WO2016000331A1

WO2016000331A1 - Image enhancement method, image enhancement device and display device

Info

Publication number: WO2016000331A1
Application number: PCT/CN2014/087517
Authority: WO
Inventors: 张晓�; 于淑环; 张丽杰
Original assignee: 京东方科技集团股份有限公司; 北京京东方视讯科技有限公司
Priority date: 2014-06-30
Filing date: 2014-09-26
Publication date: 2016-01-07
Also published as: CN104091311A

Abstract

The present disclosure provides an image enhancement method, an image enhancement device and a display device. The image enhancement method comprises that: an original image is transferred from a first color space to a second color space containing brightness information, and the brightness information of the original image is extracted; for the brightness information, an optimum transformation parameter value of the grey level transformation function for image enhancement is determined through an optimum search algorithm; a transformation curve of the grey level transformation function is constructed according to the optimum transformation parameter value, the brightness information is processed for enhancement according to the transformation curve, and the enhanced brightness information is obtained; the enhanced brightness information is transferred from the second color space to the first color space, and a final enhanced image is obtained.

Description

Image enhancement method, image enhancement device and display device

Cross-reference to related applications

The present application claims priority to Chinese Patent Application No. 201410307041.5, filed on Jun. 30, 2014, the entire content of

Technical field

The present disclosure relates to the field of image processing technologies, and in particular, to an image enhancement method, an image enhancement device, and a display device.

Background technique

The image may be affected by factors such as the dynamic range of the imaging device and the intensity of the ambient light during the acquisition process, resulting in low contrast, inconspicuous image information, color distortion, target contour or boundary information clarity. Such phenomena, which make it difficult for human visual observation and machine analysis processing, need to enhance the image.

Image enhancement refers to a method of highlighting certain information of an image according to a specific need while weakening or removing certain unwanted information. Image enhancement is the most basic means of image processing, and is often a pre-processing process for various image analysis and processing.

Specifically, image enhancement usually performs gradation transformation on the luminance information of an image through a gradation transformation function, but the transformation parameters of the gradation transformation function are usually set by empirical values or subjective observations of the human eye, and thus there is a lack thereof. Universal technical issues.

Summary of the invention

In view of this, the present disclosure provides an image enhancement method, an image enhancement apparatus, and a display apparatus, which realize adaptive enhancement of an image by automatically determining an optimal transformation parameter of a gradation transformation function.

To solve the above technical problem, the present disclosure provides an image enhancement method, including:

Converting an original image from a first color space to a second color space containing luminance information, and extracting luminance information of the original image;

For the brightness information, using a optimization algorithm to determine a gray-scale transformation function for image enhancement Optimal transformation parameter value;

Constructing a transformation curve of the gradation transformation function according to the optimal transformation parameter value, and performing enhancement processing on the luminance information according to the transformation curve to obtain enhanced luminance information;

The enhanced luminance information is converted from the second color space to the first color space to obtain a final enhanced image.

Optionally, the step of determining an optimal transform parameter value of the gray-scale transform function for the image enhancement by using the optimization algorithm specifically includes:

Establish the objective function,

The objective function is used to determine whether the current transformation parameter value is an optimal transformation parameter value, and the optimal transformation parameter value of the gradation transformation function is determined.

Optionally, the optimization algorithm is a particle swarm optimization algorithm, and the objective function is:

Where, Fitness is the current fitness value of the particle, M is the width of the image to be processed, N is the height of the image to be processed, and f(x, y) is the brightness information of the pixel at the position (x, y).

Optionally, the optimization algorithm is a genetic algorithm or a simulated annealing algorithm.

Optionally, the gradation transformation function is a non-complete Beta function;

The formula for the incomplete Beta function is:

Where B(α, β) is a Beta function,

α, β is the transformation parameter of the incomplete Beta function, t is the integral variable of the incomplete Beta function, and u is the upper limit of the range of variation of the integral variable t.

Optionally, the step of determining an optimal transform parameter value for the gray-scale transform function of the image enhancement by using the optimization algorithm further comprises: normalizing the luminance information of the original image or the original image ;

The method further includes performing inverse normalization processing on the enhanced luminance information or the final enhanced image.

The present disclosure also provides an image enhancement apparatus, including:

a first image conversion unit, configured to convert the original image from the first color space to a second color space containing the brightness information, and extract brightness information of the original image;

a transformation parameter optimization unit, configured to determine, by using a optimization algorithm, an optimal transformation parameter value of the gradation transformation function for image enhancement for the brightness information;

An enhancement processing unit, configured to construct a transformation curve of the gradation transformation function according to the optimal transformation parameter value, and perform enhancement processing on the luminance information according to the transformation curve to obtain enhanced luminance information;

The second image conversion unit is configured to convert the enhanced brightness information from the second color space to the first color space to obtain a final enhanced image.

Optionally, the transformation parameter optimization unit is specifically configured to establish an objective function, and use the objective function to determine whether the current transformation parameter value is an optimal transformation parameter value, and determine an optimal transformation of the gray transformation function. Parameter value.

Optionally, the image enhancement device further includes:

a normalization processing unit, configured to normalize brightness information of the original image or the original image;

The inverse normalization processing unit is configured to perform inverse normalization processing on the enhanced luminance information or the final enhanced image.

The present disclosure also provides a display device including the above image enhancement device.

The beneficial effects of the above technical solutions of the present disclosure are as follows:

When the image is enhanced by the gradation transformation function, the optimal transformation parameter of the gradation transformation function can be automatically determined by the optimization algorithm, thereby realizing the adaptive enhancement of the image.

DRAWINGS

1 is a schematic flow chart of an image enhancement method according to an embodiment of the present disclosure;

2 is another schematic flowchart of an image enhancement method according to an embodiment of the present disclosure;

FIG. 3 is a structural block diagram of an image enhancement apparatus according to an embodiment of the present disclosure.

detailed description

The technical problems, the technical solutions, and the advantages of the present invention will be more clearly described in conjunction with the accompanying drawings and specific embodiments.

Please refer to FIG. 1. FIG. 1 is a schematic flowchart of an image enhancement method according to an embodiment of the present disclosure, where the method includes:

Step S11: converting the original image from the first color space to the second color space containing the brightness information, and extracting brightness information of the original image;

The first color space is a red, green, blue (RGB) color space, or other color space.

The second color space is a hue, saturation, brightness (HIS) color space, a brightness, chrominance (YUV) color space, or other color space containing luminance information.

Step S12: determining, by using a optimization algorithm, an optimal transformation parameter value of the gradation transformation function for image enhancement, for the luminance information;

In the embodiment of the present disclosure, the gradation transformation function may be a non-complete Beta function, and the formula of the incomplete Beta function is:

Where B(α, β) is a Beta function,

The optimization algorithm may be a Particle Swarm Optimization (PSO), a genetic algorithm or a simulated annealing algorithm.

Step S13: construct a transformation curve of the gradation transformation function according to the optimal transformation parameter value, and perform enhancement processing on the luminance information according to the transformation curve to obtain enhanced luminance information;

Step S14: Convert the enhanced brightness information from the second color space to the first color space to obtain a final enhanced image.

Specifically, the color space conversion is performed according to the enhanced luminance information and other information of the original image in the second color space.

Based on the method provided by the above embodiment, when the image is enhanced by the gradation transformation function, the optimal transformation parameter of the gradation transformation function can be automatically determined by the optimization algorithm, thereby realizing adaptive enhancement of the image.

The image enhancement method of the present disclosure will be described below by taking the gradation transformation function as a non-complete Beta function and the optimization algorithm as an example of the particle swarm optimization algorithm.

Please refer to FIG. 2. FIG. 2 is another schematic flowchart of an image enhancement method according to an embodiment of the present disclosure, where the method includes:

Step S21: converting the original image from the RGB color space to the YUV color space containing the luminance component, and extracting luminance information of the original image;

Specifically, the image collected by the collection device is described by RGB space for each pixel of the image. Therefore, in this embodiment, the first color space is an RGB color space, and the second color space is a YUV color space.

Step S22: Establish a grayscale transformation function for enhancing the image. In this embodiment, the image is nonlinearly transformed by using a non-complete Beta function. The formula of the incomplete Beta function is:

Where B(α, β) is a Beta function,

Step S23: Establish an objective function. When the particle swarm optimization algorithm is used to determine the optimal transform parameter value of the incomplete Beta function, an objective function is used to determine whether the current transform parameter value is the optimal transform parameter value.

The objective function is:

Where, Fitness is the current fitness value of the particle, M is the width of the image to be processed, N is the height of the image to be processed, and f(x, y) is the brightness information of the pixel at the position (x, y);

Step S24: determining, by using a particle swarm optimization algorithm, an optimal transform parameter value of the incomplete Beta function for the brightness information;

The step of determining the optimal transformation parameter value of the incomplete Beta function by using the particle swarm optimization algorithm specifically includes:

Step S241 (not shown): setting the particle swarm optimization parameter: setting the particle population size to N1 in the D-dimensional space, the maximum number of iterations is N2, and the particle velocity ranges from [-v _max , v _max ], the particle The value range of the position is [p _min , p _max ].

In this embodiment, D of the D-dimensional space is the number of solutions in the solution space. In the present embodiment, the objective function Fitness has two solutions of α and β, so the value of D is 2. Set the particle population size N1 to 30, the maximum number of iterations N2 to 100, and the range of particle velocity [-v _max , v _max ] to [-1, 1]. The range of particle positions can be based on different scenarios. The range of values of the transformation parameters α and β of the incomplete Beta function is set;

Step S242 (not shown): Initialization: When the number of iterations n2=0, the initial velocity of each particle in the particle group is randomly set in the solution space.

And initial position

Initializing the global optimal position Gbest;

Step S243 (not shown): calculating the fitness value of each particle

Determine the objective function

Step S244 (not shown): for each particle, comparing its current fitness value with the fitness value of the historical optimal position of the particle to complete its individual optimal solution

Update:

Step S245 (not shown): updating the global optimal position Gbest ⁿ according to the fitness value of each particle:

Step S246 (not shown): the speed of each particle according to the following two formulas

And location

Update;

Where c ₁ is a cognitive learning factor, c ₂ is a social learning factor, c ₁ , c ₂ are usually set to a constant 2; r ₁ and r ₂ are random numbers, subject to a uniform distribution between [0, 1] . w is a non-negative number, and the value is generally between [0.1, 0.9], which controls the influence of the previous speed on the current speed. The basic PSO algorithm can be regarded as a special case of w=1. The most commonly used is the inertia weight linear weight decrement strategy. In this embodiment, the exponential inertia weight nonlinear weight reduction strategy is adopted to achieve a better balance between the global search and the local search.

Usually, the value w _ini = 0.9, w _end = 0.4; t is the current number of iterations, t _max is the total number of iterations of the algorithm, and a is the adjustment parameter, which is used to control the size of the inertia weight transformation. Excessive value of a causes the inertia weight to drop too fast, and enters the local search too early, and the value of a is too small and does not bring about an increase in convergence speed. The technique of the embodiment of the present disclosure selects a value of 15.

Step S247 (not shown): determining whether the iteration termination condition is satisfied; the termination condition is an objective function

The value is the largest.

Step S248 (not shown): If the iteration termination condition is not satisfied, the number of iterations is n2=n2+1, and the process goes to step S243 until the end of the termination condition iteration is completed, and the optimal transformation parameter value of the incomplete Beta function is obtained.

Step S25: construct a transformation curve of the gradation transformation function according to the optimal transformation parameter value, and perform enhancement processing on the luminance information according to the transformation curve to obtain enhanced luminance information;

Step S26: Converting the YUV color space enhanced by the brightness information to the RGB color space to obtain a final enhanced image.

Specifically, color space conversion is performed according to the enhanced luminance information and the chrominance information of the original image in the YUV color space to obtain a final enhanced image.

In the above embodiment, before the optimal transformation parameter value of the gradation transformation function for image enhancement is determined by using the optimization algorithm, the original image or the luminance component image may be normalized to reduce the complexity of the algorithm. Of course, after the image is subjected to enhancement processing, the enhanced luminance information or the final enhanced image also needs to be inverse normalized. The normalization process and the inverse normalization process can be implemented by those skilled in the art according to the techniques in the art, and details are not described herein again.

Please refer to FIG. 3. FIG. 3 is a structural block diagram of an image enhancement apparatus according to an embodiment of the present disclosure, where the image enhancement apparatus includes:

a first image conversion unit 301, configured to convert the original image from the first color space to include light a second color space of the degree information, and extracting brightness information of the original image;

The transformation parameter optimization unit 302 is configured to determine, by using a optimization algorithm, an optimal transformation parameter value of the gradation transformation function for image enhancement for the luminance information;

The enhancement processing unit 303 is configured to construct a transformation curve of the gradation transformation function according to the optimal transformation parameter value, and perform enhancement processing on the luminance information according to the transformation curve to obtain enhanced luminance information;

The second image converting unit 304 is configured to convert the enhanced brightness information from the second color space to the first color space to obtain a final enhanced image.

Optionally, the gradation transformation function is a non-complete Beta function.

The formula for the incomplete Beta function is:

Where B(α, β) is a Beta function,

α, β is the transformation parameter of the incomplete Beta function, t is the integral variable of the incomplete Beta function, and u is the upper limit of the variation range of the integral variable t;

Optionally, the optimization algorithm is a particle swarm optimization algorithm.

The transformation parameter optimization unit is specifically configured to establish an objective function. When the optimal transformation parameter value of the gradation transformation function is determined by using the optimization algorithm, the target function is used to determine whether the current transformation parameter value is Optimal transformation parameter value;

The objective function is:

Where, Fitness is the current fitness value of the particle, M, N are the width and height of the image to be processed, respectively, and f(x, y) is the brightness information of the pixel at the position (x, y).

Optionally, the image enhancement apparatus may further include:

a normalization processing unit (not shown) for normalizing the luminance information of the original image or the original image;

An anti-normalization processing unit (not shown) for using the enhanced luminance information or the final increase The strong image is denormalized.

Embodiments of the present disclosure also provide a display device including the image enhancement device as described above. For the structure and working principle of the image enhancement device included in the display device, refer to the foregoing embodiment, and details are not described herein again. In addition, the structure of other parts of the display device can refer to the prior art, and will not be described in detail herein. The display device may be any product or component having a display function such as a home appliance, a communication device, an engineering device, an electronic entertainment product, or the like.

The above is an alternative embodiment of the present disclosure, and it should be noted that those skilled in the art can also make several improvements and refinements without departing from the principles of the present disclosure. It should be considered as the scope of protection of this disclosure.

Claims

An image enhancement method, comprising:

Converting an original image from a first color space to a second color space containing luminance information, and extracting luminance information of the original image;

Determining, by the optimization algorithm, an optimal transformation parameter value of the gradation transformation function for image enhancement;

Constructing a transformation curve of the gradation transformation function according to the optimal transformation parameter value, and performing enhancement processing on the luminance information according to the transformation curve to obtain enhanced luminance information;

Converting the enhanced luminance information from the second color space to the first color space to obtain a final enhanced image.
The image enhancement method according to claim 1, wherein the step of determining an optimal transformation parameter value of the gradation transformation function for the image enhancement using the optimization algorithm comprises:

Establish an objective function;

The objective function is used to determine whether the current transformation parameter value is an optimal transformation parameter value, and the optimal transformation parameter value of the gradation transformation function is determined.
The image enhancement method according to claim 2, wherein the optimization algorithm is a particle swarm optimization algorithm, and the objective function is:

Where, Fitness is the current fitness value of the particle, M is the width of the image to be processed, N is the height of the image to be processed, and f(x, y) is the brightness information of the pixel at the position (x, y).
The image enhancement method according to claim 1 or 2, wherein the optimization algorithm is a genetic algorithm or a simulated annealing algorithm.
The image enhancement method according to claim 1 or 2, wherein the gradation transformation function is a non-complete Beta function;

The formula for the incomplete Beta function is:

Where B(α, β) is a Beta function,
α, β is the transformation parameter of the incomplete Beta function, t is the integral variable of the incomplete Beta function, and u is the upper limit of the range of variation of the integral variable t.
The image enhancement method according to claim 1, wherein

The step of determining an optimal transform parameter value for the gray-scale transform function of the image enhancement by using the optimization algorithm further includes: normalizing the luminance information of the original image or the original image;

The method further includes performing inverse normalization processing on the enhanced luminance information or the final enhanced image.
An image enhancement device, comprising:

a first image conversion unit, configured to convert the original image from the first color space to a second color space containing the brightness information, and extract brightness information of the original image;

a transformation parameter optimization unit, configured to determine, by using a optimization algorithm, an optimal transformation parameter value of the gradation transformation function for image enhancement for the brightness information;

An enhancement processing unit, configured to construct a transformation curve of the gradation transformation function according to the optimal transformation parameter value, and perform enhancement processing on the luminance information according to the transformation curve to obtain enhanced luminance information;

And a second image conversion unit, configured to convert the enhanced brightness information from the second color space to the first color space to obtain a final enhanced image.
The image enhancement device according to claim 7, wherein

The transformation parameter optimization unit is specifically configured to establish an objective function, and use the objective function to determine whether the current transformation parameter value is an optimal transformation parameter value, and determine an optimal transformation parameter value of the gradation transformation function.
The image enhancement device of claim 7, further comprising:

a normalization processing unit, configured to normalize brightness information of the original image or the original image;

The inverse normalization processing unit is configured to perform inverse normalization processing on the enhanced luminance information or the final enhanced image.
A display device comprising the image enhancement device of claims 7-9.