CN104374721A - Spliced beef rapid detection system and method - Google Patents

Abstract

The invention discloses a spliced beef rapid detection system and method. The system includes a controller and a detection mechanism, the detection mechanism comprises a base, the base is provided with a detection platform, a rectangular groove is arranged in the middle of the detection platform, a liquid is stored in the rectangular groove, a sample tray is arranged above the liquid, the front end and the rear end of the sample tray are respectively provided with a first permanent magnet and a second permanent magnet, the bottom of the sample tray is provided with a positioning column, the bottom of the positioning column is provided with a positioning head, the front inner side face of the rectangular groove is provided with a first electromagnet and a first proximity switch, the rear inner side face of the rectangular groove is provided with a second electromagnet, the bottom surface of the rectangular groove is provided with a plurality of positioning blocks, a first drive motor for driving the positioning blocks to rotate is arranged below the positioning blocks, the positioning blocks are provided with positioning holes, third electromagnets and second proximity switches are arranged at the bottom of the positioning holes, an optical transceiver is arranged above the sample tray, and a camera is arranged above rectangular groove. The spliced beef rapid detection system can fast, conveniently and accurately detect whether beef is spliced beef.

Description

A kind of system and method for quick detection splicing beef

Technical field

The present invention relates to technical field of food detection, particularly relate to a kind of system and method for quick detection splicing beef.

Background technology

Beef taste is delicious, and nutritious, mouthfeel is good, is first-class food materials.But because its production cost is higher, especially top grade beef block, expensive.And splice beef by the fragment beef of different size and be spliced, not only mouthfeel is bad, and the beef clod of separate sources also can form pollution each other, forms food-safety problem.

China Patent Publication No. CN1603794, publication date on April 6th, 2005, the name of invention is called the method and apparatus of rapidly detecting tenderness of beef utilizing near infrared technology, this application case discloses a kind of method and apparatus of rapidly detecting tenderness of beef utilizing near infrared technology, and it is made up of near-infrared light source, near infrared detector, diffuse reflection fiber plant, microprocessor, display and pen recorder and rotatable objective table.Its weak point is, whether this pick-up unit can not detect beef is splicing beef.

Summary of the invention

The object of the invention is to overcome current beef detection system and can not detect the technical matters whether beef be splicing beef, provide a kind of system and method for quick detection splicing beef, it can detect whether beef is splicing beef quick, easy, exactly.

In order to solve the problem, the present invention is achieved by the following technical solutions:

The system of a kind of quick detection splicing beef of the present invention, comprise controller and testing agency, described controller comprises CPU (central processing unit), touch-screen and light source control module, described testing agency comprises base, described base is provided with monitor station, rectangular recess is provided with in the middle part of described monitor station, liquid is stored in described rectangular recess, described liquid is provided with rounded sample tray, the front-end and back-end of described sample tray are respectively equipped with the first permanent magnet and the second permanent magnet, reference column is provided with bottom described sample tray, positioning head is provided with bottom described reference column, described positioning head is made up of ferrimagnet, the anteromedial surface of described rectangular recess is provided with the first electromagnet and the first proximity switch, the rear medial surface of described rectangular recess is provided with the second electromagnet, the bottom surface of described rectangular recess is provided with several locating pieces, the first drive motor driving locating piece to rotate is provided with below described locating piece, described locating piece is provided with and coordinates with positioning head and to limit the pilot hole that positioning head freely rotates, pilot hole on several locating pieces described is arranged in a linear from front to back, the 3rd electromagnet and the second proximity switch is provided with bottom described pilot hole, optical transceiver is provided with above described sample tray, described optical transceiver comprises radiation source and optical receiver, camera is provided with above described rectangular recess, described CPU (central processing unit) respectively with touch-screen, light source control module, first electromagnet, second electromagnet, 3rd electromagnet, first proximity switch, second proximity switch, first drive motor, optical receiver and camera electrical connection, described light source control module is also electrically connected with radiation source.

In the technical program, radiation source is Halogen lamp LED or generating laser.During detection, beef sample to be measured is placed on sample tray, makes the center being centrally located at sample tray of beef sample.

Central processing unit controls first work of electromagnet, first electromagnet produces suction, the first permanent magnet on sample tray is by the first electromagnet suction effect, sample tray moves to the anteromedial surface of rectangular recess, first permanent magnet is subject to suction just to the first electromagnet simultaneously, the position of calibration positioning head, ensures that positioning head can insert pilot hole.Distance between the madial wall of rectangular recess left and right is more bigger than the diameter of sample tray, makes sample tray be merely able to seesaw, and ensures the straight line upper direction that positioning head is arranged at pilot hole, is convenient to positioning head and inserts pilot hole.When sample tray moves to rectangular recess anteromedial surface place, first proximity switch sends trigger pip, CPU (central processing unit) receives the trigger pip that the first proximity switch sends, control the first electromagnet to quit work, the 3rd work of electromagnet bottom the pilot hole that command range rectangular recess anteromedial surface is nearest, 3rd electromagnet produces suction, and the positioning head bottom sample tray inserts this pilot hole by the suction of the 3rd electromagnet, and the second proximity switch in this pilot hole sends trigger pip.CPU (central processing unit) receives trigger pip, and judgement sample pallet is located successfully, and now the central point of sample tray is positioned at immediately below optical transceiver.

User selects sample tray to need the position of positioning hole inserted by touch-screen.The 3rd electromagnet bottom the pilot hole that central processing unit controls distance rectangular recess anteromedial surface is nearest quits work, and sample tray floats by buoyancy, and the positioning head bottom sample tray leaves pilot hole.Central processing unit controls second electromagnet and sample tray need the 3rd work of electromagnet bottom the pilot hole of insertion, and the second permanent magnet on sample tray is by the second electromagnet suction effect, and sample tray is to the rear medial surface motion of rectangular recess.When above sample tray moves to the pilot hole needing to insert, the positioning head below sample tray is by the 3rd electromagnet suction effect, and the positioning head bottom sample tray inserts this pilot hole, and the second proximity switch in this pilot hole sends trigger pip.CPU (central processing unit) receives trigger pip, and judgement sample pallet successfully inserts the pilot hole that user selects.Sample tray moves forward and backward in a liquid, without frictional dissipation, realizes accurately location by pilot hole.

The image information of camera collection rectangular recess, and be sent to CPU (central processing unit), CPU (central processing unit) obtains the positional information of sample tray by image procossing, monitor when sample tray moves forward and backward, if sample tray does not move to directly over the pilot hole of needs insertion, CPU (central processing unit) adjusts the position of sample tray by the first electromagnet and the second electromagnet, is convenient to the position that CPU (central processing unit) adjusts sample tray more accurately.

The locating piece at the pilot hole place that central processing unit controls sample tray inserts rotates a circle, and often horizontally rotates 30 degree of detection spectroscopic datas that stop.When detecting spectroscopic data, the intensity curve of the detection light that radiation source sends first rises to maximal value from 0 according to straight line, then drops to 0 from maximal value according to sinusoidal curve.Optical receiver detects the intensity of reflected light received, and sends it to CPU (central processing unit), after CPU (central processing unit) carries out respective handling to detection data, judges whether beef sample to be measured is splicing beef.

The principle of optical detection splicing beef: the beef lines of monoblock naturally continuously, orderliness is clear, color and luster is unified, water cut is close, and therefore it is close with the reflection absorption spectrum signal of incident angle, can as the foundation detected.Although and the different fragments splicing meat pass through arrangement, gluing and surface treatment, its inherent meat lines is also discontinuous, have overall tomography, color and luster difference is comparatively large, and water cut is unanimously very difficult, therefore can determine whether as splicing beef according to optical detecting method.

The present invention is that the incident light adopting light intensity constantly to change irradiates sample, in the intensity of incident light gradually in the large or change procedure that reduces, the absorption of different groups to respective wavelength light is gradually large or reduction, now the degree of absorption of group is in unsaturation, saturated diminuendo process, comprise more Detection Information in reflected light, thus whether enable the detection signal obtained symbolize beef be more accurately splicing.

As preferably, described positioning head is flat, and the longitudinal section of described positioning head is inverted isosceles trapezoid, and the shape of described pilot hole and the form fit of positioning head, described pilot hole top is provided with the spigot surface of inclination.Pilot hole is along direction setting from front to back, and the spigot surface of the left and right sides at pilot hole top play the guiding role, and is convenient to positioning head and inserts pilot hole.

As preferably, the distance between the madial wall of described rectangular recess left and right and the diameter of sample tray match.Distance between the madial wall of rectangular recess left and right is more bigger than the diameter of sample tray, makes sample tray be merely able to seesaw, and ensures the straight line upper direction that positioning head is arranged at pilot hole, is convenient to positioning head and inserts pilot hole.

As preferably, described light source control module comprises function memory and power amplifier, the input end of described function memory is electrically connected with CPU (central processing unit), and the output terminal of described function memory is electrically connected with the input end of power amplifier, and the output terminal of described power amplifier is electrically connected with radiation source.Function memory stores the function information that CPU (central processing unit) sends, and controls by power amplifier the detection light that radiation source sends varying strength.

As preferably, the pilot hole that on described rectangular recess bottom surface, distance rectangular recess anteromedial surface is nearest is positioned at immediately below optical transceiver.The pilot hole nearest apart from rectangular recess anteromedial surface is used for the initial alignment of sample tray, and the distance between the pilot hole of foremost and rearmost pilot hole is the radius of sample tray.

The method of a kind of quick detection splicing beef of the present invention, comprises the following steps:

S1: preparation sheet beef sample, is placed on beef sample on sample tray, makes the center being centrally located at sample tray of beef sample;

S2: the instruction that CPU (central processing unit) is inputted by touch-screen according to user, Quality control pallet inserts the pilot hole that user specifies;

S3: the detection light that CPU (central processing unit) sends certain light intensity by light source control module control radiation source is radiated on beef sample, detect light intensity and first rise to maximal value from 0 according to straight line, then 0 is dropped to from maximal value according to sinusoidal curve, optical receiver gathers reflected spectrum data Spect (t), then central processing unit controls sample tray insert pilot hole place locating piece below the first drive motor work, first drive motor controls locating piece and rotates 30 degree, sample tray is made to horizontally rotate 30 degree, optical receiver gathers the spectroscopic data of now radiation source point of irradiation, Quality control pallet horizontal rotary like this three-sixth turn, often horizontally rotate 30 degree of detection spectroscopic datas that stop, thus on beef sample, gather spectroscopic data Spect (t) of 12 diverse locations,

S4: collect 12 spectroscopic datas Spect (t) are all carried out same data processing by CPU (central processing unit), calculate 12 signal to noise ratio (S/N ratio) eigenwerts, comprise the following steps the data processing that each spectroscopic data Spect (t) is carried out:

Adopt input signal as input matrix, potential function V (x, t, α) and input signal are worked in coordination with as one deck accidental resonance model:

Wherein, V (x, t, α) is potential function, and x (t) is Brownian Particles movement locus function, and t is run duration, for periodic sinusoidal signal, N (t) grasps noise in being, A is signal amplitude, and f is signal frequency, and D is external noise intensity, and ξ (t) is external noise, for phase place,

Calculate V (x, t, α) for the first order derivative of x (t), second derivative and three order derivatives, and make equation equal 0, obtain two layers of accidental resonance model:

Setting noise intensity D=0, spect (t)=0, N (t)=0, the critical value calculating A is by A _csubstitute in one deck accidental resonance model, and set x ₀(t)=0, sn ₀=0, adopt quadravalence jade for asking rain Ge Kuta Algorithm for Solving one deck accidental resonance model, obtain:

$\begin{array}{c}{x}_{n+1}\left(t\right)=x_n\left(t\right)+1/6[{\left(k_1\right)}_n+(2-\sqrt{2}){\left(k_2\right)}_n+(2+\sqrt{2}){\left(k_3\right)}_n+{\left(k_4\right)}_n]\\ n=\mathrm{0,1},...,N-1\end{array},---\left(3\right)$

Wherein, undetermined coefficient:

(k ₁) _n＝a(αx _n-1(t)) ²b(αx _n-1(t)) ³+sn _n-1(4)

${\left(k_2\right)}_n=a({\mathrm{\αx}}_{n-1}\left(t\right)+\frac{{\left(k_1\right)}_{n-1}}{3})-b{({\mathrm{\αx}}_{n-1}\left(t\right)+\frac{{\left(k_1\right)}_{n-1}}{3})}^3+{\mathrm{sn}}_{n-1}---\left(5\right)$

${\left(k_3\right)}_n=a({\mathrm{\αx}}_{n-1}\left(t\right)+\frac{{\left(k_2\right)}_{n-1}}{3})-b{({\mathrm{\αx}}_{n-1}\left(t\right)+\frac{\sqrt{2}-1}{3}{\left(k_1\right)}_{n-1}+\frac{2-\sqrt{2}}{3}{\left(k_2\right)}_{n-1})}^3+{\mathrm{sn}}_{n+1}---\left(6\right)$

${\left(k_4\right)}_n=a(3\mathrm{\α}{x}_{n-1}\left(t\right)+{\left(k_3\right)}_{n-1})-b{({\mathrm{\αx}}_{n-1}\left(t\right)-\frac{\sqrt{2}}{3}{\left(k_2\right)}_{n-1}+\frac{2+\sqrt{2}}{3}{\left(k_3\right)}_{n-1})}^3+{\mathrm{sn}}_{n+1}---\left(7\right)$

Wherein, x _nt n order derivative that () is x (t), sn _nbe the value of n order derivative at t=0 place of S (t), a, b are the constant of setting,

Calculate x ₁(t), x ₂(t) ... x _n+1t the value of (), to x ₁(t), x ₂(t) ... x _n+1t () is carried out integration and is obtained x (t), and obtain x (t) produces resonance moment position x in the double-deck stochastic resonance system of one deck accidental resonance model and two layers of accidental resonance model composition _mvalue and x _mcorresponding resonance moment t1, the acceleration of motion α 1 corresponding with t1 and the noise D1 corresponding with t1, D1 are a value in D,

Pass through formula $\mathrm{SNR}=\sqrt{2}{\left(\mathrm{\ΔU}\frac{{4a}^3/27b}{D_1}\right)}^2{e}^{-{\left(\mathrm{\ΔU}\right)}^2/D_1}$ , calculate signal to noise ratio (S/N ratio) eigenwert SNR _feature, wherein, Δ U=a ²/ 4b;

S5: calculate 12 signal to noise ratio (S/N ratio) eigenwerts are divided into four groups, by SNR _{feature 3}, SNR _{feature 6}, SNR _{feature 9}, SNR _{feature 12}be divided into first group, by SNR _{feature 2}, SNR _{feature 4}, SNR _{feature 7}, SNR _{feature 11}be divided into second group, by SNR _{feature 1}and SNR _{feature 5}be divided into the 3rd group, by SNR _{feature 8}and SNR _{feature 10}be divided into the 4th group, and calculate the signal to noise ratio (S/N ratio) mean value of each group respectively, obtain SNR _{average 1}, SNR _{average 2}, SNR _{average 3}, SNR _{average 4},

Calculate the error QE between each signal to noise ratio (S/N ratio) eigenwert and signal to noise ratio (S/N ratio) mean value of the group corresponding to it _j, j=1 ..., 12;

Statistics meets QE _jthe number M of the signal to noise ratio (S/N ratio) eigenwert of≤2% ₁, statistics meets QE _jthe number M of the signal to noise ratio (S/N ratio) eigenwert of > 2% ₂;

S6: if then CPU (central processing unit) judges that this beef sample is not splicing beef, if then CPU (central processing unit) judges that this beef sample is splicing beef, otherwise jumps to step S3, re-starts detection to beef sample.

As preferably, described step S2 comprises the following steps:

S21: central processing unit controls first work of electromagnet, sample tray is by the first electromagnet suction action to rectangular recess anteromedial surface place, and the first proximity switch sends trigger pip;

S22: CPU (central processing unit) receives the trigger pip that the first proximity switch sends, control the first electromagnet to quit work, the 3rd work of electromagnet bottom the pilot hole that command range rectangular recess anteromedial surface is nearest, 3rd electromagnet produces suction, positioning head bottom sample tray inserts this pilot hole by the suction of the 3rd electromagnet, and the second proximity switch in this pilot hole sends trigger pip;

S23: CPU (central processing unit) receives the trigger pip that the second proximity switch sends, judgement sample pallet is located successfully, and now the central point of sample tray is positioned at immediately below optical transceiver;

S24: CPU (central processing unit) reads the position of positioning hole that user needs by the instruction determination sample tray that touch-screen inputs to insert;

S25: the 3rd electromagnet bottom the pilot hole that central processing unit controls distance rectangular recess anteromedial surface is nearest quits work, and sample tray floats by buoyancy, and the positioning head bottom sample tray leaves pilot hole;

S26: central processing unit controls second electromagnet and sample tray need the 3rd work of electromagnet bottom the pilot hole of insertion, and the positioning head bottom sample tray inserts this pilot hole, and the second proximity switch in this pilot hole sends trigger pip.

Substantial effect of the present invention is: (1) can detect whether beef is splicing beef quick, easy, exactly.(2) sample tray moves forward and backward in a liquid, without frictional dissipation, realizes accurately location by pilot hole.

Accompanying drawing explanation

Fig. 1 is a kind of structural representation of the present invention;

Fig. 2 is the structural representation of rectangular recess;

Fig. 3 is the structural representation of rectangular recess and sample tray;

Fig. 4 is the sectional view of Fig. 3;

Fig. 5 is the structural representation of pilot hole;

Fig. 6 is that circuit theory of the present invention connects block diagram;

The light intensity curve figure of the detection light that radiation source sends when Fig. 7 is detection.

In figure: 1, CPU (central processing unit), 2, touch-screen, 3, light source control module, 4, base, 5, monitor station, 6, rectangular recess, 7, sample tray, 8, first permanent magnet, 9, second permanent magnet, 10, reference column, 11, positioning head, 12, first electromagnet, 13, second electromagnet, 14, locating piece, 15, first drive motor, 16, pilot hole, 17, 3rd electromagnet, 18, first proximity switch, 19, optical transceiver, 20, radiation source, 21, optical receiver, 22, elevating mechanism, 23, spigot surface, 24, function memory, 25, power amplifier, 26, camera, 27, transverse link, 28, longitudinal rod, 29, screw mandrel, 30, support bar, 31, thread bush, 32, sleeve pipe, 33, second proximity switch.

Embodiment

Below by embodiment, and by reference to the accompanying drawings, technical scheme of the present invention is described in further detail.

Embodiment: the system of a kind of quick detection splicing beef of the present embodiment, as Fig. 1, Fig. 2, Fig. 3, shown in Fig. 4, comprise controller and testing agency, controller comprises CPU (central processing unit) 1, touch-screen 2 and light source control module 3, testing agency comprises base 4, base 4 is provided with monitor station 5, rectangular recess 6 is provided with in the middle part of monitor station 5, liquid is stored in rectangular recess 6, liquid is provided with rounded sample tray 7, the front-end and back-end of sample tray 7 are respectively equipped with the first permanent magnet and the second permanent magnet, reference column 10 is provided with bottom sample tray 7, positioning head 11 is provided with bottom reference column 10, positioning head 11 is made up of ferrimagnet, the anteromedial surface of rectangular recess 6 is provided with the first electromagnet 12 and the first proximity switch 18, the rear medial surface of rectangular recess 6 is provided with the second electromagnet 13, the bottom surface of rectangular recess 6 is provided with seven locating pieces 14, the first drive motor 15 driving locating piece 14 to rotate is provided with below locating piece 14, locating piece 14 is provided with and coordinates with positioning head 11 and to limit the pilot hole 16 that positioning head 11 freely rotates, pilot hole 16 on seven locating pieces 14 is arranged in a linear from front to back, the 3rd electromagnet 17 and the second proximity switch 33 is provided with bottom pilot hole 16, optical transceiver 19 is provided with above sample tray 7, optical transceiver 19 comprises radiation source 20 and optical receiver 21, base 4 is also provided with the elevating mechanism 22 driving optical transceiver 19 to move up and down.

As shown in Figure 6, light source control module 3 comprises function memory 24 and power amplifier 25, camera 26 is also provided with above rectangular recess 6, camera 26 is connected with elevating mechanism 22 by web member, CPU (central processing unit) 1 respectively with touch-screen 2, first electromagnet 12, second electromagnet 13, 3rd electromagnet 17, first proximity switch 18, second proximity switch 33, first drive motor 15, optical receiver 21, elevating mechanism 22, camera 26, the input end electrical connection of function memory 24, the output terminal of function memory 24 is electrically connected with the input end of power amplifier 25, the output terminal of power amplifier 25 is electrically connected with radiation source 20.

The second drive motor that elevating mechanism 22 comprises transverse link 27, longitudinal rod 28, screw mandrel 29, support bar 30 and drives screw mandrel 29 to rotate, screw mandrel 29 and support bar 30 are vertically arranged on base 4, transverse link 27 one end is provided with the thread bush 31 be set on screw mandrel 29 and the sleeve pipe 32 be set on support bar 30, described transverse link 27 other end is connected with the top of longitudinal rod 28, described optical transceiver 19 is arranged on the bottom of longitudinal rod 28, straight down, the second drive motor is electrically connected with CPU (central processing unit) 1 direction of illumination of radiation source 20.CPU (central processing unit) 1 adjusts the upper-lower position of optical transceiver 29 by elevating mechanism 22, with satisfied actual detection demand.

Positioning head 11 is in flat, and the longitudinal section of positioning head 11 is inverted isosceles trapezoid, the shape of pilot hole 16 and the form fit of positioning head 11, and as shown in Figure 5, the left and right sides, pilot hole 16 top is respectively equipped with the spigot surface 23 of inclination.Pilot hole 16 is along direction setting from front to back, and the spigot surface 23 of the left and right sides, pilot hole 16 top play the guiding role, and is convenient to positioning head 11 and inserts pilot hole 16.

Distance between rectangular recess about 6 madial wall and the diameter of sample tray 7 match.Distance between rectangular recess about 6 madial wall 2cm larger than the diameter of sample tray 7, makes sample tray 7 be merely able to seesaw, and ensures the straight line upper direction that positioning head 11 is arranged at pilot hole 16, is convenient to positioning head 11 and inserts pilot hole 16.

The pilot hole 16 that on rectangular recess 6 bottom surface, distance rectangular recess 6 anteromedial surface is nearest is positioned at immediately below optical transceiver 19.The pilot hole 16 nearest apart from rectangular recess 6 anteromedial surface is for the initial alignment of sample tray 7, and the distance between the pilot hole 16 of foremost and rearmost pilot hole 16 is the radius of sample tray 7.

The annular sidewall that sample tray 7 comprises rounded chassis and is connected with chassis outer rim, first permanent magnet and the second permanent magnet are embedded in the front-end and back-end of annular sidewall respectively, tray bottom is provided with the projection protruded downwards, and projection is spherical, and reference column 10 is arranged on the bottom of projection.Projection is spherical, and sample tray 7 can be fluctuated stably in a liquid, chassis for placing beef sample to be measured, on the beef sample to be measured that annular sidewall can prevent liquid from splashing on chassis.

Function memory 24 stores the function information that CPU (central processing unit) 1 sends, and controls by power amplifier 25 the detection light that radiation source 20 sends varying strength.Camera 26 gathers the image information of rectangular recess 6, and be sent to CPU (central processing unit) 1, CPU (central processing unit) 1 obtains the positional information of sample tray 7 by image procossing, monitor when sample tray 7 moves forward and backward, if sample tray 7 does not move to directly over the pilot hole 16 of needs insertion, CPU (central processing unit) 1 adjusts the position of sample tray 7 by the first electromagnet 12 and the second electromagnet 13, is convenient to the position that CPU (central processing unit) 1 adjusts sample tray 7 more accurately.Radiation source 20 is Halogen lamp LED.

During detection, beef sample to be measured is placed on sample tray 7, makes the center being centrally located at sample tray 7 of beef sample.

CPU (central processing unit) 1 controls the first electromagnet 12 and works, first electromagnet 12 produces suction, the first permanent magnet on sample tray 7 is by the first electromagnet 12 suction, sample tray 7 moves to the anteromedial surface of rectangular recess 6, first permanent magnet is subject to suction just to the first electromagnet 12 simultaneously, the position of calibration positioning head 11, ensures that positioning head 11 can insert pilot hole 16.Distance between rectangular recess about 6 madial wall is more bigger than the diameter of sample tray 7, makes sample tray 7 be merely able to seesaw, and ensures the straight line upper direction that positioning head 11 is arranged at pilot hole 16, is convenient to positioning head 11 and inserts pilot hole 16.When sample tray 7 contacts with rectangular recess 6 anteromedial surface (now, positioning head 11 is positioned at directly over the nearest pilot hole 16 of distance rectangular recess 6 anteromedial surface), first proximity switch 18 sends trigger pip, CPU (central processing unit) 1 receives the trigger pip that the first proximity switch 18 sends, control the first electromagnet 12 to quit work, the 3rd electromagnet 17 bottom the pilot hole 16 that command range rectangular recess 6 anteromedial surface is nearest works, 3rd electromagnet 17 produces suction, positioning head 11 bottom sample tray 7 inserts this pilot hole 16 by the suction of the 3rd electromagnet 17, the second proximity switch 33 in this pilot hole 16 sends trigger pip.CPU (central processing unit) 1 receives the trigger pip that the second proximity switch 33 sends, and judgement sample pallet 7 is located successfully, and now the central point of sample tray 7 is positioned at immediately below optical transceiver 19.

User selects sample tray 7 to need pilot hole 16 position of inserting by touch-screen 2.The 3rd electromagnet 17 bottom the pilot hole 16 that CPU (central processing unit) 1 command range rectangular recess 6 anteromedial surface is nearest quits work, and sample tray 7 floats by buoyancy, and the positioning head 11 bottom sample tray 7 leaves pilot hole 16.CPU (central processing unit) 1 controls the second electromagnet 13 and sample tray 7 needs the 3rd electromagnet 17 bottom the pilot hole 16 of insertion to work, the second permanent magnet on sample tray 7 is by the second electromagnet 13 suction, and sample tray 7 is to the rear medial surface motion of rectangular recess 6.When sample tray 7 move to need insert pilot hole 16 above time, positioning head 11 below sample tray 7 is by the 3rd electromagnet 13 suction, positioning head 11 bottom sample tray 7 inserts this pilot hole 16, and the second proximity switch 33 in this pilot hole 16 sends trigger pip.CPU (central processing unit) 1 receives trigger pip, and judgement sample pallet 7 successfully inserts the pilot hole 16 that user selects.Sample tray 7 moves forward and backward in a liquid, without frictional dissipation, realizes accurate location by pilot hole 16.

The locating piece 14 at pilot hole 16 place that CPU (central processing unit) 1 Quality control pallet 7 inserts rotates a circle, and often horizontally rotates 30 degree of detection spectroscopic datas that stop.When detecting spectroscopic data, the intensity curve of the detection light that radiation source 20 sends first rises to maximal value from 0 according to straight line, then drops to 0 from maximal value according to sinusoidal curve, as shown in Figure 7.Optical receiver 21 detects the intensity of reflected light curve received, and sends it to CPU (central processing unit) 1, after CPU (central processing unit) 1 carries out respective handling to detection data, judges whether beef sample to be measured is splicing beef.

The principle of optical detection splicing beef: the beef lines of monoblock naturally continuously, orderliness is clear, color and luster is unified, water cut is close, and therefore it is close with the reflection absorption spectrum signal of incident angle, can as the foundation detected.Although and the different fragments splicing meat pass through arrangement, gluing and surface treatment, its inherent meat lines is also discontinuous, have overall tomography, color and luster difference is comparatively large, and water cut is unanimously very difficult, therefore can determine whether as splicing beef according to optical detecting method.

The present invention is that the incident light adopting light intensity constantly to change irradiates sample, in the intensity of incident light gradually in the large or change procedure that reduces, the absorption of different groups to respective wavelength light is gradually large or reduction, now the degree of absorption of group is in unsaturation, saturated diminuendo process, comprise more Detection Information in reflected light, thus whether enable the detection signal obtained symbolize beef be more accurately splicing.

The method of a kind of quick detection splicing beef of the present embodiment, is applicable to the system of above-mentioned a kind of quick detection splicing beef, comprises the following steps:

S1: prepare the beef sample that thickness is 7mm-8mm, be placed on by beef sample on sample tray, makes the center being centrally located at sample tray of beef sample;

S2: the instruction that CPU (central processing unit) is inputted by touch-screen according to user, Quality control pallet inserts the pilot hole that user specifies;

S3: the detection light that CPU (central processing unit) sends certain light intensity by light source control module control radiation source is radiated on beef sample, detect light intensity and first rise to maximal value from 0 according to straight line, then 0 is dropped to from maximal value according to sinusoidal curve, optical receiver gathers reflected spectrum data Spect (t), then central processing unit controls sample tray insert pilot hole place locating piece below the first drive motor work, first drive motor controls locating piece and rotates 30 degree, sample tray is made to horizontally rotate 30 degree, optical receiver gathers the spectroscopic data of now radiation source point of irradiation, Quality control pallet horizontal rotary like this three-sixth turn, often horizontally rotate 30 degree of detection spectroscopic datas that stop, thus on beef sample, gather spectroscopic data Spect (t) of 12 diverse locations,

S4: collect 12 spectroscopic datas Spect (t) are all carried out same data processing by CPU (central processing unit), calculate 12 signal to noise ratio (S/N ratio) eigenwerts, comprise the following steps the data processing that each spectroscopic data Spect (t) is carried out:

Adopt input signal as input matrix,

Under adiabatic approximation condition, suppose signal amplitude minimum (A < < 1), bistable system is not when having enough energy drives, Brownian movement particle is offset in the potential well of side, signal period is more than system relaxation time length in some typical potential wells, the now appearance of cyclic drive power makes potential function run-off the straight, finally cause Brownian movement particle from a potential well to the transition of another one potential well, therefore potential function V (x, t, α) work in coordination with as one deck accidental resonance model with input signal:

Wherein, V (x, t, α) is potential function, and x (t) is Brownian Particles movement locus function, and t is run duration, for periodic sinusoidal signal, N (t) grasps noise in being, A is signal amplitude, and f is signal frequency, and D is external noise intensity, and ξ (t) is external noise, for phase place,

Calculate V (x, t, α) for the first order derivative of x (t), second derivative and three order derivatives, and make equation equal 0, obtain two layers of accidental resonance model:

Setting noise intensity D=0, spect (t)=0, N (t)=0, the critical value calculating A is at A < A _cwhen, Brownian movement particle is hovered about its original position, can not realize the transition between two potential wells, but particle is when obtaining external noise and intervening, even if A < is A _cit also can complete the transition between potential well, and the generating process of Here it is accidental resonance, by A _csubstitute in one deck accidental resonance model, and set x ₀(t)=0, sn ₀=0, adopt quadravalence jade for asking rain Ge Kuta Algorithm for Solving one deck accidental resonance model, obtain:

$\begin{array}{c}{x}_{n+1}\left(t\right)=x_n\left(t\right)+1/6[{\left(k_1\right)}_n+(2-\sqrt{2}){\left(k_2\right)}_n+(2+\sqrt{2}){\left(k_3\right)}_n+{\left(k_4\right)}_n]\\ n=\mathrm{0,1},...,N-1\end{array},---\left(3\right)$

Wherein, undetermined coefficient:

(k ₁) _n＝a(αx _n-1(t)) ²b(αx _n-1(t)) ³+sn _n-1(4)

${\left(k_2\right)}_n=a({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{{\left(k_1\right)}_{n-1}}{3})-b{({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{{\left(k_1\right)}_{n-1}}{3})}^3+{\mathrm{sn}}_{n-1}---\left(5\right)$

${\left(k_3\right)}_n=a({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{{\left(k_2\right)}_{n-1}}{3})-b{({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{\sqrt{2}-1}{3}{\left(k_1\right)}_{n-1}+\frac{2-\sqrt{2}}{3}{\left(k_2\right)}_{n-1})}^3+{\mathrm{sn}}_{n+1}---\left(6\right)$

${\left(k_4\right)}_n=a(3\mathrm{\&alpha;}{x}_{n-1}\left(t\right)+{\left(k_3\right)}_{n-1})-b{({\mathrm{\&alpha;x}}_{n-1}\left(t\right)-\frac{\sqrt{2}}{3}{\left(k_2\right)}_{n-1}+\frac{2+\sqrt{2}}{3}{\left(k_3\right)}_{n-1})}^3+{\mathrm{sn}}_{n+1}---\left(7\right)$

Wherein, x _nt n order derivative that () is x (t), sn _nbe the value of n order derivative at t=0 place of S (t), a, b are the constant of setting,

Calculate x ₁(t), x ₂(t) ... x _n+1t the value of (), to x ₁(t), x ₂(t) ... x _n+1t () is carried out integration and is obtained x (t), and obtain x (t) produces resonance moment position x in the double-deck stochastic resonance system of one deck accidental resonance model and two layers of accidental resonance model composition _mvalue and x _mcorresponding resonance moment t1, the acceleration of motion α 1 corresponding with t1 and the noise D1 corresponding with t1, D1 are a value in D,

Pass through formula $\mathrm{SNR}=\sqrt{2}{\left(\mathrm{\&Delta;U}\frac{{4a}^3/27b}{D_1}\right)}^2{e}^{-{\left(\mathrm{\&Delta;U}\right)}^2/D_1}$ , calculate signal to noise ratio (S/N ratio) eigenwert SNR _feature, wherein, Δ U=a ²/ 4b;

S5: calculate 12 signal to noise ratio (S/N ratio) eigenwerts are divided into four groups, by SNR _{feature 3}, SNR _{feature 6}, SNR _{feature 9}, SNR _{feature 12}be divided into first group, by SNR _{feature 2}, SNR _{feature 4}, SNR _{feature 7}, SNR _{feature 11}be divided into second group, by SNR _{feature 1}and SNR _{feature 5}be divided into the 3rd group, by SNR _{feature 8}and SNR _{feature 10}be divided into the 4th group, and calculate the signal to noise ratio (S/N ratio) mean value of each group respectively, obtain SNR _{average 1}, SNR _{average 2}, SNR _{average 3}, SNR _{average 4},

Calculate the error QE between each signal to noise ratio (S/N ratio) eigenwert and signal to noise ratio (S/N ratio) mean value of the group corresponding to it _j, j=1 ..., 12;

Statistics meets QE _jthe number M of the signal to noise ratio (S/N ratio) eigenwert of≤2% ₁, statistics meets QE _jthe number M of the signal to noise ratio (S/N ratio) eigenwert of > 2% ₂;

S6: if then CPU (central processing unit) judges that this beef sample is not splicing beef, if then CPU (central processing unit) judges that this beef sample is splicing beef, otherwise jumps to step S3, re-starts detection to beef sample.

Light intensity first rises to maximal value from 0 according to straight line, and being then applicable to detecting thickness from maximal value according to the detection light that sinusoidal curve drops to 0 is the beef sample of 7mm-8mm.In step S6, if with two formula are all false, then can insert the pilot hole adjacent with this pilot hole by Quality control pallet, then jump to step S3 and re-start detection to beef sample.

Step S2 comprises the following steps:

S21: central processing unit controls first work of electromagnet, sample tray is by the first electromagnet suction action to rectangular recess anteromedial surface place, and the first proximity switch sends trigger pip;

S22: CPU (central processing unit) receives the trigger pip that the first proximity switch sends, control the first electromagnet to quit work, the 3rd work of electromagnet bottom the pilot hole that command range rectangular recess anteromedial surface is nearest, 3rd electromagnet produces suction, positioning head bottom sample tray inserts this pilot hole by the suction of the 3rd electromagnet, and the second proximity switch in this pilot hole sends trigger pip;

S23: CPU (central processing unit) receives the trigger pip that the second proximity switch sends, judgement sample pallet is located successfully, and now the central point of sample tray is positioned at immediately below optical transceiver;

S24: CPU (central processing unit) reads the position of positioning hole that user needs by the instruction determination sample tray that touch-screen inputs to insert;

S25: the 3rd electromagnet bottom the pilot hole that central processing unit controls distance rectangular recess anteromedial surface is nearest quits work, and sample tray floats by buoyancy, and the positioning head bottom sample tray leaves pilot hole;

S26: central processing unit controls second electromagnet and sample tray need the 3rd work of electromagnet bottom the pilot hole of insertion, sample tray moves to above the pilot hole of needs insertion by the second electromagnet suction effect, positioning head bottom sample tray inserts this pilot hole by the 3rd electromagnet suction effect, and the second proximity switch in this pilot hole sends trigger pip.The image information of camera collection rectangular recess, and be sent to CPU (central processing unit), CPU (central processing unit) obtains the positional information of sample tray by image procossing, if sample tray is excessively mobile backward by the second electromagnet suction effect, move to the rear needing the pilot hole inserted, CPU (central processing unit) adjusts the position of sample tray by the first electromagnet and the second electromagnet, and sample tray is remained on directly over the pilot hole of needs insertion.

In the present embodiment, QE is met through detecting _jthe number M of the signal to noise ratio (S/N ratio) eigenwert of≤2% ₁=1, statistics meets QE _jthe number of the signal to noise ratio (S/N ratio) eigenwert of > 2% is M ₂=11, judge that beef sample is splicing beef.

Claims (7)

1. one kind is detected the system of splicing beef fast, it is characterized in that: comprise controller and testing agency, described controller comprises CPU (central processing unit) (1), touch-screen (2) and light source control module (3), described testing agency comprises base (4), described base (4) is provided with monitor station (5), described monitor station (5) middle part is provided with rectangular recess (6), described rectangular recess stores liquid in (6), described liquid is provided with rounded sample tray (7), the front-end and back-end of described sample tray (7) are respectively equipped with the first permanent magnet and the second permanent magnet, described sample tray (7) bottom is provided with reference column (10), described reference column (10) bottom is provided with positioning head (11), described positioning head (11) is made up of ferrimagnet, the anteromedial surface of described rectangular recess (6) is provided with the first electromagnet (12) and the first proximity switch (18), the rear medial surface of described rectangular recess (6) is provided with the second electromagnet (13), the bottom surface of described rectangular recess (6) is provided with several locating pieces (14), described locating piece (14) below is provided with the first drive motor (15) driving locating piece (14) to rotate, described locating piece (14) is provided with and coordinates with positioning head (11) and to limit the pilot hole (16) that positioning head (11) freely rotates, pilot hole (16) on described several locating pieces (14) is arranged in a linear from front to back, described pilot hole (16) bottom is provided with the 3rd electromagnet (17) and the second proximity switch (33), described sample tray (7) top is provided with optical transceiver (19), described optical transceiver (19) comprises radiation source (20) and optical receiver (21), described rectangular recess (6) top is provided with camera (26), described CPU (central processing unit) (1) respectively with touch-screen (2), light source control module (3), first electromagnet (12), second electromagnet (13), 3rd electromagnet (17), first proximity switch (18), second proximity switch (33), first drive motor (15), optical receiver (21) and camera (26) electrical connection, described light source control module (3) is also electrically connected with radiation source (20).

2. the system of a kind of quick detection splicing beef according to claim 1, it is characterized in that: described positioning head (11) is in flat, the longitudinal section of described positioning head (11) is inverted isosceles trapezoid, the shape of described pilot hole (16) and the form fit of positioning head (11), described pilot hole (16) top is provided with the spigot surface (23) of inclination.

3. the system of a kind of quick detection splicing beef according to claim 1, is characterized in that: the distance between the madial wall of described rectangular recess (6) left and right and the diameter of sample tray (7) match.

4. the system of a kind of quick detection splicing beef according to claim 1 or 2 or 3, it is characterized in that: described light source control module (3) comprises function memory (24) and power amplifier (25), the input end of described function memory (24) is electrically connected with CPU (central processing unit) (1), the output terminal of described function memory (24) is electrically connected with the input end of power amplifier (25), and the output terminal of described power amplifier (25) is electrically connected with radiation source (20).

5. the system of a kind of quick detection splicing beef according to claim 1 or 2 or 3, is characterized in that: the pilot hole (16) that on described rectangular recess (6) bottom surface, distance rectangular recess (6) anteromedial surface is nearest is positioned at immediately below optical transceiver (19).

6. detect a method for splicing beef fast, it is characterized in that, comprise the following steps:

S1: preparation sheet beef sample, is placed on beef sample on sample tray, makes the center being centrally located at sample tray of beef sample;

S2: the instruction that CPU (central processing unit) is inputted by touch-screen according to user, Quality control pallet inserts the pilot hole that user specifies;

S3: the detection light that CPU (central processing unit) sends certain light intensity by light source control module control radiation source is radiated on beef sample, detect light intensity and first rise to maximal value from 0 according to straight line, then 0 is dropped to from maximal value according to sinusoidal curve, optical receiver gathers reflected spectrum data Spect (t), then central processing unit controls sample tray insert pilot hole place locating piece below the first drive motor work, first drive motor controls locating piece and rotates 30 degree, sample tray is made to horizontally rotate 30 degree, optical receiver gathers the spectroscopic data of now radiation source point of irradiation, Quality control pallet horizontal rotary like this three-sixth turn, often horizontally rotate 30 degree of detection spectroscopic datas that stop, thus on beef sample, gather spectroscopic data Spect (t) of 12 diverse locations,

S4: collect 12 spectroscopic datas Spect (t) are all carried out same data processing by CPU (central processing unit), calculate 12 signal to noise ratio (S/N ratio) eigenwerts, comprise the following steps the data processing that each spectroscopic data Spect (t) is carried out:

Adopt input signal as input matrix, potential function V (x, t, α) and input signal are worked in coordination with as one deck accidental resonance model:

Wherein, V (x, t, α) is potential function, and x (t) is Brownian Particles movement locus function, and t is run duration, for periodic sinusoidal signal, N (t) grasps noise in being, A is signal amplitude, and f is signal frequency, and D is external noise intensity, and ξ (t) is external noise, for phase place,

Calculate V (x, t, α) for the first order derivative of x (t), second derivative and three order derivatives, and make equation equal 0, obtain two layers of accidental resonance model:

(2)

Setting noise intensity D=0, spect (t)=0, N (t)=0, the critical value calculating A is by A _csubstitute in one deck accidental resonance model, and set x ₀(t)=0, sn ₀=0, adopt quadravalence jade for asking rain Ge Kuta Algorithm for Solving one deck accidental resonance model, obtain:

$\begin{array}{c}{x}_{n+1}\left(t\right)=x_n\left(t\right)+1/6[{\left(k_1\right)}_n+(2-\sqrt{2}){\left(k_2\right)}_n+(2+\sqrt{2}){\left(k_3\right)}_n+{\left(k_4\right)}_n]\\ n=\mathrm{0,1},...,N-1\end{array},,---\left(3\right)$

Wherein, undetermined coefficient:

(k ₁) _n＝a(αx _n-1(t)) ²b(αx _n-1(t)) ³+sn _n-1(4)

${\left(k_2\right)}_n=a({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{{\left(k_1\right)}_{n-1}}{3})-b{({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{{\left(k_1\right)}_{n-1}}{3})}^3+{\mathrm{sn}}_{n-1}---\left(5\right)$

${\left(k_3\right)}_n=a({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{{\left(k_2\right)}_{n-1}}{3})-b{({\mathrm{\&alpha;x}}_{n-1}\left(t\right)+\frac{\sqrt{2}-1}{3}{\left(k_1\right)}_{n-1}+\frac{2-\sqrt{2}}{3}{\left(k_2\right)}_{n-1})}^3+{\mathrm{sn}}_{n+1}---\left(6\right)$

${\left(k_4\right)}_n=a(3\mathrm{\&alpha;}{x}_{n-1}\left(t\right)+{\left(k_3\right)}_{n-1})-b{({\mathrm{\&alpha;x}}_{n-1}\left(t\right)-\frac{\sqrt{2}}{3}{\left(k_2\right)}_{n-1}+\frac{2+\sqrt{2}}{3}{\left(k_3\right)}_{n-1})}^3+{\mathrm{sn}}_{n+1}---\left(7\right)$

Wherein, x _nt n order derivative that () is x (t), sn _nbe the value of n order derivative at t=0 place of S (t), a, b are the constant of setting,

Calculate x ₁(t), x ₂(t) ... x _n+1t the value of (), to x ₁(t), x ₂(t) ... x _n+1t () is carried out integration and is obtained x (t), and obtain x (t) produces resonance moment position x in the double-deck stochastic resonance system of one deck accidental resonance model and two layers of accidental resonance model composition _mvalue and x _mcorresponding resonance moment t1, the acceleration of motion α 1 corresponding with t1 and the noise D1 corresponding with t1, D1 are a value in D,

Pass through formula $\mathrm{SNR}=\sqrt{2}{\left(\mathrm{\&Delta;U}\frac{{4a}^3/27b}{D_1}\right)}^2{e}^{-{\left(\mathrm{\&Delta;U}\right)}^2/D_1}$ , calculate signal to noise ratio (S/N ratio) eigenwert SNR _feature, wherein, Δ U=a ²/ 4b;

S5: calculate 12 signal to noise ratio (S/N ratio) eigenwerts are divided into four groups, by SNR _{feature 3}, SNR _{feature 6}, SNR _{feature 9}, SNR _{feature 12}be divided into first group, by SNR _{feature 2}, SNR _{feature 4}, SNR _{feature 7}, SNR _{feature 11}be divided into second group, by SNR _{feature 1}and SNR _{feature 5}be divided into the 3rd group, by SNR _{feature 8}and SNR _{feature 10}be divided into the 4th group, and calculate the signal to noise ratio (S/N ratio) mean value of each group respectively, obtain SNR _{average 1}, SNR _{average 2}, SNR _{average 3}, SNR _{average 4},

Calculate the error QE between each signal to noise ratio (S/N ratio) eigenwert and signal to noise ratio (S/N ratio) mean value of the group corresponding to it _j, j=1 ..., 12;

Statistics meets QE _jthe number M of the signal to noise ratio (S/N ratio) eigenwert of≤2% ₁, statistics meets QE _jthe number M of the signal to noise ratio (S/N ratio) eigenwert of > 2% ₂;

S6: if then CPU (central processing unit) judges that this beef sample is not splicing beef, if then CPU (central processing unit) judges that this beef sample is splicing beef, otherwise jumps to step S3, re-starts detection to beef sample.

7. the method for a kind of quick detection splicing beef according to claim 6, it is characterized in that, described step S2 comprises the following steps:

S21: central processing unit controls first work of electromagnet, sample tray is by the first electromagnet suction action to rectangular recess anteromedial surface place, and the first proximity switch sends trigger pip;

S22: CPU (central processing unit) receives the trigger pip that the first proximity switch sends, control the first electromagnet to quit work, the 3rd work of electromagnet bottom the pilot hole that command range rectangular recess anteromedial surface is nearest, 3rd electromagnet produces suction, positioning head bottom sample tray inserts this pilot hole by the suction of the 3rd electromagnet, and the second proximity switch in this pilot hole sends trigger pip;

S23: CPU (central processing unit) receives the trigger pip that the second proximity switch sends, judgement sample pallet is located successfully, and now the central point of sample tray is positioned at immediately below optical transceiver;

S24: CPU (central processing unit) reads the position of positioning hole that user needs by the instruction determination sample tray that touch-screen inputs to insert;

S25: the 3rd electromagnet bottom the pilot hole that central processing unit controls distance rectangular recess anteromedial surface is nearest quits work, and sample tray floats by buoyancy, and the positioning head bottom sample tray leaves pilot hole;

S26: central processing unit controls second electromagnet and sample tray need the 3rd work of electromagnet bottom the pilot hole of insertion, and the positioning head bottom sample tray inserts this pilot hole, and the second proximity switch in this pilot hole sends trigger pip.