CN106228522A - A kind of color concentration treatment of CCD color selector - Google Patents

Abstract

The invention provides the color concentration treatment of a kind of CCD color selector, it comprises the following steps: particle image is sent into image processor and obtains the gray value of image；Determine optimal parameter α, β；The parameter of Genetic Simulated Annealing Algorithm is set；According to population scale n, according to binary mode, individuality is encoded；Optimum individual fitness is stretched；More excellent individual entrance population of future generation；Two random individualities are carried out self adaptation intersection function operation, selects the superior to enter population of future generation；Two random individualities are carried out TSP question function operation, selects the superior to enter population of future generation；Enumerator t=0 is set and minimum new explanation accepts number of times, to each individual simulation annealing operation in new population.The color concentration treatment of the CCD color selector that the present invention provides improves brightness of image, is effectively improved picture contrast, it is possible to obtains more preferably image enhancement effects, more meets image procossing and be actually subjected to.

Description

A kind of color concentration treatment of CCD color selector

Technical field

The invention belongs to CCD color selector technical field, the color concentration treatment of a kind of CCD color selector.

Background technology

At the gatherer process of the image of CCD color selector, due to technological means and environmental disturbances, the image of collection is more or less Including some noises, these noise on image quality have a negative impact, and drop image visual effect.Image enhaucament refers to use certain Technology, carries out respective handling to image, and to improve picture quality, and enhanced picture quality directly affects successive image and divides Analysis, therefore improves the great research topic that image enhancement effects is always in Computer Image Processing.In order to strengthen picture quality, Image enhaucament problem has been carried out deeply, has studied widely by Chinese scholars and research worker, different according to processing space, image Enhancement Method can be divided into frequency domain method, the big class of spatial domain method two.Space domain method is improved by changing image aerial image vegetarian refreshments gray value Picture contrast, specifically includes that linear transformation, nonlinear transformation and histogram equalization etc..Frequency domain method is to revise image Based on Fourier transformation, mainly including wavelet transformation, extra small wave conversion etc., they have multiresolution, select base the most special Point, achieves good application effect in image intensification, and such method is only capable of describing characteristic and the position of the singular point in image Put, it is impossible to " edge " edge feature of accurate description image.Spatial domain rule is so that image pixel to be directly processed as basis, permissible Effectively eliminate the noise in image, but yet suffer from keeping details and the marginal information of image in actual applications, The defects such as image definition is low.

Summary of the invention

For solving above-mentioned technical problem, the invention provides the color concentration treatment of a kind of CCD color selector, it includes following Step:

Obtain particle image by ccd image sensor, and particle image is sent into the ash of image processor acquisition image Angle value f (x, y)；

Definition image processing function is

Wherein α, β are undetermined parameter, obtain function by formula (1)

$B(\mathrm{\α},\mathrm{\β})=\underset{0}{\overset{u}{\∫}}{t}^{a-1}{(1-t)}^{\mathrm{\β}-1}dt---\left(2\right);$

Image intensity value is normalized:

Obtain z ' (x, y)=F [z (x, y)] (4)

G (x, y)=(L_max-L_min) z ' (x, y)+L_min (5)

Simulated annealing and genetic algorithm is utilized to determine optimal parameter α, β；

The parameter of Genetic Simulated Annealing Algorithm is set, mainly includes population scale n, intersect and mutation probability Pc, Pm, maximum Evolutionary generation Tmax, annealing initial temperature t0 and temperature damping's factor k；

According to population scale n, use random fashion to produce initial population, and according to binary mode, individuality is compiled Code, each individuality includes that parameter alpha, β form；

Calculate the fitness value of each individuality, and optimum individual fitness is stretched, specific as follows:

$f^{\′}=\frac{\exp \[f-f_{\max }\]}{T}---\left(6\right)$

Roulette mode is used to select part more excellent individual entrance population of future generation by a certain percentage；

Randomly choose two individualities, two individualities are carried out self adaptation intersection function operation, selects the superior to enter next For population, concrete self adaptation intersection handling function is:

$p_c=\left\{\begin{array}{c}(f_{\max }-f^{\&prime;})/(f_{\max }-f_{avg})f^{\&prime;}\&GreaterEqual;f_{avg}\\ 1f^{\&prime;}<f_{avg}\end{array}\right.$

Randomly choose two individualities, two individualities are carried out TSP question function operation, select the superior to enter next For population, concrete TSP question function is:

$p_c=\left\{\begin{array}{c}(f_{\max }-f^{\&prime;})/s(f_{\max }-f_{avg})f^{\&prime;}\&GreaterEqual;f_{avg}\\ 0.5f^{\&prime;}<f_{avg}\end{array}\right.---\left(8\right)$

Enumerator t=0 is set and minimum new explanation accepts number of times, to each individual simulation annealing behaviour in new population Making, concrete mode is:

Each individuality is carried out disturbance operation, and determines whether to accept new explanation according to formula (2), if accepted, then t= T+1, not so t is constant；If t accepts number of times more than minimum new explanation, individual after annealing operation is replaced fitness worst Body；

If evolutionary generation reaches maximum evolutionary generation, then the individuality of adaptive optimal control angle value is decoded, obtains optimum Parameter alpha, β value；

By optimized parameter α, β value substitute into z ＇ (x, y) in, obtain image through nonlinear gray convert after gray value；

According to z ＇ (x, value y) can get final image enhanced gray value g (x, y).

The method have the advantages that

The color concentration treatment of the CCD color selector that the present invention provides improves brightness of image, is effectively improved image pair Degree of ratio, it is possible to obtain more preferably image enhancement effects, more meet image procossing and be actually subjected to.

Certainly, the arbitrary product implementing the present invention it is not absolutely required to reach all the above advantage simultaneously.

Detailed description of the invention

Below in conjunction with the embodiment of the present invention, the technical scheme in the present invention is clearly and completely described, it is clear that institute The embodiment described is only a part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, All other embodiments that those of ordinary skill in the art are obtained under not making creative work premise, broadly fall into this The scope of bright protection.

The invention provides the color concentration treatment of a kind of CCD color selector, it comprises the following steps:

Obtain particle image by ccd image sensor, and particle image is sent into the ash of image processor acquisition image Angle value f (x, y)

Definition image processing function is

Wherein α, β are undetermined parameter, obtain function by formula (1)

$B(\mathrm{\&alpha;},\mathrm{\&beta;})=\underset{0}{\overset{u}{\&Integral;}}{t}^{a-1}{(1-t)}^{\mathrm{\&beta;}-1}dt---\left(2\right);$

Image intensity value is normalized:

Obtain z ' (x, y)=F [z (x, y)] (4)

G (x, y)=(L_max-L_min) z ' (x, y)+L_min (5)

Simulated annealing and genetic algorithm is utilized to determine optimal parameter α, β；

The parameter of Genetic Simulated Annealing Algorithm is set, mainly includes population scale n, intersect and mutation probability Pc, Pm, maximum Evolutionary generation Tmax, annealing initial temperature t0 and temperature damping's factor k；

According to population scale n, use random fashion to produce initial population, and according to binary mode, individuality is compiled Code, each individuality includes that parameter alpha, β form；

Calculate the fitness value of each individuality, and optimum individual fitness is stretched, specific as follows:

$f^{\&prime;}=\frac{\exp \&lsqb;f-f_{\max }\&rsqb;}{T}---\left(6\right)$

Roulette mode is used to select part more excellent individual entrance population of future generation by a certain percentage；

Randomly choose two individualities, two individualities are carried out self adaptation intersection function operation, selects the superior to enter next For population, concrete self adaptation intersection handling function is:

$p_c=\left\{\begin{array}{c}(f_{\max }-f^{\&prime;})/(f_{\max }-f_{avg})f^{\&prime;}\&GreaterEqual;f_{avg}\\ 1f^{\&prime;}<f_{avg}\end{array}\right.---\left(7\right)$

Randomly choose two individualities, two individualities are carried out TSP question function operation, select the superior to enter next For population, concrete TSP question function is:

$p_c=\left\{\begin{array}{c}(f_{\max }-f^{\&prime;})/s(f_{\max }-f_{avg})f^{\&prime;}\&GreaterEqual;f_{avg}\\ 0.5f^{\&prime;}<f_{avg}\end{array}\right.---\left(8\right)$

Enumerator t=0 is set and minimum new explanation accepts number of times, to each individual simulation annealing behaviour in new population Making, concrete mode is:

Each individuality is carried out disturbance operation, and determines whether to accept new explanation according to formula (2), if accepted, then t= T+1, not so t is constant；If t accepts number of times more than minimum new explanation, individual after annealing operation is replaced fitness worst Body；

If evolutionary generation reaches maximum evolutionary generation, then the individuality of adaptive optimal control angle value is decoded, obtains optimum Parameter alpha, β value；

By optimized parameter α, β value substitute into z ＇ (x, y) in, obtain image through nonlinear gray convert after gray value；

According to z ＇ (x, value y) can get final image enhanced gray value g (x, y).

The color concentration treatment of the CCD color selector that the present invention provides improves brightness of image, is effectively improved image pair Degree of ratio, it is possible to obtain more preferably image enhancement effects, more meet image procossing and be actually subjected to.

Present invention disclosed above preferred embodiment is only intended to help to illustrate the present invention.Preferred embodiment is the most detailed Describe all of details, be also not intended to the detailed description of the invention that this invention is only described.Obviously, according to the content of this specification, Can make many modifications and variations.These embodiments are chosen and specifically described to this specification, is to preferably explain the present invention Principle and actual application so that skilled artisan can be best understood by and utilize the present invention.The present invention is only Limited by claims and four corner thereof and equivalent.

Claims (1)

1. the color concentration treatment of a CCD color selector, it is characterised in that comprise the following steps:

Obtain particle image by ccd image sensor, and particle image is sent into the gray value f of image processor acquisition image (x, y)；

Definition image processing function is

Wherein α, β are undetermined parameter, obtain function by formula (1)

$B(\mathrm{\&alpha;},\mathrm{\&beta;})=\underset{0}{\overset{u}{\&Integral;}}{t}^{a-1}{(1-t)}^{\mathrm{\&beta;}-1}dt---\left(2\right);$

Image intensity value is normalized:

Obtain z ' (x, y)=F [z (x, y)] (4)

$g(x,y)=(L_{\max }-L_{\min })z^{\&prime;}(x,y)+L_{\min }---\left(5\right)$

Simulated annealing and genetic algorithm is utilized to determine optimal parameter α, β；

The parameter of Genetic Simulated Annealing Algorithm is set, mainly includes population scale n, intersect and mutation probability Pc, Pm, maximum evolution Algebraically Tmax, annealing initial temperature t0 and temperature damping's factor k；

According to population scale n, use random fashion to produce initial population, and according to binary mode, individuality is encoded, often Body includes that parameter alpha, β form one by one；

Calculate the fitness value of each individuality, and optimum individual fitness is stretched, specific as follows:

$f^{\&prime;}=\frac{\exp \&lsqb;f-f_{\max }\&rsqb;}{T}---\left(6\right)$

Roulette mode is used to select part more excellent individual entrance population of future generation by a certain percentage；

Randomly choose two individualities, two individualities are carried out self adaptation intersection function operation, select the superior to enter kind of future generation Group, concrete self adaptation intersection handling function is:

$p_c=\left\{\begin{array}{c}\left(f_{\max }-f^{\&prime;}\right)/\left(f_{\max }-f_{avg}\right)f^{\&prime;}\&GreaterEqual;f_{avg}\\ 1f^{\&prime;}<f_{avg}\end{array}\right.---\left(7\right)$

Randomly choose two individualities, two individualities are carried out TSP question function operation, select the superior to enter kind of future generation Group, concrete TSP question function is:

$p_c=\left\{\begin{array}{c}(f_{\max }-f^{\&prime;})/s(f_{\max }-f_{avg})f^{\&prime;}\&GreaterEqual;f_{avg}\\ 0.5f^{\&prime;}<f_{avg}\end{array}\right.---\left(8\right)$

Enumerator t=0 is set and minimum new explanation accepts number of times, to each individual annealing operation that is simulated in new population, tool Body mode is:

Each individuality is carried out disturbance operation, and determines whether to accept new explanation according to formula (2), if accepted, then t=t+1, Not so t is constant；If t accepts number of times more than minimum new explanation, by individuality worst for the individual replacement fitness after annealing operation；

If evolutionary generation reaches maximum evolutionary generation, then the individuality of adaptive optimal control angle value is decoded, obtains optimized parameter α, β value；

By optimized parameter α, β value substitute into z ＇ (x, y) in, obtain image through nonlinear gray convert after gray value；

According to z ＇ (x, value y) can get final image enhanced gray value g (x, y).