spectrum acquiring unit, for obtaining the pulp of the honeydew melon of water content within a preset range in the honeydew melon sample
Pulp spectrum of the water content beyond the honeydew melon of the preset range in spectrum and the honeydew melon sample；
First sensitive band determining unit, for contrasting the honeydew melon of water content within a preset range in the honeydew melon sample
Pulp spectrum of the pulp spectrum with water content in the honeydew melon sample beyond the honeydew melon of the preset range, according to the honeydew melon sample
The pulp spectrum of the honeydew melon of water content within a preset range exceeds the preset range with water content in the honeydew melon sample in product
Honeydew melon pulp spectrum between difference characteristic, determine the water stain weld defect control wave band in the pulp spectrum；
Second sensitive band determining unit, for carrying out first differential processing, least square fitting to the pulp spectrum
After processing, multiplicative scatter correction and filtering, soluble solid content sensitive band in the pulp spectrum is determined；
Pulp spectrum sensitive information extraction unit, for according to the water stain weld defect control wave band in the pulp spectrum and institute
Soluble solid content sensitive band in pulp spectrum is stated, extracts the water stain weld defect control spectral information in the pulp spectrum
With soluble solid content sensitivity spectrum information；
The whole melon spectrum sensitive information extraction modules, are specifically included：
Whole melon spectrum acquiring unit, for obtaining honeydew melon of the water content in the preset range in the honeydew melon sample
Whole melon spectrum of the water content beyond the honeydew melon of the preset range in pulp spectrum and the honeydew melon sample；
3rd sensitive band determining unit, for contrasting the honeydew melon of water content within a preset range in the honeydew melon sample
Whole melon spectrum of the whole melon spectrum with water content in the honeydew melon sample beyond the honeydew melon of the preset range, according to the honeydew melon sample
The whole melon spectrum of the honeydew melon of water content within a preset range exceeds the preset range with water content in the honeydew melon sample in product
Honeydew melon whole melon spectrum between difference characteristic, determine the water stain weld defect control wave band in the whole melon spectrum；
4th sensitive band determining unit, for carrying out first differential processing, least square fitting to the whole melon spectrum
After processing, multiplicative scatter correction and filtering, soluble solid content sensitive band in the whole melon spectrum is determined；
Whole melon spectrum sensitive information extraction unit, for according to the water stain weld defect control wave band in the whole melon spectrum and institute
Soluble solid content sensitive band in whole melon spectrum is stated, extracts the water stain weld defect control spectral information in the whole melon spectrum
With soluble solid content sensitivity spectrum information.
Optionally, the transformation model establishes module, specifically includes：
Pulp spectral information converting unit, for by the water stain weld defect control spectral information in the pulp spectrum and solvable
Property solid content sensitivity spectrum information is respectively converted into matrix form；
Whole melon spectral information converting unit, for by the water stain weld defect control spectral information in the whole melon spectrum and solvable
Property solid content sensitivity spectrum information is respectively converted into matrix form；
First transformation model establishes unit, for determining that the water stain defect in the whole melon spectrum is quick using spectra methods
Difference between the matrix form of water stain weld defect control spectral information in the matrix form of photosensitive spectrum information and the pulp spectrum
Different feature, establish the water stain weld defect control spectral information in the whole melon spectrum and the water stain weld defect control in the pulp spectrum
Transformation model between spectral information；
Second transformation model establishes unit, for determining the soluble solid in the whole melon spectrum using spectra methods
The matrix form of thing content sensitivity spectrum information and the soluble solid content sensitivity spectrum information in the pulp spectrum
Difference characteristic between matrix form, the soluble solid content sensitivity spectrum information established in the whole melon spectrum with it is described
Transformation model between the soluble solid content sensitivity spectrum information of pulp spectrum.
Optionally, the detection model establishes module, specifically includes：
First detection model establishes unit, for according to the water stain weld defect control spectral information in the pulp spectrum, institute
Water stain defect color characteristic, the volume and the fruit shape index are stated, establishes the water stain defect Nondestructive Testing Model of honeydew melon；
Second detection model establishes unit, for according to the soluble solid content sensitivity spectrum in the pulp spectrum
Information, the soluble solid content color characteristic, the volume and the fruit shape index, establish honeydew melon soluble solid
Content Nondestructive Testing Model；
3rd detection model establishes unit, for can according to the water stain defect Nondestructive Testing Model of the honeydew melon and the honeydew melon
Dissolubility solid content Nondestructive Testing Model, establish honeydew melon inside quality grade on-line checking model.
Compared with prior art, the beneficial effects of the invention are as follows：
The present invention proposes a kind of honeydew melon inside quality online test method and system, by extracting the water in pulp spectrum
Water stain weld defect control in stain weld defect control spectral information and soluble solid content sensitivity spectrum information and whole melon spectrum
Spectral information and soluble solid content sensitivity spectrum information, establish the spectrum modulus of conversion that whole melon spectrum is changed to pulp spectrum
Type, the influence for using pulp as detection object, eliminating pericarp and fruit caving Nei Fei edible tissues to testing result is realized, can
Improve accuracy of detection；Contained using the water stain weld defect control spectral information in pulp spectrum, the soluble solid in pulp spectrum
Amount sensitivity spectrum information, water stain defect color characteristic, soluble solid content color characteristic, the volume of honeydew melon and fruit shape refer to
Number, establishes honeydew melon inside quality grade on-line checking model, realizes and is combined the spectral information of honeydew melon pair with surface
Honeydew melon inside quality grade is detected, and can further improve accuracy of detection.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to institute in embodiment
The accompanying drawing needed to use is briefly described, it should be apparent that, drawings in the following description are only some implementations of the present invention
Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these accompanying drawings
Obtain other accompanying drawings.
Fig. 1 is the flow chart of honeydew melon inside quality online test method of the embodiment of the present invention；
Fig. 2 is the stereometer nomogram of honeydew melon sample of the embodiment of the present invention；
Fig. 3 is the course diagram of honeydew melon inside quality on-line detecting system of the embodiment of the present invention.
Embodiment
Below in conjunction with 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
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
In order to facilitate the understanding of the purposes, features and advantages of the present invention, it is below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is further detailed explanation.
Fig. 1 is the flow chart of honeydew melon inside quality online test method of the embodiment of the present invention.
Referring to Fig. 1, honeydew melon inside quality online test method includes：
Step S1：The diffusing transmission spectrum of honeydew melon sample and the image of honeydew melon sample are obtained, the diffusing transmission spectrum includes whole
Melon spectrum and pulp spectrum.
The pulp spectrum is that whole melon removes the spectrum unless being obtained after edible tissue, and whole melon goes the method unless edible tissue
It is as follows：
The pericarp of honeydew melon sample is equably pruned using spiral peeling device；Below carpopodium position, push up certainly to close to horizontal
Footpath maximum cuts a drift angle and is 60 ° of vee, and takes out intraluminal tissue；Again by cutout installation in situ.Using upper
State the honeydew melon sample after method processing and remove the pulp unless being obtained after edible tissue for whole melon.
Step S2：The image of diffusing transmission spectrum and the honeydew melon sample to the honeydew melon sample pre-processes respectively.
Specifically include：
The diffusing transmission spectrum to the honeydew melon sample pre-processes, including noise elimination, background deduction, Norris
Filtering, multiplicative scatter correction, wavelet transformation and data normalization.
The image to the honeydew melon sample pre-processes, including morphology denoising, adaptive threshold fuzziness and right
Described image is marked, and is specially：
The image of honeydew melon sample is converted into bianry image, expanded for honeydew melon feature of image by burn into, opening operation etc.
After morphology denoising, bianry image is filled；Bianry image after filling is entered using the method for adaptive threshold fuzziness
Row segmentation；The centre of form of honeydew melon sample in image is determined, the image after segmentation is marked on the basis of the centre of form.Above-mentioned filling
Purpose is to be divided according to gray level to pixel set, and each subset so obtained can form one and real scape
The corresponding region of thing, each region inside have consistent attribute.
Step S3：Water stain the weld defect control spectral information and soluble solid content extracted in the pulp spectrum is sensitive
Spectral information.
Specifically include：
Step S301：Obtain the pulp spectrum of the honeydew melon of water content within a preset range and the honey in the honeydew melon sample
Pulp spectrum of the water content beyond the honeydew melon of the preset range in melon sample.
Step S302：Contrast the pulp spectrum of the honeydew melon of water content within a preset range and the honey in the honeydew melon sample
Water content is being preset beyond the pulp spectrum of the honeydew melon of the preset range according to water content in the honeydew melon sample in melon sample
In the range of honeydew melon pulp spectrum and the honeydew melon sample in water content beyond the honeydew melon of the preset range pulp spectrum
Between difference characteristic, determine the water stain weld defect control wave band in the pulp spectrum.
Step S303：First differential processing, least square fitting processing, multiplicative scatter correction are carried out to the pulp spectrum
After filtering, soluble solid content sensitive band in the pulp spectrum is determined.
Step S304：According to soluble solid in the water stain weld defect control wave band in the pulp spectrum and the pulp spectrum
Shape thing content sensitive band, water stain the weld defect control spectral information and soluble solid content extracted in the pulp spectrum are quick
Photosensitive spectrum information.
Step S4：Water stain the weld defect control spectral information and soluble solid content extracted in the whole melon spectrum is sensitive
Spectral information.
Specifically include：
Step S401：Obtain the pulp spectrum of honeydew melon of the water content in the preset range and institute in the honeydew melon sample
State the whole melon spectrum that water content in honeydew melon sample exceeds the honeydew melon of the preset range.
Step S402：Contrast the whole melon spectrum of the honeydew melon of water content within a preset range and the honey in the honeydew melon sample
Water content is being preset beyond the whole melon spectrum of the honeydew melon of the preset range according to water content in the honeydew melon sample in melon sample
In the range of honeydew melon whole melon spectrum and the honeydew melon sample in water content beyond the honeydew melon of the preset range whole melon spectrum
Between difference characteristic, determine the water stain weld defect control wave band in the whole melon spectrum.
Step S403：First differential processing, least square fitting processing, polynary scattering school are carried out to the whole melon spectrum
Just and after filtering, soluble solid content sensitive band in the whole melon spectrum is determined.
Step S404：According to soluble solid in the water stain weld defect control wave band in the whole melon spectrum and the whole melon spectrum
Shape thing content sensitive band, water stain the weld defect control spectral information and soluble solid content extracted in the whole melon spectrum are quick
Photosensitive spectrum information.
Step S5：Establish spectrum transformation model.The spectrum transformation model is the pulp light obtained according to step S3
Water stain weld defect control spectral information and soluble solid content sensitivity spectrum information and step S4 in spectrum obtain described
What water stain weld defect control spectral information and soluble solid content sensitivity spectrum information in whole melon spectrum were established, the spectrum
Transformation model is used to whole melon spectrum being converted to pulp spectrum.
Specifically include：
Step S501：Water stain weld defect control spectral information and soluble solid content in the pulp spectrum is sensitive
Spectral information is respectively converted into matrix form.
Step S502：Water stain weld defect control spectral information and soluble solid content in the whole melon spectrum is sensitive
Spectral information is respectively converted into matrix form.
Step S503：The matrix of the water stain weld defect control spectral information in the whole melon spectrum is determined using spectra methods
Difference characteristic between the matrix form of water stain weld defect control spectral information in form and the pulp spectrum, is established described whole
Turn between the water stain weld defect control spectral information in water stain weld defect control spectral information and the pulp spectrum in melon spectrum
Mold changing type.
Step S504：Determine that the soluble solid content sensitivity spectrum in the whole melon spectrum is believed using spectra methods
Difference between the matrix form of soluble solid content sensitivity spectrum information in the matrix form of breath and the pulp spectrum
Different feature, the soluble solid content sensitivity spectrum information established in the whole melon spectrum and the solubility of the pulp spectrum
Transformation model between solid content sensitivity spectrum information.
Step S6：Extract the color characteristic of the image of the honeydew melon sample.The color characteristic includes being used to react described
The water stain defect color characteristic of the water content of honeydew melon sample and the soluble solid of the sugar content for reacting the honeydew melon sample
Thing content color characteristic.The color characteristic can be extracted using neural network algorithm, in the present embodiment, using BP nerves
Water stain the defect color characteristic and soluble solid content color characteristic of the image of honeydew melon sample described in network extraction.
Step S7：According to the image of the honeydew melon sample, the volume and fruit shape index of the honeydew melon sample are calculated.
Specifically include：
Step S701：Calculate the volume of the honeydew melon sample.Because the image approximate of honeydew melon sample is oval, for the ease of meter
Calculate, it is assumed that preferable honeydew melon two dimensional image is symmetrical on vertical footpath, and using layering Integral Thought, (pixel volume of preferable honeydew melon can be seen
Make a series of integration of round platform volumes in vertical footpath direction) honeydew melon volume is calculated, be specially
Assuming that preferable honeydew melon is on indulging the symmetrical rotary body in footpath, Fig. 2 is the volume of honeydew melon sample of the embodiment of the present invention
Figure is calculated, wherein Fig. 2 (a) is the profile diagram of honeydew melon sample, and Fig. 2 (b) is round platform figure, and referring to Fig. 2 (a), L is the vertical of honeydew melon sample
Footpath, and be vertical direction, A is summit, and straight line CD, EF are vertical with vertical footpath, and the left intersection point with profile is 2 points of C, E, with profile
Right intersection point is 2 points of D, F, and the distance between CD and EF are Δ h,When Δ h gradually becomes very little,
By EF, CD enclose part can approximation regard round platform as, as shown in Fig. 2 (b), the volume of round platform is
The volume of honeydew melon sample is considered as a series of integration of round platforms on vertical footpath direction L, therefore, the body of honeydew melon sample
Product is
V=∫ V (h) dh
Step S702：Calculate the fruit shape index of the honeydew melon sample.Because the image approximate of honeydew melon sample is oval, therefore,
The profile of honeydew melon sample can be obtained by G passage blocking cut-off process, the marginal information of honeydew melon sample is extracted, is found according to marginal information
The centre of form of honeydew melon, the fruit shape index of honeydew melon is then calculated using software slide calliper rule method, be specially
The longitudinal and transverse demeter that honeydew melon is carried out using Minimum Enclosing Rectangle method is detected, and the image of honeydew melon sample is pre-processed, and is selected
RGB color is taken, extracts G component images；Bianry image is converted images into, bianry image is carried out using canny operators
Rim detection, the edge of the image of honeydew melon sample is extracted, by calculating the minimum enclosed rectangle for including the edge, draw honeydew melon
Vertical footpath DmaxWith transverse diameter Dmin, and then calculate honeydew melon fruit shape index D=Dmax/Dmin。
Step S8：Establish honeydew melon inside quality grade on-line checking model.The honeydew melon inside quality grade on-line checking
Model is contained according to the water stain weld defect control spectral information in the pulp spectrum, the soluble solid in the pulp spectrum
Measure sensitivity spectrum information, the water stain defect color characteristic, the soluble solid content color characteristic, the volume and institute
Fruit shape index foundation is stated, is specifically included：
Step S801：It is special according to the water stain weld defect control spectral information in the pulp spectrum, the water stain defect color
Sign, the volume and the fruit shape index, establish the water stain defect Nondestructive Testing Model of honeydew melon, are specially
By the water stain weld defect control spectral information in the pulp spectrum, the water stain defect color characteristic, the volume
Mutually tied with genetic algorithm using simulated annealing algorithm as output quantity as input quantity, water stain defect with the fruit shape index
The Genetic Simulated Annealing Algorithm of conjunction establishes the water stain defect Nondestructive Testing Model of honeydew melon, the calculating step of the Genetic Simulated Annealing Algorithm
It is rapid as follows
1) initialize：The initially initial temperature of time genetic algorithm genetic algebra, mutation probability, population invariable number and simulated annealing
Iterations etc. at degree, final temperature and each temperature.
2) initial population is randomly generated：Input quantity and output quantity are encoded.
3) bring each individual in population into calculation procedure, the variate-value of input quantity and output quantity is calculated.
4) bring fitness function into, evaluate the fitness of each population.
5) individual is intersected, mutation operation.
6) selection operation is carried out to individual using simulated annealing：
A. all individuals obtained to step 5) bring object function into, calculate desired value；
B. according to neighborhood function, each individual produces new explanation, and calculates the desired value of new explanation；
C. fresh target value and former desired value are made the difference to obtain Δ E, if Δ E<0, then receive new explanation, conversely, judge T (i) with
Critical-temperature TsetBetween relation, if T (i)>Tset, a value c is randomly generated between 0~1, if e-ΔE/kT(i)＞ c,
Then receive new explanation, conversely, retaining former solution, if T (i)<Tset, then former solution is directly retained；
D. judge whether to reach maximum iteration at this temperature, if reaching, perform annealing operation and enter step 7),
Conversely, into step b).
7) judge whether to reach maximum algebraically, terminate if it is satisfied, then calculating, be unsatisfactory for, then obtained step 6)
Body re-starts step 3) operation.
Step S802：According to soluble solid content sensitivity spectrum information, the solubility in the pulp spectrum
Solid content color characteristic, the volume and the fruit shape index, establish honeydew melon soluble solid content Non-Destructive Testing mould
Type, it is specially
By the soluble solid content sensitivity spectrum information in the pulp spectrum, the soluble solid content face
Color characteristic, the volume and the fruit shape index are moved back as input quantity, soluble solid content as output quantity using simulation
Pyrogenic process algorithm establishes honeydew melon soluble solid content Non-Destructive Testing mould with the Genetic Simulated Annealing Algorithm that genetic algorithm is combined
Type, the calculation procedure of the Genetic Simulated Annealing Algorithm are as described above.
Step S803：According to the water stain defect Nondestructive Testing Model of the honeydew melon and the honeydew melon soluble solid content without
Detection model is damaged, establishes honeydew melon inside quality grade on-line checking model.
Step S9：Tested honeydew melon is detected using the honeydew melon inside quality grade on-line checking model, obtain by
Survey the inside quality grade of honeydew melon.
Specifically include：
Step S901：Tested honeydew melon is lain in a horizontal plane on the shackle of conveying device, at the uniform velocity steadily travelled forward；
Step S902：When tested honeydew melon reaches diffusing transmission spectra collection room, the indoor photoelectric sensor output put of collection
One tested honeydew melon, which reaches, feeds back signal to signaling control unit, and command triggers spectrometer, spectrum are sent by signaling control unit
Instrument starts to gather the spectral information of tested honeydew melon, is tested honeydew melon by rear, photoelectric sensor exports a tested honeydew melon and left instead
Feedback signal sends order by signaling control unit and stops spectrometer work to signaling control unit；
Step S903：When tested honeydew melon reaches honeydew melon IMAQ room, the indoor photoelectric sensor output one put of collection
Individual tested honeydew melon, which reaches, feeds back signal to signaling control unit, and two industrial cameras of command triggers are sent by signaling control unit,
Two industrial cameras gather the direct picture information and side image information of tested honeydew melon respectively, are tested honeydew melon by rear, photoelectricity
Sensor one tested honeydew melon of output, which leaves, feeds back signal to signaling control unit, is stopped by signaling control unit control industrial camera
Only work；
Step S904：The spectrum collected and image information are input to the honeydew melon inside quality grade on-line checking mould
In type, the inside quality of tested honeydew melon is obtained, and then judges and divides honeydew melon grade.
Honeydew melon inside quality online test method in the present embodiment, realizes using pulp as detection object, eliminates
The influence of pericarp and fruit caving Nei Fei edible tissues to testing result, improves accuracy of detection, and by spectral information and surface
It is combined and is detected, further increases accuracy of detection.
Present invention also offers a kind of honeydew melon inside quality on-line detecting system 30, including：
Data obtaining module 301, it is described unrestrained saturating for obtaining the diffusing transmission spectrum of honeydew melon sample and the image of honeydew melon sample
Penetrating spectrum includes whole melon spectrum and pulp spectrum, and the pulp spectrum is that whole melon removes the spectrum unless being obtained after edible tissue.
Pretreatment module 302, the pretreatment module 302 specifically include：
Pretreated spectra unit, it is described to the honey for being pre-processed to the diffusing transmission spectrum of the honeydew melon sample
The diffusing transmission spectrum of melon sample is pre-processed, including noise elimination, background deduction, Norris filtering, multiplicative scatter correction, small
Wave conversion and data normalization.
Image pre-processing unit, it is described to the honeydew melon sample for being pre-processed to the image of the honeydew melon sample
Image pre-processed, including morphology denoising, adaptive threshold fuzziness and described image is marked.
Pulp spectrum sensitive information extraction modules 303, for extracting the water stain weld defect control spectrum in the pulp spectrum
Information and soluble solid content sensitivity spectrum information.
The pulp spectrum sensitive information extraction modules 303, are specifically included：
Pulp spectrum acquiring unit, for obtaining the pulp of the honeydew melon of water content within a preset range in the honeydew melon sample
Pulp spectrum of the water content beyond the honeydew melon of the preset range in spectrum and the honeydew melon sample.
First sensitive band determining unit, for contrasting the honeydew melon of water content within a preset range in the honeydew melon sample
Pulp spectrum of the pulp spectrum with water content in the honeydew melon sample beyond the honeydew melon of the preset range, according to the honeydew melon sample
The pulp spectrum of the honeydew melon of water content within a preset range exceeds the preset range with water content in the honeydew melon sample in product
Honeydew melon pulp spectrum between difference characteristic, determine the water stain weld defect control wave band in the pulp spectrum.
Second sensitive band determining unit, for carrying out first differential processing, least square fitting to the pulp spectrum
After processing, multiplicative scatter correction and filtering, soluble solid content sensitive band in the pulp spectrum is determined.
Pulp spectrum sensitive information extraction unit, for according to the water stain weld defect control wave band in the pulp spectrum and institute
Soluble solid content sensitive band in pulp spectrum is stated, extracts the water stain weld defect control spectral information in the pulp spectrum
With soluble solid content sensitivity spectrum information.
Whole melon spectrum sensitive information extraction modules 304, for extracting the water stain weld defect control spectrum in the whole melon spectrum
Information and soluble solid content sensitivity spectrum information.
The whole melon spectrum sensitive information extraction modules 304, are specifically included：
Whole melon spectrum acquiring unit, for obtaining honeydew melon of the water content in the preset range in the honeydew melon sample
Whole melon spectrum of the water content beyond the honeydew melon of the preset range in pulp spectrum and the honeydew melon sample.
3rd sensitive band determining unit, for contrasting the honeydew melon of water content within a preset range in the honeydew melon sample
Whole melon spectrum of the whole melon spectrum with water content in the honeydew melon sample beyond the honeydew melon of the preset range, according to the honeydew melon sample
The whole melon spectrum of the honeydew melon of water content within a preset range exceeds the preset range with water content in the honeydew melon sample in product
Honeydew melon whole melon spectrum between difference characteristic, determine the water stain weld defect control wave band in the whole melon spectrum.
4th sensitive band determining unit, for carrying out first differential processing, least square fitting to the whole melon spectrum
After processing, multiplicative scatter correction and filtering, soluble solid content sensitive band in the whole melon spectrum is determined.
Whole melon spectrum sensitive information extraction unit, for according to the water stain weld defect control wave band in the whole melon spectrum and institute
Soluble solid content sensitive band in whole melon spectrum is stated, extracts the water stain weld defect control spectral information in the whole melon spectrum
With soluble solid content sensitivity spectrum information.
Transformation model establishes module 305, for according to the water stain weld defect control spectral information in the pulp spectrum and can
Dissolubility solid content sensitivity spectrum information and water stain weld defect control spectral information in the whole melon spectrum and solubility are consolidated
Shape thing content sensitivity spectrum information, establishes spectrum transformation model, and the spectrum transformation model is used to whole melon spectrum being converted to fruit
Meat spectrum.
The transformation model establishes module 305, specifically includes：
Pulp spectral information converting unit, for by the water stain weld defect control spectral information in the pulp spectrum and solvable
Property solid content sensitivity spectrum information is respectively converted into matrix form.
Whole melon spectral information converting unit, for by the water stain weld defect control spectral information in the whole melon spectrum and solvable
Property solid content sensitivity spectrum information is respectively converted into matrix form.
First transformation model establishes unit, for determining that the water stain defect in the whole melon spectrum is quick using spectra methods
Difference between the matrix form of water stain weld defect control spectral information in the matrix form of photosensitive spectrum information and the pulp spectrum
Different feature, establish the water stain weld defect control spectral information in the whole melon spectrum and the water stain weld defect control in the pulp spectrum
Transformation model between spectral information.
Second transformation model establishes unit, for determining the soluble solid in the whole melon spectrum using spectra methods
The matrix form of thing content sensitivity spectrum information and the soluble solid content sensitivity spectrum information in the pulp spectrum
Difference characteristic between matrix form, the soluble solid content sensitivity spectrum information established in the whole melon spectrum with it is described
Transformation model between the soluble solid content sensitivity spectrum information of pulp spectrum.
Color feature extracted module 306, the color characteristic of the image for extracting the honeydew melon sample, the color characteristic
Water stain defect color characteristic including the water content for reacting the honeydew melon sample and for reacting containing for the honeydew melon sample
The soluble solid content color characteristic of sugar amount.
Volume and fruit shape index computing module 307, for the image according to the honeydew melon sample, calculate the honeydew melon sample
Volume and fruit shape index.
Detection model establishes module 308, for according to the water stain weld defect control spectral information in the pulp spectrum, described
Soluble solid content sensitivity spectrum information, the water stain defect color characteristic, the soluble solid in pulp spectrum
Thing content color characteristic, the volume and the fruit shape index, establish honeydew melon inside quality grade on-line checking model.
The detection model establishes module 308, specifically includes：
First detection model establishes unit, for according to the water stain weld defect control spectral information in the pulp spectrum, institute
Water stain defect color characteristic, the volume and the fruit shape index are stated, establishes the water stain defect Nondestructive Testing Model of honeydew melon.
Second detection model establishes unit, for according to the soluble solid content sensitivity spectrum in the pulp spectrum
Information, the soluble solid content color characteristic, the volume and the fruit shape index, establish honeydew melon soluble solid
Content Nondestructive Testing Model.
3rd detection model establishes unit, for can according to the water stain defect Nondestructive Testing Model of the honeydew melon and the honeydew melon
Dissolubility solid content Nondestructive Testing Model, establish honeydew melon inside quality grade on-line checking model.
Detection module 309, for being examined using the honeydew melon inside quality grade on-line checking model to tested honeydew melon
Survey, obtain the inside quality grade of tested honeydew melon.
Honeydew melon inside quality on-line detecting system in the present embodiment, realizes using pulp as detection object, eliminates
The influence of pericarp and fruit caving Nei Fei edible tissues to testing result, improves accuracy of detection, and by spectral information and surface
It is combined and is detected, further increases accuracy of detection.
In this specification for system disclosed in embodiment, because it is corresponded to the method disclosed in Example, institute
With the fairly simple of description, related part is referring to method part illustration.
Specific case used herein is set forth to the principle and embodiment of the present invention, and above example is said
It is bright to be only intended to help the method and its core concept for understanding the present invention；Meanwhile for those of ordinary skill in the art, foundation
The thought of the present invention, in specific embodiments and applications there will be changes.In summary, this specification content is not
It is interpreted as limitation of the present invention.
Claims (10)
- A kind of 1. honeydew melon inside quality online test method, it is characterised in that including：The diffusing transmission spectrum of honeydew melon sample and the image of honeydew melon sample are obtained, the diffusing transmission spectrum includes whole melon spectrum and pulp Spectrum, the pulp spectrum are that whole melon removes the spectrum unless being obtained after edible tissue；Extract water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in the pulp spectrum；Extract water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in the whole melon spectrum；According to the water stain weld defect control spectral information in the pulp spectrum and soluble solid content sensitivity spectrum information with And water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in the whole melon spectrum, establish spectrum Transformation model, the spectrum transformation model are used to whole melon spectrum being converted to pulp spectrum；The color characteristic of the image of the honeydew melon sample is extracted, the color characteristic includes being used to react containing for the honeydew melon sample The soluble solid content color of the water stain defect color characteristic of water and the sugar content for reacting the honeydew melon sample is special Sign；According to the image of the honeydew melon sample, the volume and fruit shape index of the honeydew melon sample are calculated；According to the water stain weld defect control spectral information in the pulp spectrum, the soluble solid content in the pulp spectrum Sensitivity spectrum information, the water stain defect color characteristic, the soluble solid content color characteristic, the volume and described Fruit shape index, establish honeydew melon inside quality grade on-line checking model；Tested honeydew melon is detected using the honeydew melon inside quality grade on-line checking model, obtains the inside of tested honeydew melon Quality grade.
- 2. a kind of honeydew melon inside quality online test method according to claim 1, it is characterised in that obtain honey described After the diffusing transmission spectrum of melon sample and the image of honeydew melon sample, in addition to：Diffusing transmission spectrum and institute to the honeydew melon sample The image for stating honeydew melon sample is pre-processed respectively；The diffusing transmission spectrum to the honeydew melon sample pre-processes, including noise elimination, background deduction, Norris filters Ripple, multiplicative scatter correction, wavelet transformation and data normalization；The image to the honeydew melon sample pre-processes, including morphology denoising, adaptive threshold fuzziness and to described Image is marked.
- 3. a kind of honeydew melon inside quality online test method according to claim 1, it is characterised in that described in the extraction Water stain weld defect control spectral information and soluble solid content sensitivity spectrum information in pulp spectrum, is specifically included：Obtain aqueous in the pulp spectrum of the honeydew melon of water content within a preset range and the honeydew melon sample in the honeydew melon sample Pulp spectrum of the amount beyond the honeydew melon of the preset range；Contrast in the honeydew melon sample pulp spectrum of the honeydew melon of water content within a preset range with it is aqueous in the honeydew melon sample Amount exceeds the pulp spectrum of the honeydew melon of the preset range, according to the honeydew melon of water content within a preset range in the honeydew melon sample Pulp spectrum and the honeydew melon sample in difference of the water content between the pulp spectrum of the honeydew melon of the preset range it is special Sign, determines the water stain weld defect control wave band in the pulp spectrum；After first differential processing, least square fitting processing, multiplicative scatter correction and filtering are carried out to the pulp spectrum, it is determined that Soluble solid content sensitive band in the pulp spectrum；It is sensitive according to soluble solid content in the water stain weld defect control wave band in the pulp spectrum and the pulp spectrum Wave band, extract water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in the pulp spectrum；Water stain weld defect control spectral information and soluble solid content sensitivity spectrum letter in the extraction whole melon spectrum Breath, is specifically included：Obtain in the honeydew melon sample in the pulp spectrum of honeydew melon of the water content in the preset range and the honeydew melon sample Whole melon spectrum of the water content beyond the honeydew melon of the preset range；Contrast in the honeydew melon sample the whole melon spectrum of the honeydew melon of water content within a preset range with it is aqueous in the honeydew melon sample Amount exceeds the whole melon spectrum of the honeydew melon of the preset range, according to the honeydew melon of water content within a preset range in the honeydew melon sample Whole melon spectrum and the honeydew melon sample in difference of the water content between the whole melon spectrum of the honeydew melon of the preset range it is special Sign, determines the water stain weld defect control wave band in the whole melon spectrum；After first differential processing, least square fitting processing, multiplicative scatter correction and filtering are carried out to the whole melon spectrum, really Soluble solid content sensitive band in the fixed whole melon spectrum；It is sensitive according to soluble solid content in the water stain weld defect control wave band in the whole melon spectrum and the whole melon spectrum Wave band, extract water stain weld defect control spectral information and soluble solid content the sensitivity spectrum information in the whole melon spectrum.
- 4. a kind of honeydew melon inside quality online test method according to claim 1, it is characterised in that described to establish spectrum Transformation model, specifically include：Water stain weld defect control spectral information in the pulp spectrum and soluble solid content sensitivity spectrum information are distinguished Be converted to matrix form；Water stain weld defect control spectral information in the whole melon spectrum and soluble solid content sensitivity spectrum information are distinguished Be converted to matrix form；The matrix form of the water stain weld defect control spectral information in the whole melon spectrum and the fruit are determined using spectra methods Difference characteristic between the matrix form of water stain weld defect control spectral information in meat spectrum, the water established in the whole melon spectrum The transformation model between water stain weld defect control spectral information in stain weld defect control spectral information and the pulp spectrum；The matrix form of the soluble solid content sensitivity spectrum information in the whole melon spectrum is determined using spectra methods Difference characteristic between the matrix form of the soluble solid content sensitivity spectrum information in the pulp spectrum, establishes institute State soluble solid content sensitivity spectrum information in whole melon spectrum and the soluble solid content of the pulp spectrum is quick Transformation model between photosensitive spectrum information.
- 5. a kind of honeydew melon inside quality online test method according to claim 1, it is characterised in that described to establish honeydew melon Inside quality grade on-line checking model, is specifically included：According to the water stain weld defect control spectral information in the pulp spectrum, the water stain defect color characteristic, the volume and The fruit shape index, establish the water stain defect Nondestructive Testing Model of honeydew melon；According to the soluble solid content sensitivity spectrum information in the pulp spectrum, the soluble solid content color Feature, the volume and the fruit shape index, establish honeydew melon soluble solid content Nondestructive Testing Model；According to the water stain defect Nondestructive Testing Model of the honeydew melon and the honeydew melon soluble solid content Nondestructive Testing Model, build Vertical honeydew melon inside quality grade on-line checking model.
- A kind of 6. honeydew melon inside quality on-line detecting system, it is characterised in that including：Data obtaining module, for obtaining the diffusing transmission spectrum of honeydew melon sample and the image of honeydew melon sample, the diffusing transmission spectrum Including whole melon spectrum and pulp spectrum, the pulp spectrum is that whole melon removes the spectrum unless being obtained after edible tissue；Pulp spectrum sensitive information extraction modules, for extracting the water stain weld defect control spectral information in the pulp spectrum and can Dissolubility solid content sensitivity spectrum information；Whole melon spectrum sensitive information extraction modules, for extracting the water stain weld defect control spectral information in the whole melon spectrum and can Dissolubility solid content sensitivity spectrum information；Transformation model establishes module, for according to the water stain weld defect control spectral information in the pulp spectrum and soluble solid Water stain weld defect control spectral information and soluble solid content in thing content sensitivity spectrum information and the whole melon spectrum Sensitivity spectrum information, establishes spectrum transformation model, and the spectrum transformation model is used to whole melon spectrum being converted to pulp spectrum；Color feature extracted module, the color characteristic of the image for extracting the honeydew melon sample, the color characteristic include using In the water stain defect color characteristic of the water content of reacting the honeydew melon sample and sugar content for reacting the honeydew melon sample Soluble solid content color characteristic；Volume and fruit shape index computing module, for the image according to the honeydew melon sample, calculate the volume of the honeydew melon sample And fruit shape index；Detection model establishes module, for according to the water stain weld defect control spectral information in the pulp spectrum, the pulp light Soluble solid content sensitivity spectrum information, the water stain defect color characteristic in spectrum, the soluble solid content Color characteristic, the volume and the fruit shape index, establish honeydew melon inside quality grade on-line checking model；Detection module, for being detected using the honeydew melon inside quality grade on-line checking model to tested honeydew melon, obtain The inside quality grade of tested honeydew melon.
- 7. a kind of honeydew melon inside quality on-line detecting system according to claim 6, it is characterised in that also include pretreatment Module, the pretreatment module specifically include：Pretreated spectra unit, it is described to the honeydew melon sample for being pre-processed to the diffusing transmission spectrum of the honeydew melon sample The diffusing transmission spectrum of product is pre-processed, including noise elimination, background deduction, Norris filtering, multiplicative scatter correction, small echo change Change and data normalization；Image pre-processing unit, for being pre-processed to the image of the honeydew melon sample, the figure to the honeydew melon sample As being pre-processed, including morphology denoising, adaptive threshold fuzziness and described image is marked.
- A kind of 8. honeydew melon inside quality on-line detecting system according to claim 6, it is characterised in that the pulp spectrum Sensitive information extraction module, is specifically included：Pulp spectrum acquiring unit, for obtaining the pulp spectrum of the honeydew melon of water content within a preset range in the honeydew melon sample Pulp spectrum with water content in the honeydew melon sample beyond the honeydew melon of the preset range；First sensitive band determining unit, for contrasting the pulp of the honeydew melon of water content within a preset range in the honeydew melon sample Pulp spectrum of the spectrum with water content in the honeydew melon sample beyond the honeydew melon of the preset range, according in the honeydew melon sample The pulp spectrum of the honeydew melon of water content within a preset range exceeds the honey of the preset range with water content in the honeydew melon sample Difference characteristic between the pulp spectrum of melon, determine the water stain weld defect control wave band in the pulp spectrum；Second sensitive band determining unit, for the pulp spectrum is carried out first differential processing, least square fitting processing, After multiplicative scatter correction and filtering, soluble solid content sensitive band in the pulp spectrum is determined；Pulp spectrum sensitive information extraction unit, for according to the water stain weld defect control wave band in the pulp spectrum and the fruit Soluble solid content sensitive band in meat spectrum, extracts water stain weld defect control spectral information in the pulp spectrum and can Dissolubility solid content sensitivity spectrum information；The whole melon spectrum sensitive information extraction modules, are specifically included：Whole melon spectrum acquiring unit, for obtaining the pulp of honeydew melon of the water content in the preset range in the honeydew melon sample Whole melon spectrum of the water content beyond the honeydew melon of the preset range in spectrum and the honeydew melon sample；3rd sensitive band determining unit, for contrasting the whole melon of the honeydew melon of water content within a preset range in the honeydew melon sample Whole melon spectrum of the spectrum with water content in the honeydew melon sample beyond the honeydew melon of the preset range, according in the honeydew melon sample The whole melon spectrum of the honeydew melon of water content within a preset range exceeds the honey of the preset range with water content in the honeydew melon sample Difference characteristic between the whole melon spectrum of melon, determine the water stain weld defect control wave band in the whole melon spectrum；4th sensitive band determining unit, for the whole melon spectrum is carried out first differential processing, least square fitting processing, After multiplicative scatter correction and filtering, soluble solid content sensitive band in the whole melon spectrum is determined；Whole melon spectrum sensitive information extraction unit, for according to the water stain weld defect control wave band in the whole melon spectrum and described whole Soluble solid content sensitive band in melon spectrum, extracts the water stain weld defect control spectral information in the whole melon spectrum and can Dissolubility solid content sensitivity spectrum information.
- A kind of 9. honeydew melon inside quality on-line detecting system according to claim 6, it is characterised in that the transformation model Module is established, is specifically included：Pulp spectral information converting unit, for the water stain weld defect control spectral information in the pulp spectrum and solubility to be consolidated Shape thing content sensitivity spectrum information is respectively converted into matrix form；Whole melon spectral information converting unit, for the water stain weld defect control spectral information in the whole melon spectrum and solubility to be consolidated Shape thing content sensitivity spectrum information is respectively converted into matrix form；First transformation model establishes unit, for determining the water stain weld defect control light in the whole melon spectrum using spectra methods Difference between the matrix form of water stain weld defect control spectral information in the matrix form of spectrum information and the pulp spectrum is special Sign, establish the water stain weld defect control spectral information in the whole melon spectrum and the water stain weld defect control spectrum in the pulp spectrum Transformation model between information；Second transformation model establishes unit, for determining that the soluble solid in the whole melon spectrum contains using spectra methods Measure the matrix form and the matrix of the soluble solid content sensitivity spectrum information in the pulp spectrum of sensitivity spectrum information Difference characteristic between form, the soluble solid content sensitivity spectrum information established in the whole melon spectrum and the pulp Transformation model between the soluble solid content sensitivity spectrum information of spectrum.
- A kind of 10. honeydew melon inside quality on-line detecting system according to claim 6, it is characterised in that the detection mould Type establishes module, specifically includes：First detection model establishes unit, for according to water stain weld defect control spectral information, the water in the pulp spectrum Stain defect color characteristic, the volume and the fruit shape index, establish the water stain defect Nondestructive Testing Model of honeydew melon；Second detection model establishes unit, for according to the soluble solid content sensitivity spectrum letter in the pulp spectrum Breath, the soluble solid content color characteristic, the volume and the fruit shape index, establish honeydew melon soluble solid and contain Measure Nondestructive Testing Model；3rd detection model establishes unit, for soluble according to the water stain defect Nondestructive Testing Model of the honeydew melon and the honeydew melon Solid content Nondestructive Testing Model, establish honeydew melon inside quality grade on-line checking model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711041446.9A CN107860722B (en) | 2017-10-30 | 2017-10-30 | Method and system for online detection of internal quality of honeydew melons |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711041446.9A CN107860722B (en) | 2017-10-30 | 2017-10-30 | Method and system for online detection of internal quality of honeydew melons |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107860722A true CN107860722A (en) | 2018-03-30 |
| CN107860722B CN107860722B (en) | 2020-04-21 |
Family
ID=61696949
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711041446.9A Active CN107860722B (en) | 2017-10-30 | 2017-10-30 | Method and system for online detection of internal quality of honeydew melons |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107860722B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110017778A (en) * | 2019-04-25 | 2019-07-16 | 广州富港万嘉智能科技有限公司 | The dimension measurement method and peeling method of melon and fruit, corresponding device and storage medium |
| CN110389104A (en) * | 2019-08-01 | 2019-10-29 | 南京林业大学 | The fruit quality detection method that pericarp influences is eliminated based on spectral differences subassembly |
| CN111999255A (en) * | 2020-08-13 | 2020-11-27 | 中冶建筑研究总院（深圳）有限公司 | Marine environment steel structure coating detection method and device, server and storage medium |
| WO2021089001A1 (en) * | 2019-11-08 | 2021-05-14 | 宁波诺丁汉大学 | Method, device, and system for detecting sweetness of fruit, and storage medium |
| CN113390833A (en) * | 2021-06-23 | 2021-09-14 | 长江师范学院 | Crisp plum maturity judging method based on multiple detection modes |
| CN113390834A (en) * | 2021-06-23 | 2021-09-14 | 长江师范学院 | Crisp plum maturity detection method based on visual identification |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS61200407A (en) * | 1985-03-01 | 1986-09-05 | Hitachi Ltd | Fourier transformation type infrared film thickness measuring apparatus |
| CN201519152U (en) * | 2009-09-25 | 2010-07-07 | 深圳先进技术研究院 | Noninvasive blood sugar detecting system |
| CN103792235A (en) * | 2014-01-10 | 2014-05-14 | 内蒙古农业大学 | Diffuse transmission spectrum and image information fusion method for detecting internal quality of honeydew melons on line and device |
| CN107179283A (en) * | 2017-07-02 | 2017-09-19 | 天津大学 | A kind of method of the packed complicated solution component content of nondestructive measurement |
-
2017
- 2017-10-30 CN CN201711041446.9A patent/CN107860722B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS61200407A (en) * | 1985-03-01 | 1986-09-05 | Hitachi Ltd | Fourier transformation type infrared film thickness measuring apparatus |
| CN201519152U (en) * | 2009-09-25 | 2010-07-07 | 深圳先进技术研究院 | Noninvasive blood sugar detecting system |
| CN103792235A (en) * | 2014-01-10 | 2014-05-14 | 内蒙古农业大学 | Diffuse transmission spectrum and image information fusion method for detecting internal quality of honeydew melons on line and device |
| CN107179283A (en) * | 2017-07-02 | 2017-09-19 | 天津大学 | A kind of method of the packed complicated solution component content of nondestructive measurement |
Non-Patent Citations (6)
| Title |
|---|
| 史波林等: "应用GA-DOSC算法消除果皮影响近红外漫反射光谱分析苹果硬度的研究", 《光谱学与光谱分析》 * |
| 周水洪等: "《易腐食品冷链百科全书》", 31 December 2009, 东华大学出版社 * |
| 孔祥生等: "《植物生理学实验技术》", 29 February 2008, 中国农业出版社 * |
| 张德虎: "河套蜜瓜品质可见近红外光谱检测研究", 《中国优秀硕士学位论文全文数据库（电子期刊）》 * |
| 田海清等: "厚皮甜瓜无损检测方法的研究现状及发展趋势", 《农机化研究》 * |
| 谭峰等: "《基于光谱技术的寒地水稻稻瘟病害分析及机理研究》", 31 August 2016, 哈尔滨工程大学出版社 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110017778A (en) * | 2019-04-25 | 2019-07-16 | 广州富港万嘉智能科技有限公司 | The dimension measurement method and peeling method of melon and fruit, corresponding device and storage medium |
| CN110017778B (en) * | 2019-04-25 | 2021-07-06 | 广州富港万嘉智能科技有限公司 | Melon and fruit size measuring method and peeling method, corresponding device and storage medium |
| CN110389104A (en) * | 2019-08-01 | 2019-10-29 | 南京林业大学 | The fruit quality detection method that pericarp influences is eliminated based on spectral differences subassembly |
| WO2021089001A1 (en) * | 2019-11-08 | 2021-05-14 | 宁波诺丁汉大学 | Method, device, and system for detecting sweetness of fruit, and storage medium |
| US11965842B2 (en) | 2019-11-08 | 2024-04-23 | University Of Nottingham Ningbo China | Method, device, and system for detecting sweetness of fruit, and storage medium |
| CN111999255A (en) * | 2020-08-13 | 2020-11-27 | 中冶建筑研究总院（深圳）有限公司 | Marine environment steel structure coating detection method and device, server and storage medium |
| CN111999255B (en) * | 2020-08-13 | 2024-01-19 | 中冶建筑研究总院（深圳）有限公司 | Marine environment steel structure coating detection method, device, server and storage medium |
| CN113390833A (en) * | 2021-06-23 | 2021-09-14 | 长江师范学院 | Crisp plum maturity judging method based on multiple detection modes |
| CN113390834A (en) * | 2021-06-23 | 2021-09-14 | 长江师范学院 | Crisp plum maturity detection method based on visual identification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107860722B (en) | 2020-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107860722A (en) | A kind of honeydew melon inside quality online test method and system | |
| Yan et al. | Review of indirect optical measurements of leaf area index: Recent advances, challenges, and perspectives | |
| CN110287932B (en) | Road blocking information extraction method based on deep learning image semantic segmentation | |
| CN103954542B (en) | Based on the PM2.5 concentration detectors that non-reference picture definition is evaluated | |
| WO2019174572A1 (en) | Method for identifying frostbite status of cereal seeds using embryo spectrogram characteristic wavebands | |
| Leblanc et al. | Hemispherical photography simulations with an architectural model to assess retrieval of leaf area index | |
| Liu et al. | A vision-based robust grape berry counting algorithm for fast calibration-free bunch weight estimation in the field | |
| CN109738137B (en) | Real-time earth-rock dam leakage monitoring and rapid diagnosis method based on image contrast | |
| Bienert et al. | A voxel-based technique to estimate the volume of trees from terrestrial laser scanner data | |
| CN105021529A (en) | Spectrum information and image information fusing crop plant disease and insect pest identifying and distinguishing method | |
| CN103900972B (en) | Multi-feature fusion-based meat freshness hyperspectral image visual detection | |
| CN103065149A (en) | Netted melon fruit phenotype extraction and quantization method | |
| CN110349153B (en) | Image method tobacco peduncle content detection method based on multi-feature fusion | |
| CN101944231A (en) | Method for extracting wheatear morphological parameters | |
| CN102855485B (en) | The automatic testing method of one grow wheat heading | |
| CN111141653B (en) | Tunnel leakage rate prediction method based on neural network | |
| CN108376403A (en) | Grid Colony hybridization dividing method based on hough-circle transform | |
| CN115936532B (en) | Saline-alkali soil stability assessment method and system based on BP neural network | |
| CN112215217B (en) | Digital image recognition method and device for simulating doctor to read film | |
| CN105004722A (en) | Method for rapidly detecting maturity of tobacco leaves | |
| CN107229910A (en) | A kind of remote sensing images icing lake detection method and its system | |
| CN114419311B (en) | Multi-source information-based passion fruit maturity nondestructive testing method and device | |
| CN117274251B (en) | Tablet quality detection method in medicine production process based on image data | |
| CN106022354A (en) | SVM-based image MTF measurement method | |
| CN112974303B (en) | Hyperspectrum-based fruit quality detection method, device and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |