CN105088595A - Printing and dyeing on-line color matching equipment and method based on neural network model - Google Patents

Abstract

The invention discloses printing and dyeing on-line color matching equipment based on a neural network model. The equipment comprises a standard color atla acquisition module, a dye pool image acquisition module, an ultrasonic liquid level measuring instrument, a data processing module and a flow solenoid valve, wherein the dye pool image acquisition module is used for collecting actual color images of a dye pool to be processed; the ultrasonic liquid level measuring instrument is used for measuring the height of the dye pool to be processed and calculating and obtaining the volume of dye liquid which already exists in the dye pool to be processed according to the height; the data processing module is used for calculating color differences between the actual color images of the dye pool to be processed and standard color atla images and determining flow values required by dye in all colors according to the color differences and the volume of the dye pool to be processed and based on the neural network model; flow solenoid valve is used for utilizing the fuzzy control technology for controlling dye flow according to the flow values required by the dye in all colors so that the dye on-line distribution can be achieved, and the actual color of the dye pool to be processed after stirring is made to satisfy the color requirements of the standard color atla. The invention further discloses a printing and dyeing on-line color matching method based on the neural network model. By means of the equipment and method, the color matching accuracy is improved, and the human cost is reduced.

Description

The online color matching apparatus of printing and dyeing based on neural network model and method

Technical field

The present invention relates to technical field of image processing, particularly relate to the online color matching apparatus of a kind of printing and dyeing based on neural network model and method.

Background technology

In field of printing and dyeing, the development of color matching instrument is the focus of research always, but due to condition restriction, manufacturing enterprises a lot of so far still mainly relies on visual judgement and color matching experience to complete the color blending technique of printing and dyeing.But this kind of method itself also exists certain human error risk, and the color matching skilled worker quantity of printworks in recent years just gradually reduces, the human cost of this work is improved, the color matching work of high strength also makes the artificial collimation error be in higher level simultaneously, have impact on the dyeing and printing products quality of manufacturing enterprise and the further lifting of shipment efficiency to a certain extent, there is many drawbacks.And the color of dyestuff with use the dye colour in raw material and current dye cell to have certain relation, but do not have mathematical models to adopt.

Summary of the invention

Technical problem to be solved by this invention is, provides the online color matching apparatus of a kind of printing and dyeing based on neural network model and method, improves color matching accuracy rate, improves dyeing and printing products quality and shipment efficiency, has saved human cost.

In order to solve the problems of the technologies described above, the invention provides the online color matching apparatus of a kind of printing and dyeing based on neural network model, comprising:

Standard color card acquisition module, for gathering standard color card image;

Dye cell image capture module, is arranged at pending dye cell surrounding, for gathering the actual color image of pending dye cell;

Ultrasonic Liquid Level Indicator, is arranged on pending dye cell, for measuring the liquid level of pending dye cell, and obtains the volume of existing dye liquid in pending dye cell according to described high computational;

Data processing module, be connected with described standard color card acquisition module, described dye cell color acquisition module and described Ultrasonic Liquid Level Indicator respectively, for calculating the aberration of pending dye cell actual color image and standard color card image, and determine flow value needed for assorted dyestuff according to the volume of described aberration and described pending dye cell based on neural network model;

Flow solenoid valve, be connected with described data processing module, utilize fuzzy control technology to control dyestuff flow for flow value needed for the assorted dyestuff determined according to described data processing module to realize dyestuff and distribute online, after making to stir, pending dye cell actual color meets standard color card color requirements.

Further, described standard color card acquisition module comprises standard color card and gathers draw-in groove and gather with described standard color card the high resolution industrial camera that draw-in groove is connected; Described dye cell color acquisition module is specially high resolution industrial camera; Described data processing module is specially computer.

Present invention also offers the online color matching method of a kind of printing and dyeing based on neural network model, described method is based on above-mentioned color matching apparatus, then described method comprises step:

S101, experimentally room toning experimental data sets up toning scheme neural network model;

S102, standard color card inserted gather draw-in groove, by camera collection standard color card color image, and be sent to computer and carry out preliminary treatment and RGB and analyze;

The actual color image of the pending dye cell of camera Real-time Collection of S103, pending dye cell surrounding, is sent to computer and carries out preliminary treatment and RGB analysis;

S104, Ultrasonic Liquid Level Indicator measure the liquid level of existing dye liquid in pending dye cell, and according to having the volume of dye liquid in the pending dye cell of the areal calculation of pending dye cell;

S105, difference processing is carried out to the rgb value of the standard color card obtained respectively after step S102 and S103 process and pending dye cell;

S106, volume according to dye liquid existing in difference result and pending dye cell, be input to the toning scheme neural network model determined in step S101, calculate the required flow value of single toning that red, green, blue three kinds of dyestuffs supplement respectively;

S107, the required flow value of single toning supplemented respectively according to red, green, blue three kinds of dyestuffs of calculating, use fuzzy control technology to control flow solenoid valve and realize feed, and control mixer and stir;

S108, stirred after, repeat step S102-S107, until dye colour meets standard color card color requirements in pending dye cell.

Further, described S101 specifically comprises step:

The experiment of toning repeatedly of S1011, by experiment room obtains one group of toning data as training sample X=[x ₁, x ₂..., x _n] and Y=[y ₁, y ₂..., y _n], wherein, n represents the number of sample, x _i=[r _d, g _d, b _d, v], i=1,2,3 ..., n, r _d, g _d, b _drepresent the chromaticity coordinate of rgb value, v is dye liquid volume, y _j=[y _red, y _green, y _blue], j=1,2,3 ..., n, y _red, y _green, y _bluerepresent flow value needed for red, green, blue three look dyestuff respectively;

S1012, according to formula x _n=2* (x _i-minx _i)/(maxx _i-minx _i)-1 pair of training sample be normalized, initialize weights, the threshold value between hidden layer and input layer, produce symmetrical random number to the neuron with tangent S type activation primitive; Wherein, maxx _i, minx _ibe respectively x _imaximum and minimum of a value, x _nfor the data after mapping;

S1013, take X as input, Y is desired output, setting deviation e=1e ^-5, initialized by training parameter, adopt rapid bp algorithm neural network training, training obtains the connection weights of in the middle of neutral net three layers, determines the toning scheme neural network model of this batch of red, green, blue three look dyestuff.

Further, described S102 specifically comprises step:

S1021, by standard color card insert gather draw-in groove, by camera collection standard color card color image P _s, and be sent to computer;

S1022, computer adopt median filtering method, remove standard color card color image P _snoise, obtain muting standard color card image P _s';

S1023, calculating P _s' height H _s;

S1024, successively from image P _s' height locate from left to right scan image, wherein n is positive integer, i _s=1,3,5 ..., 2n-1;

S1025, rgb value analysis is carried out to each pixel searched after average and obtain value, in this, as the rgb value of standard color card.

Further, described S103 specifically comprises step:

S1031, pending dye cell four different industrial camera Real-time Collection pending dye cell actual color image Thursday, be respectively P _r1, P _r2, P _r3, P _r4, and be sent to computer;

S1032, computer adopt median filtering method, remove the noise of the actual color image of pending dye cell, obtain the actual color image P of muting pending dye cell _r1', P _r2', P _r3' and P _r4';

S1033, successively from image P _r1', P _r2', P _r3' and P _r4' height locate from left to right scan image, wherein n is positive integer, i _r=1,3,5 ..., 2n-1;

S1034, rgb value analysis is carried out to each pixel searched after average, obtain image P _r1', P _r2', P _r3' and P _r4' value, wherein, i=1,2,3,4;

S1035, according to formula: $\stackrel{\‾}{R_R}=\frac{\stackrel{\‾}{R_{R1}}+\stackrel{\‾}{R_{R2}}+\stackrel{\‾}{R_{R3}}+\stackrel{\‾}{R_{R4}}}{4},\stackrel{\‾}{G_R}=\frac{\stackrel{\‾}{G_{R1}}+\stackrel{\‾}{G_{R2}}+\stackrel{\‾}{G_{R3}}+\stackrel{\‾}{G_{R4}}}{4},$ calculate the RGB mean value of the actual color image of pending dye cell value.

Further, described S105 specifically comprises step:

S1051, by standard color card image P _srGB mean value color image P real-time with dye cell _rrGB mean value carry out difference processing, obtain difference

S1052, to solve each leisure relative scale in total amount, is used chromaticity coordinate r _d, g _d, b _drepresent, chromaticity coordinate computing formula is specially:

$r_D=\frac{{\stackrel{\‾}{R}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D},g_D=\frac{{\stackrel{\‾}{G}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D},b_D=\frac{{\stackrel{\‾}{B}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D}.$

Further, described S107 specifically comprises step:

S1071, for the control of orchil flow solenoid valve, it is F that the flow sensor module of orchil flow solenoid valve measures the dyestuff flow flowing through flow solenoid valve in real time _red.

The micro controller module of S1072, orchil flow solenoid valve with the flow of dyestuff for control object, with target flow V _redwith measured discharge F _rederror be deviation e input quantity, the rate of change of deviation e as ec input quantity, motor rotate angle be microcontroller output quantity u;

S1073, to arrange fuzzy variable corresponding to e be fuzzy variable that E, ec are corresponding be the fuzzy variable that EC, u are corresponding is U, and the fringe of 3 variablees is selected as follows:

E＝{NB，NM，NS，NZ，PZ，PS，PM，PB}；

EC＝{NB，NM，NS,Z，PS，PM，PB}；

U＝{NB，NM，NS，Z，PS，PM，PB}。

S1074, the grade classification of each fuzzy variable in its domain be set be respectively:

E＝{e}＝{-6，-4，-2，0，2，4，6}；

EC＝{ec}＝{-6，-4，-2，0，2，4，6}；

U＝{u}＝{-6，-4，-2，0，2，4，6}。

S1075, the membership function arranging each fuzzy variable are triangle continuous type membership function, and defuzzification adopts maximum membership degree mean value method, generate fuzzy controller.

The micro controller module of S1076, orchil flow solenoid valve, using the rate of change of deviation e and deviation e as input quantity, according to the fuzzy controller determination microcontroller output quantity u generated, control the orchil that flow solenoid valve pours into corresponding flow in dye cell.

Implement the present invention, there is following beneficial effect:

(1), the present invention is based on machine vision technique and carry out standard color card and pending dye cell colour recognition, the requirement of dyeing and printing products color change can be realized fast, carry out the collection of pending dye cell color in real time, avoid the human error and high human cost that rely on visual judgement to also exist;

(2) printing and dyeing that, the invention provides based on neural network model are mixed colours and dyestuff distribution method and equipment online, utilize neural network toning model, the color accuracy of dyestuff greatly can be improved in color matching operation, promote dyeing and printing products quality and shipment efficiency, be convenient to the further expansion of automated production scale;

(3), the present invention's control of utilizing fuzzy control technology to realize final batching, according to sensor information, in conjunction with the output data of neural network toning model, the degree of accuracy and the batching efficiency of final toning can be improved further.

(4), the present invention proposes the machine vision method that a kind of printing and dyeing are newly matched colors and prepared burden, namely computer mould apery class visual identity color of image difference is utilized, the formulation of scheme of colour is carried out by extracting aberration, there is intelligent, the flexibility of height, significantly improve the degree of accuracy of printing and dyeing color matching and batching, there is wide market application foreground.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

Fig. 1 is the schematic diagram of an embodiment of the online color matching apparatus of the printing and dyeing based on neural network model provided by the invention;

Fig. 2 is the schematic flow sheet of an embodiment of the online color matching method of the printing and dyeing based on neural network model that the present invention also provides.

Detailed description of the invention

Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

Fig. 1 is the schematic diagram of an embodiment of the online color matching apparatus of the printing and dyeing based on neural network model provided by the invention, as shown in Figure 1, comprising:

Standard color card acquisition module, for gathering standard color card image;

Dye cell image capture module, is arranged at pending dye cell surrounding, for gathering the actual color image of pending dye cell;

Ultrasonic Liquid Level Indicator, is arranged on pending dye cell, for measuring the liquid level of pending dye cell, and obtains the volume of existing dye liquid in pending dye cell according to described high computational;

Data processing module, be connected with described standard color card acquisition module, described dye cell color acquisition module and described Ultrasonic Liquid Level Indicator respectively, for calculating the aberration of pending dye cell actual color image and standard color card image, and determine flow value needed for assorted dyestuff according to the volume of described aberration and described pending dye cell based on neural network model;

Flow solenoid valve, be connected with described data processing module, utilize fuzzy control technology to control dyestuff flow for flow value needed for the assorted dyestuff determined according to described data processing module to realize dyestuff and distribute online, after making to stir, pending dye cell actual color meets standard color card color requirements.

Wherein, described standard color card acquisition module comprises standard color card and gathers draw-in groove and gather with described standard color card the high resolution industrial camera that draw-in groove is connected; Described dye cell color acquisition module is specially high resolution industrial camera; Described data processing module is specially computer.

Fig. 2 is the schematic flow sheet of an embodiment of the online color matching method of the printing and dyeing based on neural network model that the present invention also provides, and described method is based on above-mentioned color matching apparatus, then described method comprises step:

S101, experimentally room toning experimental data sets up toning scheme neural network model.

Concrete, described S101 specifically comprises step:

The experiment of toning repeatedly of S1011, by experiment room obtains one group of toning data as training sample X=[x ₁, x ₂..., x _n] and Y=[y ₁, y ₂..., y _n], wherein, n represents the number of sample, x _i=[r _d, g _d, b _d, v], i=1,2,3 ..., n, r _d, g _d, b _drepresent the chromaticity coordinate of rgb value, v is dye liquid volume, y _j=[y _red, y _green, y _blue], j=1,2,3 ..., n, y _red, y _green, y _bluerepresent flow value needed for red, green, blue three look dyestuff respectively;

S1012, according to formula x _n=2* (x _i-minx _i)/(maxx _i-minx _i)-1 pair of training sample be normalized, initialize weights, the threshold value between hidden layer and input layer, produce symmetrical random number to the neuron with tangent S type activation primitive; Wherein, maxx _i, minx _ibe respectively x _imaximum and minimum of a value, x _nfor the data after mapping;

S1013, take X as input, Y is desired output, setting deviation e=1e ^-5, initialized by training parameter, adopt rapid bp algorithm neural network training, training obtains the connection weights of in the middle of neutral net three layers, determines the toning scheme neural network model of this batch of red, green, blue three look dyestuff.

S102, standard color card inserted gather draw-in groove, by camera collection standard color card color image, and be sent to computer and carry out preliminary treatment and RGB and analyze.

Concrete, described S102 specifically comprises step:

S1021, by standard color card insert gather draw-in groove, by camera collection standard color card color image P _s, and be sent to computer;

S1022, computer adopt median filtering method, remove standard color card color image P _snoise, obtain muting standard color card image P _s';

S1023, calculating P _s' height H _s;

S1024, successively from image P _s' height locate from left to right scan image, wherein n is positive integer, i _s=1,3,5 ..., 2n-1;

S1025, rgb value analysis is carried out to each pixel searched after average and obtain value, in this, as the rgb value of standard color card.

The actual color image of the pending dye cell of camera Real-time Collection of S103, pending dye cell surrounding, is sent to computer and carries out preliminary treatment and RGB analysis.

Concrete, described S103 specifically comprises step:

S1031, pending dye cell four different industrial camera Real-time Collection pending dye cell actual color image Thursday, be respectively P _r1, P _r2, P _r3, P _r4, and be sent to computer;

S1032, computer adopt median filtering method, remove the noise of the actual color image of pending dye cell, obtain the actual color image P of muting pending dye cell _r1', P _r2', P _r3' and P _r4';

S1033, successively from image P _r1', P _r2', P _r3' and P _r4' height locate from left to right scan image, wherein n is positive integer, i _r=1,3,5 ..., 2n-1;

S1034, rgb value analysis is carried out to each pixel searched after average, obtain image P _r1', P _r2', P _r3' and P _r4' value, wherein, i=1,2,3,4;

S1035, according to formula: $\stackrel{\‾}{R_R}=\frac{\stackrel{\‾}{R_{R1}}+\stackrel{\‾}{R_{R2}}+\stackrel{\‾}{R_{R3}}+\stackrel{\‾}{R_{R4}}}{4},\stackrel{\‾}{G_R}=\frac{\stackrel{\‾}{G_{R1}}+\stackrel{\‾}{G_{R2}}+\stackrel{\‾}{G_{R3}}+\stackrel{\‾}{G_{R4}}}{4},$ calculate the RGB mean value of the actual color image of pending dye cell value.

S104, Ultrasonic Liquid Level Indicator measure the height of existing dye liquid in pending dye cell, and according to having the volume of dye liquid in the pending dye cell of the areal calculation of pending dye cell.

S105, difference processing is carried out to the rgb value of the standard color card obtained respectively after step S102 and S103 process and pending dye cell.

Concrete, described S105 specifically comprises step:

S1051, by standard color card image P _srGB mean value color image P real-time with dye cell _rrGB mean value carry out difference processing, obtain difference

S1052, to solve each leisure relative scale in total amount, is used chromaticity coordinate r _d, g _d, b _drepresent, chromaticity coordinate computing formula is specially:

$r_D=\frac{{\stackrel{\‾}{R}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D},g_D=\frac{{\stackrel{\‾}{G}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D},b_D=\frac{{\stackrel{\‾}{B}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D}.$

S106, volume according to dye liquid existing in difference result and pending dye cell, be input to the toning scheme neural network model determined in step S101, calculate the required flow value of single toning that red, green, blue three kinds of dyestuffs supplement respectively;

S107, the required flow value of single toning supplemented respectively according to red, green, blue three kinds of dyestuffs of calculating, use fuzzy control technology to control flow solenoid valve and realize feed, and control mixer and stir.

Concrete, described S107 specifically comprises step:

S1071, for the control of orchil flow solenoid valve, it is F that the flow sensor module of orchil flow solenoid valve measures the dyestuff flow flowing through flow solenoid valve in real time _red.

The micro controller module of S1072, orchil flow solenoid valve with the flow of dyestuff for control object, with target flow V _redwith measured discharge F _rederror be deviation e input quantity, the rate of change of deviation e as ec input quantity, motor rotate angle be microcontroller output quantity u;

S1073, to arrange fuzzy variable corresponding to e be fuzzy variable that E, ec are corresponding be the fuzzy variable that EC, u are corresponding is U, and the fringe of 3 variablees is selected as follows:

E＝{NB，NM，NS，NZ，PZ，PS，PM，PB}；

EC＝{NB，NM，NS,Z，PS，PM，PB}；

U＝{NB，NM，NS，Z，PS，PM，PB}。

S1074, the grade classification of each fuzzy variable in its domain be set be respectively:

E＝{e}＝{-6，-4，-2，0，2，4，6}；

EC＝{ec}＝{-6，-4，-2，0，2，4，6}；

U＝{u}＝{-6，-4，-2，0，2，4，6}。

S1075, the membership function arranging each fuzzy variable are triangle continuous type membership function, and defuzzification adopts maximum membership degree mean value method, generate fuzzy controller.

The micro controller module of S1076, orchil flow solenoid valve, using the rate of change of deviation e and deviation e as input quantity, according to the fuzzy controller determination microcontroller output quantity u generated, control the orchil that flow solenoid valve pours into corresponding flow in dye cell.

S108, stirred after, repeat step S102-S107, until dye colour meets standard color card color requirements in pending dye cell.

Implement the present invention, there is following beneficial effect:

(1), the present invention is based on machine vision technique and carry out standard color card and pending dye cell colour recognition, the requirement of dyeing and printing products color change can be realized fast, carry out the collection of pending dye cell color in real time, avoid the human error and high human cost that rely on visual judgement to also exist;

(2) printing and dyeing that, the invention provides based on neural network model are mixed colours and dyestuff distribution method and equipment online, utilize neural network toning model, the color accuracy of dyestuff greatly can be improved in color matching operation, promote dyeing and printing products quality and shipment efficiency, be convenient to the further expansion of automated production scale;

(3), the present invention's control of utilizing fuzzy control technology to realize final batching, according to sensor information, in conjunction with the output data of neural network toning model, the degree of accuracy and the batching efficiency of final toning can be improved further.

(4), the present invention proposes the machine vision method that a kind of printing and dyeing are newly matched colors and prepared burden, namely computer mould apery class visual identity color of image difference is utilized, the formulation of scheme of colour is carried out by extracting aberration, there is intelligent, the flexibility of height, significantly improve the degree of accuracy of printing and dyeing color matching and batching, there is wide market application foreground.

It should be noted that, in this article, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or device and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or device.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the device comprising this key element and also there is other identical element.

The invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.

In several embodiments that the application provides, should be understood that, disclosed system and method can realize by another way.Such as, system embodiment described above is only schematic, such as, the division of described unit, be only a kind of logic function to divide, actual can have other dividing mode when realizing, such as multiple unit or assembly can in conjunction with or another system can be integrated into, or some features can be ignored, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be by some interfaces, and the INDIRECT COUPLING of device or unit or communication connection can be electrical, machinery or other form.

Professional can also recognize further, in conjunction with unit and the algorithm steps of each example of embodiment disclosed herein description, can realize with electronic hardware, computer software or the combination of the two, in order to the interchangeability of hardware and software is clearly described, generally describe composition and the step of each example in the above description according to function.These functions perform with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme.Professional and technical personnel can use distinct methods to realize described function to each specifically should being used for, but this realization should not thought and exceeds scope of the present invention.

The software module that the method described in conjunction with embodiment disclosed herein or the step of algorithm can directly use hardware, processor to perform, or the combination of the two is implemented.Software module can be placed in the storage medium of other form any known in random access memory (RAM), internal memory, read-only storage (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field.

To the above-mentioned explanation of the disclosed embodiments, professional and technical personnel in the field are realized or uses the present invention.To be apparent for those skilled in the art to the multiple amendment of these embodiments, General Principle as defined herein can without departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention can not be restricted to these embodiments shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.

Claims (8)

1., based on the online color matching apparatus of printing and dyeing of neural network model, it is characterized in that, comprising:

Standard color card acquisition module, for gathering standard color card image;

Dye cell image capture module, is arranged at pending dye cell surrounding, for gathering the actual color image of pending dye cell;

Ultrasonic Liquid Level Indicator, is arranged on pending dye cell, for measuring the liquid level of pending dye cell, and obtains the volume of existing dye liquid in pending dye cell according to described high computational;

Data processing module, be connected with described standard color card acquisition module, described dye cell color acquisition module and described Ultrasonic Liquid Level Indicator respectively, for calculating the aberration of pending dye cell actual color image and standard color card image, and determine flow value needed for assorted dyestuff according to the volume of described aberration and described pending dye cell based on neural network model;

Flow solenoid valve, be connected with described data processing module, utilize fuzzy control technology to control dyestuff flow for flow value needed for the assorted dyestuff determined according to described data processing module to realize dyestuff and distribute online, after making to stir, pending dye cell actual color meets standard color card color requirements.

2., as claimed in claim 1 based on the online color matching apparatus of printing and dyeing of neural network model, it is characterized in that,

Described standard color card acquisition module comprises standard color card and gathers draw-in groove and gather with described standard color card the high resolution industrial camera that draw-in groove is connected;

Described dye cell color acquisition module is specially high resolution industrial camera;

Described data processing module is specially computer.

3. based on the online color matching method of printing and dyeing of neural network model, it is characterized in that, described method is based on color matching apparatus according to claim 2, then described method comprises step:

S101, experimentally room toning experimental data sets up toning scheme neural network model;

S102, standard color card inserted gather draw-in groove, by camera collection standard color card color image, and be sent to computer and carry out preliminary treatment and RGB and analyze;

The actual color image of the pending dye cell of camera Real-time Collection of S103, pending dye cell surrounding, is sent to computer and carries out preliminary treatment and RGB analysis;

S104, Ultrasonic Liquid Level Indicator measure the liquid level of existing dye liquid in pending dye cell, and according to having the volume of dye liquid in the pending dye cell of the areal calculation of pending dye cell;

S105, difference processing is carried out to the rgb value of the standard color card obtained respectively after step S102 and S103 process and pending dye cell;

S106, volume according to dye liquid existing in difference result and pending dye cell, be input to the toning scheme neural network model determined in step S101, calculate the required flow value of single toning that red, green, blue three kinds of dyestuffs supplement respectively;

S107, the required flow value of single toning supplemented respectively according to red, green, blue three kinds of dyestuffs of calculating, use fuzzy control technology to control flow solenoid valve and realize feed, and control mixer and stir;

S108, stirred after, repeat step S102-S107, until dye colour meets standard color card color requirements in pending dye cell.

4., as claimed in claim 3 based on the online color matching method of printing and dyeing of neural network model, it is characterized in that, described S101 specifically comprises step:

The experiment of toning repeatedly of S1011, by experiment room obtains one group of toning data as training sample X=[x ₁, x ₂..., x _n] and Y=[y ₁, y ₂..., y _n], wherein, n represents the number of sample, x _i=[r _d, g _d, b _d, v], i=1,2,3 ..., n, r _d, g _db _drepresent the chromaticity coordinate of rgb value, v is dye liquid volume, y _j=[y _red, y _green, y _blue], j=1,2,3 ..., n, y _red, y, _greeny _bluerepresent flow value needed for red, green, blue three look dyestuff respectively;

S1012, according to formula x _n=2* (x _i-minx _i)/(maxx _i-minx _i)-1 pair of training sample be normalized, initialize weights, the threshold value between hidden layer and input layer, produce symmetrical random number to the neuron with tangent S type activation primitive; Wherein, maxx _i, minx _ibe respectively x _imaximum and minimum of a value, x _nfor the data after mapping;

S1013, take X as input, Y is desired output, setting deviation e=1e ^-5, initialized by training parameter, adopt rapid bp algorithm neural network training, training obtains the connection weights of in the middle of neutral net three layers, determines the toning scheme neural network model of this batch of red, green, blue three look dyestuff.

5., as claimed in claim 3 based on the online color matching method of printing and dyeing of neural network model, it is characterized in that, described S102 specifically comprises step:

S1021, by standard color card insert gather draw-in groove, by camera collection standard color card color image P _s, and be sent to computer;

S1022, computer adopt median filtering method, remove standard color card color image P _snoise, obtain muting standard color card image P _s';

S1023, calculating P _s' height H _s;

S1024, successively from image P _s' height locate from left to right scan image, wherein n is positive integer, i _s=1,3,5 ..., 2n-1;

S1025, rgb value analysis is carried out to each pixel searched after average and obtain value, in this, as the rgb value of standard color card.

6., as claimed in claim 3 based on the online color matching method of printing and dyeing of neural network model, it is characterized in that, described S103 specifically comprises step:

S1031, pending dye cell four different industrial camera Real-time Collection pending dye cell actual color image Thursday, be respectively P _r1, P _r2, P _r3, P _r4, and be sent to computer;

S1032, computer adopt median filtering method, remove the noise of the actual color image of pending dye cell, obtain the actual color image P of muting pending dye cell _r1', P _r2', P _r3' and P _r4';

S1033, successively from image P _r1', P _r2', P _r3' and P _r4' height locate from left to right scan image, wherein n is positive integer, i _r=1,3,5 ..., 2n-1;

S1034, rgb value analysis is carried out to each pixel searched after average, obtain image P _r1', P _r2', P _r3' and P _r4' value, wherein, i=1,2,3,4;

S1035, according to formula: $\stackrel{\‾}{R_R}=\frac{\stackrel{\‾}{R_{R1}}+\stackrel{\‾}{R_{R2}}+\stackrel{\‾}{R_{R3}}+\stackrel{\‾}{R_{R4}}}{4},$ $\stackrel{\‾}{G_R}=\frac{\stackrel{\‾}{G_{R1}}+\stackrel{\‾}{G_{R2}}+\stackrel{\‾}{G_{R3}}+\stackrel{\‾}{G_{R4}}}{4},$ calculate the RGB mean value of the actual color image of pending dye cell value.

7., as claimed in claim 6 based on the online color matching method of printing and dyeing of neural network model, it is characterized in that, described S105 specifically comprises step:

S1051, by standard color card image P _srGB mean value color image P real-time with dye cell _rrGB mean value carry out difference processing, obtain difference

S1052, to solve each leisure relative scale in total amount, is used chromaticity coordinate r _d, g _d, b _drepresent, chromaticity coordinate computing formula is specially:

$r_D=\frac{{\stackrel{\‾}{R}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D},$ $g_D=\frac{{\stackrel{\‾}{G}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D},$ $b_D=\frac{{\stackrel{\‾}{B}}_D}{{\stackrel{\‾}{R}}_D+{\stackrel{\‾}{G}}_D+{\stackrel{\‾}{B}}_D}.$

8., as claimed in claim 3 based on the online color matching method of printing and dyeing of neural network model, it is characterized in that, described S107 specifically comprises step:

S1071, for the control of orchil flow solenoid valve, it is F that the flow sensor module of orchil flow solenoid valve measures the dyestuff flow flowing through flow solenoid valve in real time _red.

The micro controller module of S1072, orchil flow solenoid valve with the flow of dyestuff for control object, with target flow V _redwith measured discharge F _rederror be deviation e input quantity, the rate of change of deviation e as ec input quantity, motor rotate angle be microcontroller output quantity u;

S1073, to arrange fuzzy variable corresponding to e be fuzzy variable that E, ec are corresponding be the fuzzy variable that EC, u are corresponding is U, and the fringe of 3 variablees is selected as follows:

E＝{NB，NM，NS，NZ，PZ，PS，PM，PB}；

EC＝{NB，NM，NS,Z，PS，PM，PB}；

U＝{NB，NM，NS，Z，PS，PM，PB}。

S1074, the grade classification of each fuzzy variable in its domain be set be respectively:

E＝{e}＝{-6，-4，-2，0，2，4，6}；

EC＝{ec}＝{-6，-4，-2，0，2，4，6}；

U＝{u}＝{-6，-4，-2，0，2，4，6}。

S1075, the membership function arranging each fuzzy variable are triangle continuous type membership function, and defuzzification adopts maximum membership degree mean value method, generate fuzzy controller.

The micro controller module of S1076, orchil flow solenoid valve, using the rate of change of deviation e and deviation e as input quantity, according to the fuzzy controller determination microcontroller output quantity u generated, control the orchil that flow solenoid valve pours into corresponding flow in dye cell.