CN101603927B - Device for nondestructive testing of defects of hoxiu pears and use method - Google Patents
Device for nondestructive testing of defects of hoxiu pears and use method Download PDFInfo
- Publication number
- CN101603927B CN101603927B CN2009101813041A CN200910181304A CN101603927B CN 101603927 B CN101603927 B CN 101603927B CN 2009101813041 A CN2009101813041 A CN 2009101813041A CN 200910181304 A CN200910181304 A CN 200910181304A CN 101603927 B CN101603927 B CN 101603927B
- Authority
- CN
- China
- Prior art keywords
- image
- sample
- camera
- pears
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention relates to a device for nondestructive testing of defects of hoxiu pears and a use method, which belong to the technical field of agricultural engineering. The device for the nondestructive testing of internal and external defects of the hoxiu pears is characterized in that the device comprises an illumination system, a transfer system, a detection system and an analytical judgment system; a camera and an image acquisition card acquire a fruit body image, and each sample is knocked by a knocking stick with a force of 15N for two times at both ends of the equator of a fruit body; and an audio signal and an image signal of the sample recorded by a sound level meter are converted and enter an industrial computer for image and acoustic analysis treatment after the fast Fourier transform to obtain a classification result of the sample. The device optimally utilizes computer vision information and vibration spectrum information to perform system identification and detection on the internal and external defects of the hoxiu pears, and has the advantages of quickness, accuracy, stability, broad identification range and the like. The device can be used for aspects of the hoxiu pears such as separation, control of processing process and the like, and is favorable for improving the sales added value of the hoxiu pears.
Description
Technical field
The present invention relates to a kind of device and usage of Non-Destructive Testing abundance of water pears defective, is a kind of method and apparatus based on computer vision and vibration frequency specturm analysis fusion Non-Destructive Testing abundance of water pears defective at the abundance of water pears, belongs to the agricultural engineering technology field.
Background technology
Raising along with people's living standard, the consumption conversion of the direction from the tuple amount to the heavy amount to fruit, how to change simultaneously the low present situation of agricultural products in China outlet added value, for peasant and fruit processing enterprise extra earning become fruit circulation in recent years, storage and detect maximum problem of paying close attention to.
The operatic circle belongs to the rose family, pears subfamily pear deciduous fruit tree.China is pear tree cultivated area maximum, country that output is the highest in the world.But because for a long time, China's the operatic circle cultivation technique falls behind, extensive management, and the operatic circle is adopted aftertreatment link weakness at present in addition, causes the operatic circle quality uneven, and its value and competitive power are greatly reduced, and is difficult to adapt to the sale situation of market fruit competition.One of main path of enhancing competitiveness is to improve product quality, and the important means that guarantees product quality to be exactly the Quality Detection technology of application of advanced carry out classification by its quality to fruit in the storage processing and the process of circulation.
Because the operatic circle is subjected to the restriction such as factors such as seed, region, weather, soil, nutrition, disease and pests in growth course, make gather and the postharvest storage process in be easy to generate various types of defectives, if these defectives can not identify in time, the sale of the operatic circle will be influenced greatly before the packing packing.
Computer vision technique is to utilize imageing sensor to obtain the image of object, and image is changed into data matrix, analyzes with computing machine, finishes simultaneously and the vision related task.The history in existing more than 40 year of computer vision, at present, computer vision technique mainly is confined to external sort for the detection of quality of agricultural product, and this wherein mainly concentrates on aspects such as size, shape, color, surface imperfection.
The acoustic characteristic of agricultural product is meant reflection characteristic, scattering properties, transmissison characteristic, absorption characteristic, attenuation coefficient and velocity of propagation and itself acoustic impedance and the natural frequency etc. of agricultural product under acoustic wave action, and they have reflected sound wave and the interactional basic law of agricultural product.Utilize acoustic characteristic that its quality is carried out Non-Destructive Testing, and can discern the inherent vice of agricultural product in conjunction with spectrum analysis preferably.
Multi-sensor information fusion is multistage, a multifaceted data handling procedure, mainly finish to the data from a plurality of information sources detect automatically, processing such as related, relevant, estimation and combination.In multiple heat transfer agent resource merged, may there be mutual crossover phenomenon in each heat transfer agent resource.It according to certain criterion to these information analyze, comprehensive and balance, in the hope of obtaining several best generalized variables of simplifying.Compare with single detection means, it have contain much information, robustness is good and with meet advantages such as human cognitive process.
It is fast to have a detection speed based on the detection technique of computer vision and vibration frequency specturm analysis, characteristics such as the result is accurate.Domestic in recent years researcher has done a large amount of work, but all is to utilize image information or sound spectrum information that the outside or the inherent vice of fruit are detected classification separately.Do not see research with the fusion Non-Destructive Testing abundance of water pears defective of computer vision technique and vibration frequency specturm analysis.
Summary of the invention
A kind of method and apparatus of the present invention at defective inside and outside the Multi-sensor Fusion Non-Destructive Testing of abundance of water pears, utilize computer vision and vibration frequency specturm analysis to gather the fusion treatment that relevant information is carried out two levels more respectively to the various defectives inside and outside the abundance of water pears, can comprehensively discern the inside and outside defective of abundance of water pears fast.
Technical scheme
1. the device of the inside and outside defective of Non-Destructive Testing abundance of water pears is characterized in that: comprise A illumination system, B transfer system, C detection system and D analysis judgment system, wherein:
The A illumination system
Illumination system is made up of four 25W halogen tungsten photoelectric tubes (7), four halogen tungsten photoelectric tubes (7) be installed in camera (1) around, apart from the camera Center Length is 10cm, the stiff end of camera and photoelectric tube posts the ater film, be used to eliminate the noise that reflected light causes, this stiff end is connected through the hinge on support (17);
The B transfer system
Transfer system is made up of travelling belt (10), pallet (9), collection chamber (12) and main stepper motor (13), and travelling belt (10) is by being installed in inner main stepper motor (13) control of apparatus housing (15), fixed tray on the travelling belt (9);
The C detection system
Detection system comprises image capture module and vibration frequency specturm analysis module, wherein image capture module is made of camera (1), image data acquiring card (2), halogen tungsten photoelectric tube (7), camera (1) and sound meter (3) are fixed on the support (17), camera (1) distance of camera lens pallet 25cm;
The vibration frequency specturm analysis module is made up of sound meter (3), Wave data capture card (4), stamp (8), auxilliary stepper motor (14), silencer pad (16), stamp (8) fixed rotating shaft is apart from pallet 15cm, stamp PVC material (8) length 8cm, diameter 5mm; Auxilliary stepper motor (14) is installed in the position of the inner close stamp (8) of casing (15), control stamp (8) is to the lasting hammer action of sample, knock at every turn and produce an effective sound wave signal, and quilt collection enters Wave data capture card (4), silencer pad (16) is fixed in 5cm place, stamp below, is used for eliminating the issuable mixed and disorderly acoustic signal that does not meet acquisition range of the process of knocking;
The D analysis judgment system
Analysis judgment system comprises traffic pilot (5), industrial computer (6) and classification determining device (11), wherein traffic pilot (5) is connected with signal wire respectively with the vibration frequency specturm analysis module with image capture module in the detection system, schedule on the support (17) with hinges fixing, its output signal transfers to industrial computer (6) with the USB line, traffic pilot (5) imports conversion back (analog-digital conversion) in the set of signals into industrial computer (6) and carries out analyzing and processing, obtains classification results after the analysis and judgement.
In the said apparatus, camera (1) is the TK-C1380CCD of JVC company, image pick-up card (2) is Canadian MatroxMeteror-II/Standard, sound meter (3) is HS5633A type sound meter (a Zhejiang red sound equipment factory), the spectrum signal analytic system adopts SD150 dynamic analysis system (ring Electronic Instrument, Limited produces in the Tianjin), it is Ling Hua EBC-1200MediaGX1 that industrial computer (6) is embedded in formula mainboard model, traffic pilot (5) closes good electric industry Products Co., Ltd for the Zhongshan city and produces, model is SH912A, the built-in NPLC-15 industrial control board of classification determining device (11).Said apparatus is used for the method for the inside and outside defective of Non-Destructive Testing abundance of water pears, it is characterized in that:
Sample abundance of water pears are placed on the pallet (9), and travelling belt (10) brings into operation, the travelling speed 0.2m/s of travelling belt (10), interval time 0.6s; Each sample is gathered piece image by camera (1), and reaches image pick-up card (2);
Stamp (8) knocks sample simultaneously, and when knocking and sample place tangent plane is 30 degree angles, and each sample knocks twice at two ends, fruit body equator, knocks the power that produces 15N at every turn;
Note sound signal by sound meter (3), reach Wave data capture card (4) by signal wire, import multiple signals converter (5) into by the USB line together with picture signal again and carry out conversion of signals, analog picture signal transfers data image signal to, analoging sound signal is converted into digital audio signal, and makes to enter at last after the fast Flourier FFT conversion data storage area on the industrial computer (6);
Control at the virtual instrument of industrial computer (6) and general image disposal system Image Sys3.0 (Beijing modern Fu Boke skill company limited) and spectrum signal analytic system to be housed on the platform (encircling Electronic Instrument, Limited in the Tianjin produces, model is SD150) carry out image and acoustic analysis processing, extract four characteristic parameters of sample: the defect area area A
S, gradation of image entropy ENT, characteristic peak frequency average μ
fAnd characteristic peak frequency coefficient of variation CV
f, substitution is suc as formula in 13 the formula,
Obtain the classification results of this sample.
Beneficial effect:
1. the method and apparatus based on computer vision and vibration frequency specturm analysis fusion Non-Destructive Testing abundance of water pears defective utilizes computer image analysis technology and vibration frequency specturm analysis technology to extract image information and rumble spectrum information, utilize sensor fusion (data Layer merges and the decision-making level merges) technology of two levels, make up the nondestructive inspection parameters that can better reflect the inside and outside defective of abundance of water pears the operatic circle, set up the comprehensive Non-Destructive Testing hierarchy system of abundance of water pears defective.Thereby the quick nondestructive that carries out the inside and outside defective of abundance of water pears detects.
2. based on the method and apparatus of computer vision and vibration frequency specturm analysis fusion Non-Destructive Testing abundance of water pears defective, the inside and outside defective of abundance of water pears is carried out system's row identification and detection, have advantages such as quick, accurate, that identification range is wide.The aspect such as sorting, process control that can be used for the abundance of water pears.Help improving it and sell added value.For the contactless fast detecting of abundance of water pears and classification grading provide high-level efficiency, the solution of high precision and high reliability.
Four, description of drawings
Fig. 1: general structure of the present invention and principle of device synoptic diagram
Among Fig. 1: chamber, (13) main stepper motor, (14) auxilliary stepper motor, (15) casing, (16) silencer pad, (17) frame are gathered in (1) camera, (2) image data acquiring card, (3) sound meter, (4) Wave data capture card, (5) traffic pilot, (6) industrial computer, (7) photoelectric tube, (8) stamp, (9) pallet, (10) travelling belt, (11) categorised collection device, (12)
Fig. 2: apparatus module figure
Fig. 3: sensor fusion model construction process of the present invention
Five, embodiment
Further describe the present invention below in conjunction with the drawings and specific embodiments.Mainly comprise model construction process, device design, and three parts of application example.Existing division is as follows:
(1) the abundance of water pears defect rank discrimination model building process of the present invention's design.As described below:
At sample abundance of water pears of the present invention.As shown in Figure 3, the present invention utilizes computer image analysis technology and vibration frequency specturm analysis technology to extract image information and rumble spectrum information respectively, utilize sensor fusion (data Layer merges and the decision-making level merges) technology of two levels, carry out the Non-Destructive Testing of the inside and outside defective of abundance of water pears.
There are various inside and outside defective abundance of water the operatic circle samples in selection.Gather abundance of water pears surface visible light picture, image is handled (self-adaptation is cut apart [Zhen Ziyang for Butterworth low-pass filtering, image sharpening; The kingly way ripple; Adapting to image dividing method [J] photoelectron of elongated degree particle swarm optimization such as Liu Wenbo fuzzy clustering. laser, 2009,20 (1): 100~102)]), color of image information (picture tone, the average of brightness and saturation degree and the standard deviation μ of extraction fault location
H, μ
S, μ
I, Std
H, Std
S, Std
I), [Pang Jiangwei is based on research [D] Zhejiang University of the navel orange surface common deficiency kind identification of computer vision for texture information (gradation of image entropy ENT and picture contrast CON); 47~55] and shape information (defect area area A
S).Sample is carried out vibration percussion test (each sample knocks twice, and two characteristic peaks are arranged), gather sound frequency-domain with the sound waveform capture card.The nondestructive inspection parameters that adopts fast fourier transform to carry out extracting after the Signal Pretreatment to the analysis of frequency-region signal comprises: characteristic peak frequency (f) the average μ of abundance of water pears
f, characteristic peak frequency coefficient of variation CV
f, and characteristic peak area A
f[Jiang Rui relates to the Wang Jun egg and knocks response characteristic and eggshell crack detection [J] agricultural mechanical journal 2005.36 (3): 75~78].
The Physical Quantity Calculation formula or the method that relate to wherein are as described below:
Picture tone (H), brightness (I) and saturation degree (S) 3~6 are calculated as follows by the RGB component of the image extraction of camera collection.Adopt self-compiling program to get a little, and add up its standard deviation and average (μ
H, μ
S, μ
I, Std
H, Std
S, Std
I).
(formula 4)
(formula 5)
(formula 6)
The defect area area:
I, j-refer to the number of the pixel of its inside (comprising on the border) that defect area splits.
Gradation of image entropy ENT and picture contrast CON calculate according to formula 8~9
Abs (i-j)=n wherein, (i j) is the gray level co-occurrence matrixes of image to P, and L is the number of greyscale levels of matrix.
Characteristic peak frequency coefficient of variation CV
fWith characteristic peak frequency average μ
f Pressing following formula 10~11 calculates:
μ
f=f
1+ f
2(formula 11)
Std wherein
fStandard deviation μ for the sample characteristics frequency
fMean value characteristic peak area A for the sample characteristics frequency
fCalculate by formula 12:
Pi-detects the highest frequency of the power spectrum amplitude n-detection of each frequency in the frequency range, this test n=8900Hz
After carrying out Non-Destructive Testing, above sample is carried out carrying out destructive hand inspection with reference to national standard (GB/T10650-1989), write down its inside and outside defect type and degree, and according to the GB classification.The Non-Destructive Testing characteristic parameter that extracts is carried out association analysis with conventional defect situation, the External Defect of abundance of water pears and the sensor fusion of conventional inherent vice are detected, adopt the data Layer fusion method fusion nondestructive inspection parameters of isolated component and support vector machine associating to detect, by convergence analysis [Zheng Junhua; ICA-SVM modeling method [J] information and the control of Wu Tiejun blast furnace molten iron silicon content forecast, 2008,37 (2) 247~250] the nondestructive inspection parameters collection is M={CV in this process of determining of back
f, A
S.The characteristic parameter that designs in this process is CV
f, A
SUnconventional inherent vice (mainly due to the defective from inside to outside of microorganisms) for the abundance of water pears.The decision-making level's fusion method that adopts evidence theory to merge merges nondestructive inspection parameters and detects, by convergence analysis [Di Lisi, application [J] the Beijing Institute of Technology journal of Pan Xu peak D-S evidence theory in data fusion, 1997,17 (2): 198] the nondestructive inspection parameters collection determined of back is N={ μ
f, ENT}.The characteristic parameter that designs in this process is MAX (μ
f, ENT), expression is got the higher value computing to the image entropy of characteristic frequency average and defective.
Detection discrimination model with above internal-external defective carries out comprehensively at last, by the leading weight (scale-up factor see formula 1) of clear and definite each model in the synthetic determination model of analyzing.Add at last and after set up the comprehensive nondestructive detection system model of abundance of water pears defective, represent with PMED.
PMED=0.312*CV
f+ 0.477A
S+ 0.211MAX (μ
f, ENT) (formula 1)
In conjunction with the grade of pears sample being judged, and carry out relatedly, divide according to national standard and to differentiate interval (GB/T10650-1989) with calculated value to the decision threshold PMED of sample.The design abundance of water pears defective comprehensively application platform of harmless hierarchy system is formula 2 described discrimination models.Wherein PMED is for merging the comprehensive threshold value of decision model.Need only the PMED value substitution formula 2 described discrimination models that to calculate during utilization.Can differentiate the grade of abundance of water pears.
The abundance of water pears defective comprehensively application platform of harmless hierarchy system is the described discrimination model of following formula.
(2) device design
The device design of the inside and outside defective of Non-Destructive Testing abundance of water pears comprises A illumination system, B transfer system, C detection system and four parts of D analysis judgment system.
The A illumination system
Illumination system is made up of four halogen tungsten photoelectric tubes (7) (25W, Philip).Four halogen tungsten photoelectric tubes (7) be installed in camera (1) (wherein camera with the TK-C1380CCD of JVC company) around, be used for making up the visible light environment that sample surfaces is gathered.Apart from the camera Center Length is 10cm.The stiff end of camera and photoelectric tube posts the ater film.Be used to eliminate the noise that reflected light causes.This stiff end is connected through the hinge on support (17).
The B transfer system
Transfer system is made up of travelling belt (by the customization of the fast plant equipment in Jinan company limited, low-carbon steel material) (10), pallet (9) (PVC material), collection chamber (12) and main stepper motor (model is GL8825A220-331) (13).Travelling belt (10) has been fixed 50 pallets (9) by being installed in inner main stepper motor (13) control of apparatus housing (15) on the travelling belt, each is 20cm at interval.The travelling speed of travelling belt and interval time are by the macro assembly language control of establishment voluntarily, and and the image capture module of detection system and the acquisition rate coupling (0.6s detects a sample, travelling belt travelling speed 0.2m/s) of vibration frequency specturm analysis module.Detect end back travelling belt sample is transported to off-sorting station, sort, enter behind the sample classification and gather chamber (12) by the instruction control classification determining device (11) (built-in NPLC-15 industrial control board) that obtains judged result.
The C detection system
Detection system comprises image capture module and vibration frequency specturm analysis module.Wherein camera (1) and sound meter (3) (sound level is counted HS5633A type sound meter) are fixed on the support (17).Camera (1) distance of camera lens pallet 25cm.Image capture module is made of camera (1), image data acquiring card (2) (image pick-up card is Canadian Matrox Meteror-II/Standard), halogen tungsten photoelectric tube (7), and and transfer system cooperating.Can accomplish to gather 1 width of cloth picture every 0.2s.The vibration frequency specturm analysis module is by sound meter (3), Wave data capture card (4) (SD150 type, encircle Electronic Instrument, Limited in the Tianjin), stamp (8) (PVC material, heavy 14g, long 8cm), auxilliary stepper motor (model is GL8825A40-287) (14), silencer pad (self-control, the asbestos material, thickness 6cm) (16) are formed, and and transfer system cooperating.Stamp (8) fixed rotating shaft is apart from pallet 15cm.In stamp (8) when work, be 30 degree angles with sample place tangent plane, knocks twice back and forth at two ends, fruit body equator respectively during work, and stamp in this process (8) spin 300 is spent.Whenever after carrying out keystroke action, reset to former attitude.Auxilliary stepper motor (14) is installed in the position of the inner close stamp (8) of casing (15), and control stamp (8) is to the lasting hammer action of sample.Can produce an effective sound wave signal every 0.2s, and be entered Wave data capture card (4) by collection.Silencer pad (16) is fixed in 5cm place, stamp below, is used for eliminating the issuable hash that does not meet acquisition range of the process of knocking.
The D analysis judgment system
Analysis judgment system comprises traffic pilot (5) (model is SH912A, and good electric industry Products Co., Ltd is closed by the Zhongshan city), industrial computer (6) (industrial computer embedded main board model Ling Hua EBC-1200MediaGX1) and classification determining device (11).Wherein image capture module in the detection system and vibration frequency specturm analysis module are connected to traffic pilot (5) with dedicated signal lines respectively.On support (17), its output signal transfers to industrial computer with the USB line to traffic pilot (5) with hinges fixing.Traffic pilot carries out analyzing and processing with importing industrial computer (6) into after changing in the set of signals.Industrial computer calls the comprehensive nondestructive detection system of having set up of abundance of water pears defective to carry out obtaining classification results after the analysis and judgement.The output result also sends instruction control classification determining device (11) and sorts, and enters behind the sample classification and gathers chamber (12).Industrial computer carries out data transmission with USB line and categorised collection.Classification determining device (11) is positioned at the travelling belt end.Gather chamber (12) and be positioned at two ends, the casing afterbody left and right sides, sample enters according to classification results that it is indoor.Industrial computer inside is equipped with the virtual instrument that designs and controls platform [Liu Junhua, Bai Peng, Jia Hui celery .Lab Windows/CVI virtual instrument programming language study course [M]. Beijing: Electronic Industry Press, 2001. chapter 7], can independently control two motors (13), the operation of (14) and rotating speed.The also work of two of control and detection system detection modules and stopping.After the model that the information via that finally collects is set up in advance carried out the computing judgement, the result who obtains also controlled on the platform at this and shows, reached classification determining device (11) simultaneously.
(3) application example
Introduce the once complete course of work of the present invention below in conjunction with accompanying drawing 1,2:
1) virtual instrument that at first starts on the industrial computer (6) is controlled platform, starts image capture module and vibration frequency specturm analysis module.Open four halogen tungsten photoelectric tubes (7) switch that is positioned at the casing inboard.Sample is put to pallet (9).After checking that tray position and detection module are in good condition.Start the virtual instrument that is positioned on the industrial computer (6) and control the motor control switch of platform.
2) travelling belt (10) brings into operation, and sample enters detection system, pause 1s.Gather piece image by camera (1), and reach image pick-up card (2) (average gather and transmission time 0.2s), while stamp (8) is 30 degree angles sample is carried out keystroke action, rotate up 300 degree backs and carry out keystroke action with same angle again, note sound signal (average deadline 0.4s) by sound meter (3) in another survey of sample.Reach Wave data capture card (4) by dedicated signal lines, importing multiple signals converter (5) into by the USB line together with picture signal again carries out conversion of signals (analog picture signal transfers data image signal to, analoging sound signal is converted into digital audio signal, and does the FFT conversion) after enter data storage area on the industrial computer (6) at last.
3) control at the virtual instrument of industrial computer (6) and call general image disposal system Image Sys3.0 (Beijing modern Fu Boke skill company limited) and spectrum signal analytic system on the platform respectively (encircling Electronic Instrument, Limited in the Tianjin produces, model is SD150) carry out image and acoustic analysis processing, extract four characteristic parameters of sample: defect area area (A
S), gradation of image entropy (ENT), characteristic peak frequency average (μ
f) and the characteristic peak frequency coefficient of variation (CV
f).Import to algorithm as shown in the formula in the computing module shown in 13 (branch's syntactic structure design is adopted in the C Programming with Pascal Language).Obtain the classification results of this sample.The output result also sends instruction control classification determining device (11) and sorts, and enters behind the sample classification and gathers chamber (12).
Its Chinese style 13 is by following fortran
PMED=0.312*CV
f+ 0.477A
S+ 0.211MAX (μ
f, ENT) (formula 1)
Claims (3)
1. the device of the inside and outside defective of Non-Destructive Testing abundance of water pears is characterized in that: comprise A illumination system, B transfer system, C detection system and D analysis judgment system, wherein:
The A illumination system
Illumination system is made up of four 25W halogen tungsten photoelectric tubes (7), four halogen tungsten photoelectric tubes (7) be installed in camera (1) around, apart from the camera Center Length is 10cm, the stiff end of camera and photoelectric tube posts the ater film, be used to eliminate the noise that reflected light causes, this stiff end is connected through the hinge on support (17);
The B transfer system
Transfer system is made up of travelling belt (10), pallet (9), collection chamber (12) and main stepper motor (13), and travelling belt (10) is by being installed in inner main stepper motor (13) control of apparatus housing (15), fixed tray on the travelling belt (9);
The C detection system
Detection system comprises image capture module and vibration frequency specturm analysis module, wherein image capture module is made of camera (1), image data acquiring card (2), halogen tungsten photoelectric tube (7), camera (1) and sound meter (3) are fixed on the support (17), camera (1) distance of camera lens pallet 25cm;
The vibration frequency specturm analysis module is made up of sound meter (3), Wave data capture card (4), stamp (8), auxilliary stepper motor (14), silencer pad (16), stamp (8) fixed rotating shaft is apart from pallet 15cm, stamp (8) is the PVC material, length 8cm, diameter 5mm; Auxilliary stepper motor (14) is installed in the position of the inner close stamp (8) of casing (15), control stamp (8) is to the lasting hammer action of sample, knock at every turn and produce an effective sound wave signal, and quilt collection enters Wave data capture card (4), silencer pad (16) is fixed in 5cm place, stamp below, is used for eliminating the issuable mixed and disorderly acoustic signal that does not meet acquisition range of the process of knocking;
The D analysis judgment system
Analysis judgment system comprises traffic pilot (5), industrial computer (6) and classification determining device (11), wherein traffic pilot (5) is connected with signal wire respectively with the vibration frequency specturm analysis module with image capture module in the detection system, schedule on the support (17) with hinges fixing, its output signal transfers to industrial computer (6) with the USB line, traffic pilot (5) imports industrial computer (6) into after with signal lumped modelling one digital conversion and carries out analyzing and processing, obtains classification results after classification determining device (11) analysis and judgement.
2. device according to claim 1 is characterized in that,
Camera (1) is the TK-C1380CCD of JVC company, image data acquiring card (2) is Canadian Matrox Meteror-II/Standard, sound meter (3) is a HS5633A type sound meter, the vibration frequency specturm analysis module adopts the SD150 dynamic analysis system, it is Ling Hua EBC-1200MediaGX1 that industrial computer (6) is embedded in formula mainboard model, traffic pilot (5) closes good electric industry Products Co., Ltd for the Zhongshan city and produces, and model is SH912A, the built-in NPLC-15 industrial control board of classification determining device (11).
3. claim 1 or 2 described devices are used for the method for the inside and outside defective of Non-Destructive Testing abundance of water pears, it is characterized in that:
Sample abundance of water pears are placed on the pallet (9), and travelling belt (10) brings into operation, the travelling speed 0.2m/s of travelling belt (10), interval time 0.6s; Each sample is gathered piece image by camera (1), and reaches image data acquiring card (2);
Stamp (8) knocks sample, and when knocking and sample place tangent plane is 30 degree angles, and each sample knocks twice at two ends, fruit body equator, knocks the power that produces 15N at every turn;
Note sound signal by sound meter (3), reach Wave data capture card (4) by signal wire, import traffic pilot (5) into by the USB line together with picture signal again and carry out conversion of signals, analog picture signal transfers data image signal to, analoging sound signal is converted into digital audio signal, and makes to enter at last after the fast Flourier FFT conversion data storage area on the industrial computer (6);
Control at the virtual instrument of industrial computer (6) that general image disposal system Image Sys3.0 and model are housed on the platform is that the vibration frequency specturm analysis module of SD150 is carried out image and acoustic analysis is handled, extract four characteristic parameters of sample: the defect area area A
s, gradation of image entropy ENT, characteristic peak frequency average μ
fAnd characteristic peak frequency coefficient of variation CV
f, substitution is suc as formula in 13 the formula, characteristic parameter MAX (μ in the formula
f, ENT) expression is got the higher value computing to the image entropy of characteristic frequency average and defective:
Obtain the classification results of this sample.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101813041A CN101603927B (en) | 2009-07-17 | 2009-07-17 | Device for nondestructive testing of defects of hoxiu pears and use method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101813041A CN101603927B (en) | 2009-07-17 | 2009-07-17 | Device for nondestructive testing of defects of hoxiu pears and use method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101603927A CN101603927A (en) | 2009-12-16 |
CN101603927B true CN101603927B (en) | 2011-07-20 |
Family
ID=41469748
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009101813041A Expired - Fee Related CN101603927B (en) | 2009-07-17 | 2009-07-17 | Device for nondestructive testing of defects of hoxiu pears and use method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101603927B (en) |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101915767B (en) * | 2010-07-13 | 2011-10-19 | 南京农业大学 | Method for systematically recognizing egg cracks |
CN102455324A (en) * | 2010-10-18 | 2012-05-16 | 河南工业大学 | DCT based method for extracting acoustical signal characteristics of grain and oil, and system thereof |
CN102445456B (en) * | 2011-10-09 | 2013-07-03 | 河海大学常州校区 | Device and method for nondestructively detecting egg quality based on DSP (Dynamic Signal Processing) machine vision |
CN103406286B (en) * | 2013-08-16 | 2015-06-24 | 常熟理工学院 | Online grading method for external quality of fruit based on LabVIEW |
CN104677499B (en) * | 2013-09-06 | 2017-06-23 | 闻泰通讯股份有限公司 | The system and method for fruit is selected using electronic equipment |
CN103487147B (en) * | 2013-09-06 | 2016-03-23 | 闻泰通讯股份有限公司 | Electronic equipment is utilized to select the system and method for fruit |
CN104475347B (en) * | 2014-12-12 | 2023-05-26 | 重庆瑶红食品有限公司 | Detection and sorting device for food |
CN108029599B (en) * | 2018-01-08 | 2024-07-02 | 江苏省农业科学院 | Crab disease discriminating method and diagnostic device |
CN109570051B (en) * | 2019-01-16 | 2020-09-11 | 郑州轻工业学院 | Chinese chestnut wormhole detection device based on machine vision, laser and acoustics |
US11113808B2 (en) * | 2019-02-04 | 2021-09-07 | Walmart Apollo, Llc | System and method for assessing quality of produce |
CN109827971B (en) * | 2019-03-19 | 2021-09-24 | 湖州灵粮生态农业有限公司 | Method for nondestructive detection of fruit surface defects |
CN110132857B (en) * | 2019-05-20 | 2021-09-07 | 南京林业大学 | Internal detection method for navel orange all-dimensional imaging |
CN110501065A (en) * | 2019-07-24 | 2019-11-26 | 南京农业大学 | Hybrid rice based on collision characteristic splits clever Seed inspection method |
CN110976337B (en) * | 2019-12-09 | 2021-06-22 | 商丘工学院 | Bearing detection device |
CN111291800A (en) * | 2020-01-21 | 2020-06-16 | 青梧桐有限责任公司 | House decoration type analysis method and system, electronic device and readable storage medium |
CN113051992B (en) * | 2020-11-16 | 2022-01-18 | 山东米捷软件有限公司 | Uniform speed identification system applying transparent card slot |
CN113588787B (en) * | 2021-07-28 | 2023-09-19 | 广东工业大学 | Watermelon maturity detector and detection method |
CN113663926A (en) * | 2021-08-18 | 2021-11-19 | 中山众信智能自动化设备有限公司 | Apple sorting unit with multiple detection function |
CN114563347A (en) * | 2022-01-20 | 2022-05-31 | 浙江大学 | Nondestructive testing system and method for hollowness of watermelon |
CN115032272B (en) * | 2022-06-08 | 2023-04-07 | 广西大学 | Structural defect detection device and method integrating knocking force signal and sound signal |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN2587534Y (en) * | 2002-12-27 | 2003-11-26 | 浙江大学 | Machine vision based fruit sorting machine |
CN1664568A (en) * | 2005-03-28 | 2005-09-07 | 浙江大学 | Fruit quality detection system based on image information fusion technology |
-
2009
- 2009-07-17 CN CN2009101813041A patent/CN101603927B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN2587534Y (en) * | 2002-12-27 | 2003-11-26 | 浙江大学 | Machine vision based fruit sorting machine |
CN1664568A (en) * | 2005-03-28 | 2005-09-07 | 浙江大学 | Fruit quality detection system based on image information fusion technology |
Non-Patent Citations (2)
Title |
---|
JP特开平10-115602A 1998.05.06 |
洪添胜等.基于高光谱图像技术的雪花梨品质无损检测.《农业工程学报》.2007,第23 卷(第2 期),151-155. * |
Also Published As
Publication number | Publication date |
---|---|
CN101603927A (en) | 2009-12-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101603927B (en) | Device for nondestructive testing of defects of hoxiu pears and use method | |
CN101915767B (en) | Method for systematically recognizing egg cracks | |
CN106442525B (en) | Online detection method for walnut internal shriveling defect | |
CN101413928A (en) | Fowl egg crack rapid on-line nondestructive detection device and method based on acoustic characteristic | |
CN103983190B (en) | Visual inspection classification device and method for appearance dimensions of group origin duck eggs | |
CN102147402A (en) | Machine vision technology based method for rapidly detecting egg freshness | |
CN107553219A (en) | A kind of Tool Wear Monitoring method based on multiple types sensor composite signal | |
CN110689042A (en) | Tunnel leakage grade identification method and device, storage medium and electronic device | |
CN103308525B (en) | A kind of online test method for metal wire rod production and device | |
CN108663367A (en) | A kind of egg quality lossless detection method based on egg unit weight | |
CN104280349A (en) | Method for identifying hollowness of white radishes based on hyperspectral image | |
CN105092579A (en) | Mango quality non-destructive testing device | |
CN104237235A (en) | Rapid detection method based on near-infrared imaging technology for food-borne pathogens | |
CN104655585B (en) | A kind of PSE meat screening technique based near infrared spectrum | |
CN109541031A (en) | Fruit hardness detection method based on acoustics and vibration characteristics | |
CN204613160U (en) | A kind of bird egg crack detection device | |
CN104849350B (en) | Defects in timber method for identifying and classifying based on multiple features | |
CN102928513B (en) | Ultrasonic device for nondestructive examination of watermelon maturity | |
Yang et al. | Wood veneer defect detection system based on machine vision | |
Li et al. | Non-destructive detection on the egg crack based on wavelet transform | |
CN104749249B (en) | A kind of method of the detection physical purity of seed based on ultrasonic technology | |
CN104833724A (en) | Poultry eggshell crack detection method and device | |
CN109406447A (en) | A kind of near infrared detection method of tannin in sorghum | |
CN105352555A (en) | Portable detection device for quickly identifying storage time of fowl eggs, and application method thereof | |
CN105954309B (en) | A kind of Fast nondestructive evaluation abalone is drying and the method for reconstitution process moisture |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110720 Termination date: 20180717 |