WO2014169579A1 - 一种色彩增强方法及装置 - Google Patents

一种色彩增强方法及装置 Download PDF

Info

Publication number
WO2014169579A1
WO2014169579A1 PCT/CN2013/084377 CN2013084377W WO2014169579A1 WO 2014169579 A1 WO2014169579 A1 WO 2014169579A1 CN 2013084377 W CN2013084377 W CN 2013084377W WO 2014169579 A1 WO2014169579 A1 WO 2014169579A1
Authority
WO
WIPO (PCT)
Prior art keywords
component
color
reflection
processed image
image
Prior art date
Application number
PCT/CN2013/084377
Other languages
English (en)
French (fr)
Inventor
翁迪望
肖进胜
易本顺
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to JP2015511930A priority Critical patent/JP5864817B2/ja
Priority to EP13834368.6A priority patent/EP2806395B1/en
Priority to KR1020147007380A priority patent/KR101552894B1/ko
Priority to US14/331,544 priority patent/US9196024B2/en
Publication of WO2014169579A1 publication Critical patent/WO2014169579A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/90Dynamic range modification of images or parts thereof
    • G06T5/94Dynamic range modification of images or parts thereof based on local image properties, e.g. for local contrast enhancement
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/20Image enhancement or restoration using local operators
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/70Denoising; Smoothing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10024Color image

Definitions

  • the present invention relates to the field of image processing, and in particular, to a color enhancement method and apparatus.
  • the American physicist Edwin Land's Retinex theory based on color constancy believes that the color of objects perceived by humans is closely related to the reflective properties of the surface of the object, and is not subject to uncertain factors such as intensity of the light source and uneven illumination. influences.
  • a multi-scale color enhancement scheme based on constant color tone is proposed.
  • the basic principle of the scheme is to extract the luminance component from the original image, and obtain the illumination of the original image by performing multi-scale Gaussian filtering on the luminance component of the original image.
  • the component is thus separated from the reflected component of the original image, and then the brightness enhancement ratio is obtained from the reflection component and the luminance component.
  • the three color channels of the original image are multiplied by the brightness enhancement ratio to obtain an enhanced image.
  • the embodiment of the invention provides a color enhancement method and device, which solves the over-enhancement phenomenon of a region with relatively high brightness in the original image caused by the enhanced processing, and eliminates the region with low brightness in the original image after the enhancement processing. A lot of noise is present.
  • a color enhancement method comprising:
  • the performing the enhancement adjustment on the first reflection component to obtain the second reflection component includes:
  • the first reflection component after linear stretching is subjected to enhancement adjustment to obtain the second reflection component.
  • the method further includes:
  • the color component set includes an R component, a G component, and a B component; and performing each color component in the color component set separately Performing wavelet low-pass filtering processing of the same scale to obtain a set of illumination components; wherein the illumination component set includes an illumination component of an R channel of the currently processed image, an illumination component of a G channel of the currently processed image, and the current Processing the illumination component of the B channel of the image;
  • the first reflection component set includes a first reflection component of an R channel of the currently processed image, and a G of the currently processed image a first reflected component of the channel, and a first reflected component of the B channel of the currently processed image;
  • a second enhanced image is acquired from the second set of reflected components after linear stretching.
  • the acquiring, by the color component set and the illumination component set, the first reflection component set includes:
  • the current processed image is enhanced according to the brightness gain to obtain a first enhanced image, and according to linear stretching After the second set of reflection components obtains the second enhanced image, the method further includes:
  • An enhanced image of the currently processed image is acquired based on the first enhanced image and the second enhanced image.
  • a color enhancement apparatus including:
  • a first acquiring unit configured to acquire a luminance component of the currently processed image
  • a first filtering unit configured to perform Gaussian filtering processing on the luminance component obtained by the first acquiring unit, to obtain an illumination component of the currently processed image
  • a second acquiring unit configured to acquire, according to the brightness component obtained by the first acquiring unit and the illumination component obtained by the first filtering unit, a first reflection component of the currently processed image
  • An adjusting unit configured to perform enhancement adjustment on the first reflection component obtained by the second acquiring unit, to obtain a second reflection component
  • a third obtaining unit configured to acquire a brightness gain according to the brightness component obtained by the first acquiring unit and the second reflection component obtained by the adjusting unit;
  • an enhancement processing unit configured to perform enhancement processing on the current processed image according to the brightness gain obtained by the third acquiring unit, to obtain a first enhanced image.
  • the adjusting unit includes: a stretching module, configured to linearly stretch the first reflection component;
  • an adjustment module configured to perform an enhancement adjustment on the first reflection component after the linear stretching, to obtain the second reflection component.
  • the method further includes:
  • a fourth acquiring unit configured to decompose the current processed image according to an RGB spatial principle to obtain a color component set; wherein the color component set includes an R component, a G component, and a B component;
  • a second filtering unit configured to perform wavelet low-pass filtering processing on the same scale for each color component in the color component set obtained by the fourth acquiring unit, to obtain an illumination component set;
  • the illumination component set An illumination component including an R channel of the currently processed image, an illumination component of a G channel of the currently processed image, and an illumination component of a B channel of the currently processed image;
  • a fifth acquiring unit configured to acquire, according to the color component set obtained by the fourth acquiring unit, the second filtering unit, the first component of the first component, and the first component of the first component; And including a first reflection component of the R channel of the currently processed image, a first reflection component of the G channel of the currently processed image, and a first reflection component of the B channel of the currently processed image;
  • a recovery processing unit configured to perform color recovery processing on the reflection components of each of the first reflection component sets obtained by the fifth acquisition unit, to obtain a second reflection component set
  • a stretching unit configured to linearly stretch the reflection components of each of the second reflection component sets obtained by the recovery processing unit
  • a sixth obtaining unit configured to obtain the second enhanced image according to the second reflected component set after the linear stretching.
  • the fifth obtaining unit includes:
  • a processing module configured to respectively enlarge each color component in the color component set obtained by the fourth acquiring unit by the same ratio;
  • an obtaining module configured to acquire the first reflection component set according to the processed color component set and the illumination component set obtained by the second filtering unit.
  • the method further includes:
  • a seventh obtaining unit configured to obtain, by the enhancement processing unit, the first processed image by performing enhancement processing on the current processed image according to the brightness gain
  • the sixth acquiring unit is configured according to the second after linear stretching
  • the enhanced image of the currently processed image is acquired according to the first enhanced image obtained by the enhancement processing unit and the second enhanced image obtained by the sixth acquiring unit.
  • the color enhancement method and device provided by the embodiment of the present invention obtains an illumination component of a currently processed image by performing Gaussian filtering processing on the obtained luminance component, and then acquires a first reflection component of the currently processed image according to the luminance component and the illumination component, and The first reflection component is subjected to enhancement adjustment to obtain a second reflection component, and then the luminance gain is obtained according to the luminance component and the second reflection component, and finally, the current processed image is subjected to enhancement processing according to the luminance gain to obtain an enhanced image, which is provided by the embodiment of the present invention.
  • the scheme not only ensures the fidelity of the image color, but also solves the over-enhancement phenomenon of the region with higher brightness in the original image caused by the enhanced processing, and eliminates the region where the brightness is relatively low in the original image after the enhancement processing. A lot of noise.
  • FIG. 1 is a flowchart of a color enhancement method according to an embodiment of the present invention
  • FIG. 2 is a flow chart of a color enhancement method according to another embodiment of the present invention
  • FIG. 3 is a schematic diagram of a color enhancement device according to another embodiment of the present invention
  • FIG. 5 is a schematic structural diagram of still another color enhancement device according to another embodiment of the present invention.
  • An embodiment of the present invention provides a color enhancement method. As shown in FIG. 1, the method may include:
  • the luminance component is processed by Gaussian filtering to obtain the illumination component of the currently processed image.
  • the first reflected component of the currently processed image is acquired according to the luminance component and the illumination component, the first reflected component may be subjected to enhancement processing to obtain a second reflected component, thereby avoiding the phenomenon of over-enhancement and noise.
  • the color enhancement method provided by the embodiment of the present invention by entering the obtained brightness component Row Gaussian filtering process obtains an illumination component of the currently processed image, and then acquires a first reflection component of the currently processed image according to the luminance component and the illumination component, and performs enhancement adjustment on the first reflection component to obtain a second reflection component, and then according to the luminance component and the The two reflection components acquire the brightness gain, and finally the enhancement processing is performed on the current processed image according to the brightness gain to obtain an enhanced image.
  • the solution provided by the embodiment of the present invention not only ensures the fidelity of the image color, but also solves the enhanced processing.
  • the over-enhancement phenomenon of the region where the brightness is relatively high in the original image is caused, and a large amount of noise appearing in the region where the luminance is relatively low in the original image after the enhancement processing is eliminated.
  • Another embodiment of the present invention provides a color enhancement method. As shown in FIG. 2, the method may include:
  • the Retinex theory based on the color constant theory is based on the method theory of illumination component estimation.
  • the theory holds that the image is composed of the product of the reflection component and the illumination component.
  • the essential information that can reflect the original image is the reflection component, and the illumination component is substantial.
  • the interference on the image causes the visual effect of the image collected by the image to be unsatisfactory. Therefore, an algorithm can be used to calculate the illumination component, and then the illumination component is removed from the original image, that is, the interference portion of the image is removed.
  • the reflected component of the image is obtained to enhance the image.
  • the enhancement of the image is also implemented based on the Retinex theory, and the specific implementation method may include the following steps:
  • the luminance component of the currently processed image is obtained according to the HSV spatial model. For example, for any pixel of all pixels currently processed, the maximum value of the R, G, and B components of the pixel is taken as the luminance component of the pixel, and then for all pixels in the currently processed image, ⁇ By doing the same processing, the luminance components of all the pixels are finally obtained, so that the luminance components of all the pixels are combined to obtain the luminance component of the currently processed image.
  • Gaussian filtering processing may be performed on the luminance component to obtain an illumination component of the currently processed image.
  • Gaussian filtering can effectively guarantee the fidelity of image color.
  • the first reflection component of the current processed image may be acquired according to the obtained luminance component and the illuminance component.
  • the method for obtaining the first reflection component may be according to the following formula. obtain.
  • the first reflection component in order to compress the dynamic range of the obtained first reflection component, the first reflection component may be linearly stretched. Specifically, the mean and the mean square error of all the pixels of the currently processed image may be calculated first, then the maximum and minimum values of the linear stretch are obtained, and then the first reflected component is performed according to the maximum and minimum values of the linear stretch. Linear stretching treatment.
  • the specific calculation formula can be as follows:
  • Max—Min where, is the mean value of the currently processed image, ⁇ is the mean square error of the currently processed image. According to the normal distribution of “3 ⁇ , the law, it can be considered that the pixel value of most pixels of the currently processed image will fall. In the range of [ ⁇ -3 ⁇ + 3 ⁇ ], in order to improve the contrast of the enhanced image, it is generally possible to take about 2.0, which is the first reflection component of the currently processed image, and R is the first reflection component after linear stretching. .
  • the first reflection component after the linear stretching can be enhanced and adjusted to obtain the second reflection component.
  • the first reflection component after linear stretching can be enhanced and adjusted according to the following formula:
  • R R Sin ⁇ - 255 Where 2 is the second reflected component of the currently processed image, R is the first reflected component after linear stretching, and / is the luminance component of the currently processed image.
  • the luminance gain can be obtained according to the second reflection component and the luminance component.
  • the brightness gain can be obtained according to the following formula:
  • the current processed image may be enhanced according to the brightness gain. Specifically, the R, G, and B color channels of the currently processed image may be respectively multiplied by the brightness gain, and finally, An enhanced image, the first enhanced image, is obtained.
  • the Gaussian filtering process is used to estimate the illumination component of the currently processed image, such that the edge region where the image brightness changes sharply causes a halo phenomenon after enhancement, so in order to make the enhanced image look more natural, in the embodiment of the present invention.
  • Obtaining a second enhanced image by performing the following steps 208-211, and combining the results of the first enhanced image and the second enhanced image by performing step 214 to obtain a final enhanced image, so that the enhanced effect of the image is maximized good.
  • the color component set includes an R component, a G component, and a B component.
  • each color component in the color component set is respectively subjected to the same scale.
  • the wavelet low-pass filtering process obtains a set of illumination components of the currently processed image, the illumination component set including an illumination component of the R channel of the currently processed image, an illumination component of the G channel of the currently processed image, and an illumination component of the B channel of the currently processed image, In this way, the halo phenomenon can be avoided, and the efficiency of the wavelet low-pass filtering process is also relatively high, that is, the time for acquiring the illumination component set can be reduced.
  • the first reflection component set may be acquired according to the color component set and the illumination component set, where the first reflection component set includes a first reflection component of the R channel of the currently processed image, The first reflected component of the G channel of the currently processed image, and the first reflected component of the B channel of the currently processed image.
  • the enhancement of the image is achieved by performing enhancement processing on the three color channels in the currently processed image, since the three color channels of the image are related to each other, if only simple enhancement processing is performed, it will inevitably result in The color is distorted. Therefore, in the embodiment of the present invention, after the reflection component is restored, the color component is adjusted to ensure the color fidelity of the image as much as possible. Specifically, the following steps 210a and 210b may be performed.
  • each color component in the color component set amplifying the same ratio for each color component in the color component set.
  • the color distortion of the enhanced image is large, and the color of the enhanced image is substantially the same as that of the original image.
  • Each of the color components in the set of color components may be separately magnified by the same ratio.
  • each color component in the set of color components may be respectively amplified by 1 + , where is the magnification ratio.
  • the first reflective component set can be acquired according to the processed color component set and the illumination component set.
  • the specific one can be obtained by the following calculation formula:
  • R rl (l + )log(/ r )-log( )
  • is the first reflection component of the R channel, and is the R component, which is the illumination component of the R channel, which is a ratio of the proportion of each color component in the set of enlarged color components, and may have a value range of 0 ⁇ 1.
  • the reflection component of each channel in the first reflection component set may be separately subjected to color restoration processing to obtain a second reflection component set, and the specific color restoration processing may be as follows:
  • the color component is the R component as an example:
  • R r2 C r xR r
  • C r log , which is the R component, / g is the G component, and / 3 ⁇ 4 is the 8 component, which is the first reflection component of 1 channel.
  • the reflection components of each channel in the second reflection component set can be linearly stretched separately.
  • the second enhancement image may be acquired according to the linearly stretched second reflection component set.
  • step 201 - step 207 and step 208 - step 213 is not sequential, and the order of execution of step 201 - step 207 and step 208 - step 213 is performed by the embodiment of the present invention. No restrictions.
  • the processed image can be integrated twice, that is, the first enhanced image and the second enhanced image can be integrated to obtain an enhanced image of the currently processed image. You can get the final enhanced image based on the following companies:
  • the color enhancement method provided by the embodiment of the present invention obtains an illumination component of a currently processed image by performing Gaussian filtering processing on the obtained luminance component, and then acquires a first reflection component of the currently processed image according to the luminance component and the illumination component, and is first The reflection component is subjected to enhanced adjustment to obtain a second reflection component, and then the luminance gain is obtained according to the luminance component and the second reflection component, and finally, the current processed image is subjected to enhancement processing according to the luminance gain to obtain an enhanced image, and the solution provided by the embodiment of the present invention is used.
  • the single-scale filtering method reduces the amount of calculation, and the wavelet component low-pass filtering is used to obtain the illumination component, which can effectively avoid the halo phenomenon and reduce the processing time for obtaining the illumination component.
  • the wavelet component low-pass filtering is used to obtain the illumination component, which can effectively avoid the halo phenomenon and reduce the processing time for obtaining the illumination component.
  • FIG. 3 Another embodiment of the present invention provides a color enhancement apparatus, as shown in FIG. 3, including: a first acquisition unit 301, a first filtering unit 302, a second acquisition unit 303, an adjustment unit 304, a third acquisition unit 305, and an enhancement. Processing unit 306.
  • the first obtaining unit 301 is configured to acquire a luminance component of the currently processed image.
  • the first filtering unit 302 is configured to perform Gaussian filtering processing on the luminance component obtained by the first acquiring unit 301 to obtain an illumination component of the currently processed image.
  • a second obtaining unit 303 configured to acquire the current location according to the luminance component obtained by the first acquiring unit 301 and the illumination component obtained by the first filtering unit 302 The first reflection component of the image.
  • the adjusting unit 304 is configured to perform enhancement adjustment on the first reflection component obtained by the second acquiring unit 303 to obtain a second reflection component.
  • the third obtaining unit 305 is configured to obtain a brightness gain according to the brightness component obtained by the first acquiring unit 301 and the second reflection component obtained by the adjusting unit 304.
  • the enhancement processing unit 306 is configured to perform enhancement processing on the current processed image according to the brightness gain obtained by the third acquiring unit 305 to obtain a first enhanced image.
  • the adjusting unit 304 may include: a stretching module 3041 and an adjusting module 3042.
  • the stretching module 3041 is configured to linearly stretch the first reflection component.
  • the adjusting module 3042 is configured to perform an enhancement adjustment on the first reflection component after linear stretching to obtain the second reflection component.
  • the device may further include: a fourth obtaining unit 307, a second filtering unit 308, a fifth obtaining unit 309, a recovery processing unit 310, a stretching unit 311, and a sixth obtaining unit 312.
  • the fourth obtaining unit 307 is configured to decompose the current processed image according to an RGB spatial principle to obtain a color component set.
  • the color component set includes an R component, a G component, and a B component.
  • a second filtering unit 308, configured to perform wavelet low-pass filtering processing on the same scale for each color component in the color component set obtained by the fourth acquiring unit 307, to obtain an illumination component set; wherein, the illumination The component set includes an illumination component of the R channel of the currently processed image, an illumination component of the G channel of the currently processed image, and an illumination component of the B channel of the currently processed image.
  • the fifth obtaining unit 309 is configured to obtain the first reflection component set according to the color component set obtained by the fourth obtaining unit 307 and the second filtering unit 308, where the first reflection component set is obtained.
  • the set of reflected components includes a first reflected component of the R channel of the currently processed image, a first reflected component of the G channel of the currently processed image, and a first reflected component of the B channel of the currently processed image.
  • the recovery processing unit 310 is further configured to be used by the fifth obtaining unit 309.
  • a reflection component of each channel in a set of reflection components is subjected to color recovery processing to obtain a second set of reflection components.
  • the stretching unit 31 1 is configured to linearly stretch the reflection components of each of the second reflection component sets obtained by the recovery processing unit 310.
  • the sixth obtaining unit 312 is configured to obtain a second enhanced image according to the second reflected component set after linear stretching.
  • the fifth obtaining unit 309 may include: a processing module 3091, and an obtaining module 3092.
  • the processing module 3091 is configured to respectively enlarge each color component in the color component set obtained by the fourth acquiring unit 307 by the same ratio.
  • the obtaining module 3092 is configured to acquire the first reflection component set according to the processed color component set and the illumination component set obtained by the second filtering unit 308.
  • the device may further include: a seventh acquiring unit 3 13 .
  • a seventh obtaining unit 313, configured to perform enhancement processing on the current processed image according to the brightness gain to obtain a first enhanced image
  • the color enhancement device obtained by the embodiment of the present invention obtains an illumination component of a currently processed image by performing Gaussian filtering processing on the obtained luminance component, and acquires a first reflection component of the currently processed image according to the luminance component and the illumination component, and is first The reflection component is subjected to enhanced adjustment to obtain a second reflection component, and then the luminance gain is obtained according to the luminance component and the second reflection component, and finally, the current processed image is subjected to enhancement processing according to the luminance gain to obtain an enhanced image, and the solution provided by the embodiment of the present invention is used.
  • the single-scale filtering method reduces the amount of calculation, and the wavelet component low-pass filtering is used to obtain the illumination component, which can effectively avoid the halo phenomenon and reduce the acquisition of the illumination component.
  • the processing time is longer, after the reflection component is recovered, the color fidelity of the image is further ensured by adjusting the color component, and finally the results obtained by the two processing methods are combined to obtain the final enhanced image, and the result is compared. Both are excellent.
  • FIG. 5 Another embodiment of the present invention provides a color enhancement apparatus, as shown in FIG. 5, including: at least one processor 41, a memory 42, a communication interface 43, and a bus 44, the at least one processor 41, the memory 42, and the communication interface 43. Connected via bus 44 and completed communication with each other, where:
  • the bus 44 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (ESA) bus.
  • ISA Industry Standard Architecture
  • PCI Peripheral Component Interconnect
  • ESA Extended Industry Standard Architecture
  • the bus 44 can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in Figure 5, but it does not mean that there is only one bus or one type of bus.
  • the memory 42 is for storing executable program code, the program code including computer operating instructions.
  • Memory 42 may contain high speed RAM memory and may also include non-volatile memory, such as at least one disk memory.
  • the processor 41 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. .
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • the communication interface 43 is mainly used to implement communication between devices of the embodiment.
  • the processor 41 is further configured to acquire a luminance component of the currently processed image, perform Gaussian filtering processing on the luminance component, obtain an illumination component of the current processed image, and acquire the location according to the luminance component and the illumination component. Determining a first reflection component of the current processed image; performing enhancement adjustment on the first reflection component to obtain a second reflection component, acquiring a luminance gain according to the luminance component and the second reflection component, according to the luminance gain
  • the current processed image is subjected to enhancement processing to obtain a first enhanced image.
  • the processor 41 is further configured to linearly stretch the first reflection component, and perform enhancement adjustment on the first reflection component after linear stretching to obtain the second reflection component.
  • the processor 41 is further configured to: view the current according to an RGB space principle Processing the image for decomposition to obtain a color component set; wherein, the color component set includes an R component, a G component, and a B component; respectively performing wavelet low-pass filtering processing on the same scale for each color component in the color component set Obtaining an illumination component set; wherein the illumination component set includes an illumination component of an R channel of the currently processed image, an illumination component of a G channel of the currently processed image, and an illumination component of a B channel of the currently processed image Obtaining a first reflection component set according to the color component set and the illumination component set; wherein the first reflection component set includes a first reflection component of an R channel of the currently processed image, the current processed image a first reflection component of the G channel, and a first reflection component of the B channel of the currently processed image; performing color recovery processing on the reflection components of each of the first reflection component sets to obtain a second reflection component a set of reflection components for each of the second set of reflected components Do not perform linear stretching;
  • the processor 41 is further configured to separately enlarge each color component in the color component set by the same ratio, and acquire the first according to the processed color component set and the illumination component set. A collection of reflection components.
  • the processor 41 is further configured to obtain a first enhanced image by performing enhancement processing on the current processed image according to the brightness gain, and acquiring according to the second stretched component set after linear stretching After the second enhanced image, the enhanced image of the currently processed image is acquired according to the first enhanced image and the second enhanced image.
  • the color enhancement device obtained by the embodiment of the present invention obtains an illumination component of a currently processed image by performing Gaussian filtering processing on the obtained luminance component, and acquires a first reflection component of the currently processed image according to the luminance component and the illumination component, and is first The reflection component is subjected to enhanced adjustment to obtain a second reflection component, and then the luminance gain is obtained according to the luminance component and the second reflection component, and finally, the current processed image is subjected to enhancement processing according to the luminance gain to obtain an enhanced image, and the solution provided by the embodiment of the present invention is used.
  • the single-scale filtering method reduces the amount of calculation, and the wavelet component low-pass filtering is used to obtain the illumination component, which can effectively avoid the halo phenomenon and reduce the acquisition of the illumination component.
  • the processing time is longer, after the reflection component is recovered, the color fidelity of the image is further ensured by adjusting the color component, and finally the results obtained by the two processing methods are combined to obtain the final enhanced image, and the result is compared. Both are excellent.
  • the present invention can be implemented by means of software plus necessary general hardware, and of course, by hardware, but in many cases, the former is a better implementation. .
  • the technical solution of the present invention which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a readable storage medium, such as a floppy disk of a computer.
  • a hard disk or optical disk or the like includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments of the present invention.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Processing Of Color Television Signals (AREA)

Abstract

本发明实施例公开了一种色彩增强方法及装置,涉及图像处理领域,解决了经过增强处理后导致的原图像中亮度比较高的区域的过增强现象,并消除了经过增强处理后原图像中亮度比较低的区域出现的大量噪声。具体方案为:获取当前处理图像的亮度分量;对亮度分量进行高斯滤波处理,获得当前处理图像的照射分量;根据亮度分量和照射分量获取当前处理图像的第一反射分量;对第一反射分量进行增强调整,获得第二反射分量;根据亮度分量和第二反射分量获取亮度增益;根据亮度增益对当前处理图像进行增强处理,获得第一增强图像。本发明用于色彩增强的过程中。

Description

一种色彩增强方法及装置 技术领域
本发明涉及图像处理领域, 尤其涉及一种色彩增强方法及装置。
背景技术
随着信息技术的不断发展, 人们对图像质量的要求也越来越高。 众所周知, 在图像的获取过程中, 经常会由于拍摄设备、 光照等各方 面因素的影响导致图像不清晰, 因此为了保证人们可以观看到清晰的 图像, 则需要对图像进行增强处理, 这样便可以通过对图像的动态范 围的有效的压缩来提高图像的对比度, 进而改善图像的质量。
美国物理学家 Edwin Land基于色彩恒常性提出的 Retinex理论认 为人类所感知到的物体的色彩与物体表面的反射性质有着密切的联 系, 而不会受到光源强度和照射不均匀等不确定的因素的影响。 现有 技术中提出一种基于色调恒定的多尺度色彩增强方案, 该方案的基本 原理是从原图像中提取出亮度分量, 通过对原图像的亮度分量进行多 尺度的高斯滤波得到原图像的照射分量, 从而分离出原图像的反射分 量, 然后由反射分量和亮度分量得到亮度增强比例, 最后再对原图像 的三个颜色通道分别乘以亮度增强比例, 便可以得到增强的图像。
在实现上述图像增强的过程中, 发明人发现现有技术中至少存在 如下问题: 通过对原图像的三个颜色通道分别乘以亮度增强比例实现 对图像的增强, 但是这样会导致原图像中亮度比较高的区域变得过亮, 出现过增强的现象, 同时也会丟失这部分的某些细节信息, 并且也会 导致原图像中亮度比较低的区域出现大量的噪声。
发明内容
本发明的实施例提供一种色彩增强方法及装置, 解决了经过增强 处理后导致的原图像中亮度比较高的区域的过增强现象, 并消除了经 过增强处理后原图像中亮度比较低的区域出现的大量噪声。
为达到上述目的, 本发明的实施例釆用如下技术方案: 本发明的第一方面, 提供一种色彩增强方法, 包括:
获取当前处理图像的亮度分量;
对所述亮度分量进行高斯滤波处理, 获得所述当前处理图像的照 射分量;
根据所述亮度分量和所述照射分量获取所述当前处理图像的第一 反射分量;
对所述第一反射分量进行增强调整, 获得第二反射分量; 根据所述亮度分量和所述第二反射分量获取亮度增益;
根据所述亮度增益对所述当前处理图像进行增强处理, 获得第一 增强图像。
结合第一方面, 在一种可能的实现方式中, 所述对所述第一反射 分量进行增强调整, 获得第二反射分量, 包括:
对所述第一反射分量进行线性拉伸;
对进行线性拉伸后的所述第一反射分量进行增强调整, 获得所述 第二反射分量。
结合第一方面和上述可能的实现方式, 在另一种可能的实现方式 中, 还包括:
根据 RGB空间原理对所述当前处理图像进行分解, 获取色彩分量 集合; 其中, 所述色彩分量集合包括 R分量、 G分量、 以及 B分量; 对所述色彩分量集合中的每个色彩分量分别进行相同尺度的小波 低通滤波处理, 获得照射分量集合; 其中, 所述照射分量集合包括所 述当前处理图像的 R通道的照射分量、 所述当前处理图像的 G通道的 照射分量、 以及所述当前处理图像的 B通道的照射分量;
根据所述色彩分量集合和所述照射分量集合获取第一反射分量集 合; 其中, 所述第一反射分量集合包括所述当前处理图像的 R通道的 第一反射分量、 所述当前处理图像的 G通道的第一反射分量、 以及所 述当前处理图像的 B通道的第一反射分量;
对所述第一反射分量集合中的每个通道的反射分量分别进行彩色 恢复处理, 获得第二反射分量集合; 对所述第二反射分量集合中的每个通道的反射分量分别进行线性 拉伸;
根据线性拉伸后的所述第二反射分量集合获取第二增强图像。
结合第一方面和上述可能的实现方式, 在另一种可能的实现方式 中, 所述根据所述色彩分量集合和所述照射分量集合获取第一反射分 量集合, 包括:
对所述色彩分量集合中的每个色彩分量分别放大同样比例; 根据处理后的所述色彩分量集合和所述照射分量集合获取所述第 一反射分量集合。
结合第一方面和上述可能的实现方式, 在另一种可能的实现方式 中, 在所述根据所述亮度增益对所述当前处理图像进行增强处理, 获 得第一增强图像, 并根据线性拉伸后的所述第二反射分量集合获取第 二增强图像之后, 还包括:
根据所述第一增强图像和所述第二增强图像获取所述当前处理图 像的增强图像。
本发明的第二方面, 提供一种色彩增强装置, 包括:
第一获取单元, 用于获取当前处理图像的亮度分量;
第一滤波单元, 用于对所述第一获取单元得到的所述亮度分量进 行高斯滤波处理, 获得所述当前处理图像的照射分量;
第二获取单元, 用于根据所述第一获取单元得到的所述亮度分量 和所述第一滤波单元得到的所述照射分量获取所述当前处理图像的第 一反射分量;
调整单元, 用于对所述第二获取单元得到的所述第一反射分量进 行增强调整, 获得第二反射分量;
第三获取单元, 用于根据所述第一获取单元得到的所述亮度分量 和所述调整单元得到的所述第二反射分量获取亮度增益;
增强处理单元, 用于根据所述第三获取单元得到的所述亮度增益 对所述当前处理图像进行增强处理, 获得第一增强图像。
结合第二方面, 在一种可能的实现方式中, 所述调整单元, 包括: 拉伸模块, 用于对所述第一反射分量进行线性拉伸;
调整模块, 用于对进行线性拉伸后的所述第一反射分量进行增强 调整, 获得所述第二反射分量。
结合第二方面和上述可能的实现方式, 在另一种可能的实现方式 中, 还包括:
第四获取单元, 用于根据 RGB空间原理对所述当前处理图像进行 分解, 获取色彩分量集合; 其中, 所述色彩分量集合包括 R 分量、 G 分量、 以及 B分量;
第二滤波单元, 用于对所述第四获取单元得到的所述色彩分量集 合中的每个色彩分量分别进行相同尺度的小波低通滤波处理, 获得照 射分量集合; 其中, 所述照射分量集合包括所述当前处理图像的 R通 道的照射分量、 所述当前处理图像的 G通道的照射分量、 以及所述当 前处理图像的 B通道的照射分量;
第五获取单元, 用于根据所述第四获取单元得到的所述色彩分量 集合和所述第二滤波单元得到所述照射分量集合获取第一反射分量集 合; 其中, 所述第一反射分量集合包括所述当前处理图像的 R通道的 第一反射分量、 所述当前处理图像的 G通道的第一反射分量、 以及所 述当前处理图像的 B通道的第一反射分量;
恢复处理单元, 还用于对所述第五获取单元得到的所述第一反射 分量集合中的每个通道的反射分量分别进行彩色恢复处理, 获得第二 反射分量集合;
拉伸单元, 用于对所述恢复处理单元得到的所述第二反射分量集 合中的每个通道的反射分量分别进行线性拉伸;
第六获取单元, 用于根据线性拉伸后的所述第二反射分量集合获 取第二增强图像。
结合第二方面和上述可能的实现方式, 在另一种可能的实现方式 中, 所述第五获取单元, 包括:
处理模块, 用于对所述第四获取单元得到的所述色彩分量集合中 的每个色彩分量分别放大同样比例; 获取模块, 用于根据处理后的所述色彩分量集合和所述第二滤波 单元得到的所述照射分量集合获取所述第一反射分量集合。
结合第二方面和上述可能的实现方式, 在另一种可能的实现方式 中, 还包括:
第七获取单元, 用于在所述增强处理单元根据所述亮度增益对所 述当前处理图像进行增强处理获得第一增强图像, 并所述第六获取单 元根据线性拉伸后的所述第二反射分量集合获取第二增强图像之后, 根据所述增强处理单元得到的所述第一增强图像和所述第六获取单元 得到的所述第二增强图像获取所述当前处理图像的增强图像。
本发明实施例提供的色彩增强方法及装置, 通过对获取到的亮度 分量进行高斯滤波处理获得当前处理图像的照射分量, 再根据亮度分 量和照射分量获取当前处理图像的第一反射分量, 并对第一反射分量 进行增强调整获得第二反射分量, 然后根据亮度分量和第二反射分量 获取亮度增益, 最后根据亮度增益对当前处理图像进行增强处理获得 增强的图像, 釆用本发明实施例提供的方案, 不仅保证了图像颜色的 保真度, 还解决了经过增强处理后导致的原图像中亮度比较高的区域 的过增强现象, 并消除了经过增强处理后原图像中亮度比较低的区域 出现的大量噪声。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案, 下面 将对实施例或现有技术描述中所需要使用的附图作简单地介绍, 显而 易见地, 下面描述中的附图仅仅是本发明的一些实施例, 对于本领域 普通技术人员来讲, 在不付出创造性劳动性的前提下, 还可以根据这 些附图获得其他的附图。
图 1为本发明一实施例提供的一种色彩增强方法流程图;
图 2为本发明另一实施例提供的一种色彩增强方法流程图; 图 3为本发明另一实施例提供的一种色彩增强装置组成示意图; 图 4为本发明另一实施例提供的另一种色彩增强装置组成示意图; 图 5为本发明另一实施例提供的又一种色彩增强装置组成示意图。
具体实施方式
下面将结合本发明实施例中的附图, 对本发明实施例中的技术方 案进行清楚、 完整地描述, 显然, 所描述的实施例仅仅是本发明一部 分实施例, 而不是全部的实施例。 基于本发明中的实施例, 本领域普 通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例, 都属于本发明保护的范围。
本发明一实施例提供一种色彩增强方法, 如图 1 所示, 该方法可 以包括:
101、 获取当前处理图像的亮度分量。
102、 对亮度分量进行高斯滤波处理, 获得当前处理图像的照射分 量。
其中, 由于高斯滤波可以有效的保证当前处理图像颜色的保真度, 所以, 在本发明实施例中釆用高斯滤波对亮度分量进行处理, 以便获 得当前处理图像的照射分量。
103、 根据亮度分量和照射分量获取当前处理图像的第一反射分 量。
104、 对第一反射分量进行增强调整, 获得第二反射分量。
其中, 为了可以避免对当前处理图像中亮度比较高的区域的过增 强现象的出现, 以及避免当前处理图像中亮度比较低的区域经过增强 处理出现大量噪声的情况发生, 在本发明实施例中, 当根据亮度分量 和照射分量获取到当前处理图像的第一反射分量之后, 对第一反射分 量可以进行增强处理, 得到第二反射分量, 这样便可以避免过增强和 噪声的现象。
105、 根据亮度分量和第二反射分量获取亮度增益。
106、 根据亮度增益对当前处理图像进行增强处理, 获得第一增强 图像。
本发明实施例提供的色彩增强方法, 通过对获取到的亮度分量进 行高斯滤波处理获得当前处理图像的照射分量, 再根据亮度分量和照 射分量获取当前处理图像的第一反射分量, 并对第一反射分量进行增 强调整获得第二反射分量, 然后根据亮度分量和第二反射分量获取亮 度增益, 最后根据亮度增益对当前处理图像进行增强处理获得增强的 图像, 釆用本发明实施例提供的方案, 不仅保证了图像颜色的保真度, 还解决了经过增强处理后导致的原图像中亮度比较高的区域的过增强 现象, 并消除了经过增强处理后原图像中亮度比较低的区域出现的大 量噪声。
本发明另一实施例提供一种色彩增强方法, 如图 2 所示, 该方法 可以包括:
基于色彩恒常理论提出的 Retinex理论是基于照射分量估计的方法 理论, 该理论认为图像是由反射分量和照射分量的乘积组成, 能够反 应原图像的本质信息的是其中的反射分量, 而照射分量实质上会对图 像进行干扰导致釆集到的图像的视觉效果不理想, 因此可以釆用某种 算法对照射分量进行计算, 然后从原图像中将照射分量去除, 也就是 去除图像的干扰部分, 从而得到图像的反射分量, 实现图像的增强。 在本发明实施例中, 也是基于 Retinex理论实现对图像的增强的, 具体 的实现方法可以包括以下步骤:
201、 获取当前处理图像的亮度分量。
其中, 根据 HSV空间模型获取当前处理图像的亮度分量。 例如, 对于当前处理图像的所有像素点中的任意一个像素点, 取该像素点的 R、 G、 B 分量的最大值作为该像素点的亮度分量, 然后对当前处理图 像中的所有像素, ^做同样的处理, 最终得到所有像素点的亮度分量, 这样所有像素点的亮度分量组合在一起便可以得到当前处理图像的亮 度分量。
202、 对亮度分量进行高斯滤波处理, 获得当前处理图像的照射分 量。
其中, 当获得当前处理图像的亮度分量之后, 便可以对该亮度分 量进行高斯滤波处理, 得到当前处理图像的照射分量。 釆用高斯滤波 可以有效保证图像颜色的保真度。 203、 根据亮度分量和照射分量获取当前处理图像的第一反射分 量。
其中, 当获得当前处理图像的亮度分量以及照射分量之后, 便可 以根据获得的亮度分量和照度分量获取当前处理图像的第一反射分 量, 具体的, 第一反射分量的获得方法可以是根据以下公式获得。
^ = 1ο§(/) - 1ο§(Ζ)
其中, 为当前处理图像的第一反射分量, /为当前处理图像的亮 度分量, 为当前处理图像的照射分量。
204、 对第一反射分量进行线性拉伸。
其中, 为了对获得的第一反射分量的动态范围进行压缩, 可以对 第一反射分量进行线性拉伸。 具体的, 可以是先计算当前处理图像的 所有像素点的均值和均方差, 然后求出线性拉伸的最大值和最小值, 再根据线性拉伸的最大值和最小值对第一反射分量进行线性拉伸处 理。 具体的计算公式可以如下所示:
Μαχ = μ + χσ ,Min = μ- χσ
Max—Min 其中, 为当前处理图像的均值, σ为当前处理图像的均方差, 根 据正态分布的 "3 σ,,法则, 可以认为当前处理图像的绝大多数像素点的 像素值都会落在 [^ -3σ^ + 3σ]的范围内, 因此为了提高增强后图像的对 比度, —般可以取 2.0左右, 为当前处理图像的第一反射分量, R为 经过线性拉伸后的第一反射分量。
205、 对进行线性拉伸后的第一反射分量进行增强调整, 获得第二 反射分量。
其中, 为了可以有效的避免噪声和过增强的现象, 使得图像的色 彩以及显示效果更好, 可以对进行线性拉伸后的第一反射分量进行增 强调整, 获得第二反射分量。 具体的, 可以根据下述公式对进行线性 拉伸后的第一反射分量进行增强调整:
R = R Sin^^- 255 其中, ?2为当前处理图像的第二反射分量, R为经过线性拉伸后的 第一反射分量, /为当前处理图像的亮度分量。
206、 根据亮度分量和第二反射分量获取亮度增益。
其中, 当对进行线性拉伸后的第一反射分量进行增强调整得到第 二反射分量之后, 便可以根据第二反射分量和亮度分量获得亮度增益。 具体的, 可以根据下述公式得到亮度增益:
I 其中, 为亮度增益, ?2为当前处理图像的第二反射分量, /为当 前处理图像的亮度分量。
207、 根据亮度增益对当前处理图像进行增强处理, 获得第一增强 图像。
其中, 当获取到亮度增益之后, 便可以根据亮度增益对当前处理 图像进行增强处理, 具体的, 可以是对当前处理图像的 R、 G、 B三个 颜色通道分别乘以亮度增益, 最终便可以得到增强的图像, 即第一增 强图像。
众所周知, 釆用高斯滤波处理来估计当前处理图像的照射分量会 使得图像亮度变化剧烈的边缘区域在增强后出现光晕现象, 因此为了 使得增强后的图像看起来更加自然, 在本发明实施例中, 通过执行以 下步骤 208-步骤 213获得第二增强图像, 并通过执行步骤 214将第一 增强图像和第二增强图像的结果结合在一起得到最终的增强后的图 像, 使得图像的增强效果达到最佳。
208、 根据 RGB 空间原理对当前处理图像进行分解, 获取色彩分 量集合。
其中, 色彩分量集合包括 R分量、 G分量、 以及 B分量。
209、 对色彩分量集合中的每个色彩分量分别进行相同尺度的小波 低通滤波处理, 获得照射分量集合。
其中, 为了改进釆用高斯滤波处理获得照射分量出现的光晕现象, 本发明实施例中对色彩分量集合中的每个色彩分量分别进行相同尺度 的小波低通滤波处理获得当前处理图像的照射分量集合, 照射分量集 合包括当前处理图像的 R通道的照射分量、 当前处理图像的 G通道的 照射分量、 以及当前处理图像的 B 通道的照射分量, 这样便可以避免 光晕现象, 并且釆用小波低通滤波处理的效率也比较高, 也就是说可 以减小获取照射分量集合的时间。
210、 根据色彩分量集合和照射分量集合获取第一反射分量集合。 其中, 当获取到色彩分量集合和照射分量集合之后, 便可以根据 色彩分量集合和照射分量集合获取第一反射分量集合, 其中第一反射 分量集合包括当前处理图像的 R通道的第一反射分量、 当前处理图像 的 G通道的第一反射分量、 以及当前处理图像的 B通道的第一反射分 量。
众所周知, 在通过对当前处理图像中的三个颜色通道分别进行增 强处理达到对图像的增强时, 由于图像的三个颜色通道之间是相互关 联的, 如果只是进行简单的增强处理, 必然会导致颜色失真, 因此在 本发明实施例中, 在对反射分量进行恢复之后, 通过对色彩分量进行 调整,来尽量保证图像的颜色保真度。具体的可以是执行以下步骤 210a 和步骤 210b。
210a, 对色彩分量集合中的每个色彩分量分别放大同样的比例。 其中, 由于釆用小波低通滤波处理来获取照射分量集合, 会导致 增强后的图像的颜色失真较大, 为了保证增强后的图像的颜色和原图 像的颜色基本保持一致, 在本发明实施例中可以对色彩分量集合中的 每个色彩分量分别放大同样的比例, 例如, 可以将色彩分量集合中的 每个色彩分量分别放大 1 + , 其中 为放大比例。
210b , 根据处理后的色彩分量集合和照射分量集合获取第一反射 分量集合。
其中, 当对色彩分量集合中的每个色彩分量分别放大同样的比例 之后, 便可以根据处理后的色彩分量集合和照射分量集合获取第一反 射分量集合。 以彩色分量为 R分量为例, 具体的可以是通过下述计算 公式得到的:
Rrl = (l + )log(/r)-log( ) 其中, ^为 R通道的第一反射分量, 为 R分量, 为 R通道的 照射分量, 为放大色彩分量集合中的每个色彩分量的比例数, 且 的 取值范围可以是 0< < 1。
21 1、 对第一反射分量集合中的每个通道的反射分量分别进行彩色 恢复处理, 获得第二反射分量集合。
其中, 当获取到第一反射分量集合之后, 便可以对第一反射分量 集合中的每个通道的反射分量分别进行彩色恢复处理, 获得第二反射 分量集合, 具体的色彩恢复处理可以是按照以下公式进行的, 以彩色 分量为 R分量为例:
Rr2 = Cr xRr
其中, 2为 R通道的第二反射分量, 为 R通道的色彩恢复因子,
Cr =log , 为 R分量, /g为 G分量, /¾为 8分量, 为 1 通道 的第一反射分量。
212、 对第二反射分量集合中的每个通道的反射分量分别进行线性 拉伸。
其中, 当获取到第二反射分量集合之后, 便可以对第二反射分量 集合中的每个通道的反射分量分别进行线性拉伸。
需要说明的是, 线性拉伸的具体参数的描述可以参考本发明实施 例中步骤 204的对应参数的描述, 本发明实施例在此不再——赘述。
213、 根据线性拉伸后的第二反射分量集合获取第二增强图像。 其中, 当获得线性拉伸后的第二反射分量集合之后, 便可以根据 线性拉伸后的第二反射分量集合获取第二增强图像。
需要说明的是, 在本发明实施例中步骤 201 -步骤 207与步骤 208- 步骤 213 的执行顺序没有先后关系, 本发明实施例在此对步骤 201 -步 骤 207与步骤 208-步骤 213的执行顺序不作限制。
214、 根据第一增强图像和第二增强图像获取当前处理图像的增强 图像。 其中, 当获得第一增强图像和第二增强图像之后, 便可以两次增 强处理后的图像进行综合, 也就是说可以综合第一增强图像和第二增 强图像获取当前处理图像的增强图像, 具体的可以根据以下公司获得 最终的增强图像:
。„i = Wx/。Mil + (卜 w)x/。„i2
其中, /。„,为最终得到的增强后的图像, /。Μί1为第一增强图像, /。„ί2为 第二增强图像, w为第一增强图像和第二增强图像进行加权计算的权 值, 一般情况下可以取 0.5。
本发明实施例提供的色彩增强方法, 通过对获取到的亮度分量进 行高斯滤波处理获得当前处理图像的照射分量, 再根据亮度分量和照 射分量获取当前处理图像的第一反射分量, 并对第一反射分量进行增 强调整获得第二反射分量, 然后根据亮度分量和第二反射分量获取亮 度增益, 最后根据亮度增益对当前处理图像进行增强处理获得增强的 图像, 釆用本发明实施例提供的方案, 不仅保证了图像颜色的保真度, 还解决了经过增强处理后导致的原图像中亮度比较高的区域的过增强 现象, 并消除了经过增强处理后原图像中亮度比较低的区域出现的大 量噪声。
并且, 釆用单尺度的滤波方法, 减少了计算量, 釆用小波低通滤 波获取照射分量, 可以有效避免了光晕现象, 并减少了获取照射分量 的处理时长, 在对反射分量进行恢复之后, 通过对色彩分量进行调整 来进一步的保证了图像的颜色保真度, 最终将两种处理方法得到的结 果综合之后得到最后的增强图像, 结果相较于两者都要优秀。
本发明另一实施例提供一种色彩增强装置, 如图 3 所示, 包括: 第一获取单元 301、 第一滤波单元 302、 第二获取单元 303、 调整单元 304、 第三获取单元 305、 增强处理单元 306。
第一获取单元 301 , 用于获取当前处理图像的亮度分量。
第一滤波单元 302 ,用于对所述第一获取单元 301得到的所述亮度 分量进行高斯滤波处理, 获得所述当前处理图像的照射分量。
第二获取单元 303 ,用于根据所述第一获取单元 301得到的所述亮 度分量和所述第一滤波单元 302 得到的所述照射分量获取所述当前处 理图像的第一反射分量。
调整单元 304 ,用于对所述第二获取单元 303得到的所述第一反射 分量进行增强调整, 获得第二反射分量。
第三获取单元 305 ,用于根据所述第一获取单元 301得到的所述亮 度分量和所述调整单元 304得到的所述第二反射分量获取亮度增益。
增强处理单元 306 ,用于根据所述第三获取单元 305得到的所述亮 度增益对所述当前处理图像进行增强处理, 获得第一增强图像。
进一步的, 如图 4所示, 所述调整单元 304 可以包括: 拉伸模块 3041、 调整模块 3042。
拉伸模块 3041 , 用于对所述第一反射分量进行线性拉伸。
调整模块 3042 , 用于对进行线性拉伸后的所述第一反射分量进行 增强调整, 获得所述第二反射分量。
进一步的, 所述装置还可以包括: 第四获取单元 307、 第二滤波单 元 308、 第五获取单元 309、 恢复处理单元 310、 拉伸单元 311、 第六 获取单元 312。
第四获取单元 307 , 用于根据 RGB空间原理对所述当前处理图像 进行分解, 获取色彩分量集合; 其中, 所述色彩分量集合包括 R分量、 G分量、 以及 B分量。
第二滤波单元 308 ,用于对所述第四获取单元 307得到的所述色彩 分量集合中的每个色彩分量分别进行相同尺度的小波低通滤波处理, 获得照射分量集合; 其中, 所述照射分量集合包括所述当前处理图像 的 R通道的照射分量、 所述当前处理图像的 G通道的照射分量、 以及 所述当前处理图像的 B通道的照射分量。
第五获取单元 309 ,用于根据所述第四获取单元 307得到的所述色 彩分量集合和所述第二滤波单元 308 得到所述照射分量集合获取第一 反射分量集合; 其中, 所述第一反射分量集合包括所述当前处理图像 的 R通道的第一反射分量、 所述当前处理图像的 G通道的第一反射分 量、 以及所述当前处理图像的 B通道的第一反射分量。
恢复处理单元 310 ,还用于对所述第五获取单元 309得到的所述第 一反射分量集合中的每个通道的反射分量分别进行彩色恢复处理, 获 得第二反射分量集合。
拉伸单元 31 1 ,用于对所述恢复处理单元 310得到的所述第二反射 分量集合中的每个通道的反射分量分别进行线性拉伸。
第六获取单元 312 ,用于根据线性拉伸后的所述第二反射分量集合 获取第二增强图像。
进一步的, 所述第五获取单元 309可以包括: 处理模块 3091、 获 取模块 3092。
处理模块 3091 , 用于对所述第四获取单元 307得到的所述色彩分 量集合中的每个色彩分量分别放大同样的比例。
获取模块 3092 , 用于根据处理后的所述色彩分量集合和所述第二 滤波单元 308得到的所述照射分量集合获取所述第一反射分量集合。
进一步的, 所述装置还可以包括: 第七获取单元 3 13。
第七获取单元 313 ,用于在所述增强处理单元 306根据所述亮度增 益对所述当前处理图像进行增强处理获得第一增强图像, 并所述第六 获取单元 3 12 根据线性拉伸后的所述第二反射分量集合获取第二增强 图像之后, 根据所述增强处理单元 306 得到的所述第一增强图像和所 述第六获取单元 312 得到的所述第二增强图像获取所述当前处理图像 的增强图像。
本发明实施例提供的色彩增强装置, 通过对获取到的亮度分量进 行高斯滤波处理获得当前处理图像的照射分量, 再根据亮度分量和照 射分量获取当前处理图像的第一反射分量, 并对第一反射分量进行增 强调整获得第二反射分量, 然后根据亮度分量和第二反射分量获取亮 度增益, 最后根据亮度增益对当前处理图像进行增强处理获得增强的 图像, 釆用本发明实施例提供的方案, 不仅保证了图像颜色的保真度, 还解决了经过增强处理后导致的原图像中亮度比较高的区域的过增强 现象, 并消除了经过增强处理后原图像中亮度比较低的区域出现的大 量噪声。
并且, 釆用单尺度的滤波方法, 减少了计算量, 釆用小波低通滤 波获取照射分量, 可以有效避免了光晕现象, 并减少了获取照射分量 的处理时长, 在对反射分量进行恢复之后, 通过对色彩分量进行调整 来进一步的保证了图像的颜色保真度, 最终将两种处理方法得到的结 果综合之后得到最后的增强图像, 结果相较于两者都要优秀。
本发明另一实施例提供一种色彩增强装置, 如图 5 所示, 包括: 至少一个处理器 41、 存储器 42、 通信接口 43和总线 44 , 该至少一个 处理器 41、存储器 42和通信接口 43通过总线 44连接并完成相互间的 通信, 其中:
所述总线 44 可以是工业标准体系结构 ( Industry Standard Architecture , ISA ) 总线、 夕卜部设备互连 ( Peripheral Component Interconnect , PCI ) 总线或扩展工业标准体系结构 ( Extended Industry Standard Architecture , EISA ) 总线等。 该总线 44可以分为地址总线、 数据总线、 控制总线等。 为便于表示, 图 5 中仅用一条粗线表示, 但 并不表示仅有一根总线或一种类型的总线。
所述存储器 42用于存储可执行程序代码, 该程序代码包括计算机 操作指令。 存储器 42可能包含高速 RAM存储器, 也可能还包括非易 失性存储器 ( non-volatile memory ) , 例如至少一个磁盘存储器。
所述处理器 41可能是一个中央处理器 (Central Processing Unit, CPU ) , 或者是特定集成电路 ( Application Specific Integrated Circuit, ASIC ) , 或者是被配置成实施本发明实施例的一个或多个集成电路。
所述通信接口 43 , 主要用于实现本实施例的设备之间的通信。 所述处理器 41 , 还用于获取当前处理图像的亮度分量, 并对所述 亮度分量进行高斯滤波处理, 获得所述当前处理图像的照射分量, 根 据所述亮度分量和所述照射分量获取所述当前处理图像的第一反射分 量; 对所述第一反射分量进行增强调整, 获得第二反射分量, 根据所 述亮度分量和所述第二反射分量获取亮度增益, 根据所述亮度增益对 所述当前处理图像进行增强处理, 获得第一增强图像。
进一步的, 所述处理器 41 , 还用于对所述第一反射分量进行线性 拉伸, 并对进行线性拉伸后的所述第一反射分量进行增强调整, 获得 所述第二反射分量。
进一步的, 所述处理器 41 , 还用于根据 RGB空间原理对所述当前 处理图像进行分解, 获取色彩分量集合; 其中, 所述色彩分量集合包 括 R分量、 G分量、 以及 B分量; 对所述色彩分量集合中的每个色彩 分量分别进行相同尺度的小波低通滤波处理, 获得照射分量集合; 其 中, 所述照射分量集合包括所述当前处理图像的 R通道的照射分量、 所述当前处理图像的 G通道的照射分量、 以及所述当前处理图像的 B 通道的照射分量; 根据所述色彩分量集合和所述照射分量集合获取第 一反射分量集合; 其中, 所述第一反射分量集合包括所述当前处理图 像的 R通道的第一反射分量、 所述当前处理图像的 G通道的第一反射 分量、 以及所述当前处理图像的 B 通道的第一反射分量; 对所述第一 反射分量集合中的每个通道的反射分量分别进行彩色恢复处理, 获得 第二反射分量集合; 对所述第二反射分量集合中的每个通道的反射分 量分别进行线性拉伸; 根据线性拉伸后的所述第二反射分量集合获取 第二增强图像。
进一步的, 所述处理器 41 , 还用于对所述色彩分量集合中的每个 色彩分量分别放大同样的比例, 根据处理后的所述色彩分量集合和所 述照射分量集合获取所述第一反射分量集合。
进一步的, 所述处理器 41 , 还用于在所述根据所述亮度增益对所 述当前处理图像进行增强处理获得第一增强图像, 并根据线性拉伸后 的所述第二反射分量集合获取第二增强图像之后, 根据所述第一增强 图像和所述第二增强图像获取所述当前处理图像的增强图像。
本发明实施例提供的色彩增强装置, 通过对获取到的亮度分量进 行高斯滤波处理获得当前处理图像的照射分量, 再根据亮度分量和照 射分量获取当前处理图像的第一反射分量, 并对第一反射分量进行增 强调整获得第二反射分量, 然后根据亮度分量和第二反射分量获取亮 度增益, 最后根据亮度增益对当前处理图像进行增强处理获得增强的 图像, 釆用本发明实施例提供的方案, 不仅保证了图像颜色的保真度, 还解决了经过增强处理后导致的原图像中亮度比较高的区域的过增强 现象, 并消除了经过增强处理后原图像中亮度比较低的区域出现的大 量噪声。
并且, 釆用单尺度的滤波方法, 减少了计算量, 釆用小波低通滤 波获取照射分量, 可以有效避免了光晕现象, 并减少了获取照射分量 的处理时长, 在对反射分量进行恢复之后, 通过对色彩分量进行调整 来进一步的保证了图像的颜色保真度, 最终将两种处理方法得到的结 果综合之后得到最后的增强图像, 结果相较于两者都要优秀。
通过以上的实施方式的描述, 所属领域的技术人员可以清楚地了 解到本发明可借助软件加必需的通用硬件的方式来实现, 当然也可以 通过硬件, 但很多情况下前者是更佳的实施方式。 基于这样的理解, 本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软 件产品的形式体现出来, 该计算机软件产品存储在可读取的存储介质 中, 如计算机的软盘, 硬盘或光盘等, 包括若干指令用以使得一台计 算机设备 (可以是个人计算机, 服务器, 或者网络设备等) 执行本发 明各个实施例所述的方法。
以上所述, 仅为本发明的具体实施方式, 但本发明的保护范围并不 局限于此, 任何熟悉本技术领域的技术人员在本发明揭露的技术范围 内, 可轻易想到的变化或替换, 都应涵盖在本发明的保护范围之内。 因 此, 本发明的保护范围应以所述权利要求的保护范围为准。

Claims

权 利 要 求
1、 一种色彩增强方法, 其特征在于, 包括:
获取当前处理图像的亮度分量;
对所述亮度分量进行高斯滤波处理, 获得所述当前处理图像的照 射分量;
根据所述亮度分量和所述照射分量获取所述当前处理图像的第一 反射分量;
对所述第一反射分量进行增强调整, 获得第二反射分量; 根据所述亮度分量和所述第二反射分量获取亮度增益;
根据所述亮度增益对所述当前处理图像进行增强处理, 获得第一 增强图像。
2、 根据权利要求 1所述的色彩增强方法, 其特征在于, 所述对所 述第一反射分量进行增强调整, 获得第二反射分量, 包括:
对所述第一反射分量进行线性拉伸;
对进行线性拉伸后的所述第一反射分量进行增强调整, 获得所述 第二反射分量。
3、 根据权利要求 1所述的色彩增强方法, 其特征在于, 还包括: 根据 RGB空间原理对所述当前处理图像进行分解, 获取色彩分量 集合; 其中, 所述色彩分量集合包括 R分量、 G分量、 以及 B分量; 对所述色彩分量集合中的每个色彩分量分别进行相同尺度的小波 低通滤波处理, 获得照射分量集合; 其中, 所述照射分量集合包括所 述当前处理图像的 R通道的照射分量、 所述当前处理图像的 G通道的 照射分量、 以及所述当前处理图像的 B通道的照射分量;
根据所述色彩分量集合和所述照射分量集合获取第一反射分量集 合; 其中, 所述第一反射分量集合包括所述当前处理图像的 R通道的 第一反射分量、 所述当前处理图像的 G通道的第一反射分量、 以及所 述当前处理图像的 B通道的第一反射分量;
对所述第一反射分量集合中的每个通道的反射分量分别进行彩色 恢复处理, 获得第二反射分量集合;
对所述第二反射分量集合中的每个通道的反射分量分别进行线性 拉伸; 根据线性拉伸后的所述第二反射分量集合获取第二增强图像。
4、 根据权利要求 3所述的色彩增强方法, 其特征在于, 所述根据 所述色彩分量集合和所述照射分量集合获取第一反射分量集合, 包括: 对所述色彩分量集合中的每个色彩分量分别放大同样比例; 根据处理后的所述色彩分量集合和所述照射分量集合获取所述第 一反射分量集合。
5、 根据权利要求 1 -4中任一项所述的色彩增强方法, 其特征在于, 在所述根据所述亮度增益对所述当前处理图像进行增强处理, 获得第 一增强图像, 并根据线性拉伸后的所述第二反射分量集合获取第二增 强图像之后, 还包括:
根据所述第一增强图像和所述第二增强图像获取所述当前处理图 像的增强图像。
6、 一种色彩增强装置, 其特征在于, 包括:
第一获取单元, 用于获取当前处理图像的亮度分量;
第一滤波单元, 用于对所述第一获取单元得到的所述亮度分量进 行高斯滤波处理, 获得所述当前处理图像的照射分量;
第二获取单元, 用于根据所述第一获取单元得到的所述亮度分量 和所述第一滤波单元得到的所述照射分量获取所述当前处理图像的第 一反射分量;
调整单元, 用于对所述第二获取单元得到的所述第一反射分量进 行增强调整, 获得第二反射分量;
第三获取单元, 用于根据所述第一获取单元得到的所述亮度分量 和所述调整单元得到的所述第二反射分量获取亮度增益;
增强处理单元, 用于根据所述第三获取单元得到的所述亮度增益 对所述当前处理图像进行增强处理, 获得第一增强图像。
7、 根据权利要求 6所述的色彩增强装置, 其特征在于, 所述调整 单元, 包括:
拉伸模块, 用于对所述第一反射分量进行线性拉伸;
调整模块, 用于对进行线性拉伸后的所述第一反射分量进行增强 调整, 获得所述第二反射分量。
8、 根据权利要求 6所述的色彩增强装置, 其特征在于, 还包括: 第四获取单元, 用于根据 RGB空间原理对所述当前处理图像进行 分解, 获取色彩分量集合; 其中, 所述色彩分量集合包括 R 分量、 G 分量、 以及 B分量;
第二滤波单元, 用于对所述第四获取单元得到的所述色彩分量集 合中的每个色彩分量分别进行相同尺度的小波低通滤波处理, 获得照 射分量集合; 其中, 所述照射分量集合包括所述当前处理图像的 R通 道的照射分量、 所述当前处理图像的 G通道的照射分量、 以及所述当 前处理图像的 B通道的照射分量;
第五获取单元, 用于根据所述第四获取单元得到的所述色彩分量 集合和所述第二滤波单元得到所述照射分量集合获取第一反射分量集 合; 其中, 所述第一反射分量集合包括所述当前处理图像的 R通道的 第一反射分量、 所述当前处理图像的 G通道的第一反射分量、 以及所 述当前处理图像的 B通道的第一反射分量;
恢复处理单元, 还用于对所述第五获取单元得到的所述第一反射 分量集合中的每个通道的反射分量分别进行彩色恢复处理, 获得第二 反射分量集合;
拉伸单元, 用于对所述恢复处理单元得到的所述第二反射分量集 合中的每个通道的反射分量分别进行线性拉伸;
第六获取单元, 用于根据线性拉伸后的所述第二反射分量集合获 取第二增强图像。
9、 根据权利要求 8所述的色彩增强装置, 其特征在于, 所述第五 获取单元, 包括:
处理模块, 用于对所述第四获取单元得到的所述色彩分量集合中 的每个色彩分量分别放大同样比例;
获取模块, 用于根据处理后的所述色彩分量集合和所述第二滤波 单元得到的所述照射分量集合获取所述第一反射分量集合。
10、根据权利要求 6 - 9中任一项所述的色彩增强装置,其特征在于, 还包括:
第七获取单元, 用于在所述增强处理单元根据所述亮度增益对所述当 前处理图像进行增强处理获得第一增强图像, 并所述第六获取单元根据 线性拉伸后的所述第二反射分量集合获取第二增强图像之后, 根据所述 增强处理单元得到的所述第一增强图像和所述第六获取单元得到的所述 第二增强图像获取所述当前处理图像的增强图像。
PCT/CN2013/084377 2013-04-19 2013-09-27 一种色彩增强方法及装置 WO2014169579A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2015511930A JP5864817B2 (ja) 2013-04-19 2013-09-27 カラーをエンハンスするための方法及び装置
EP13834368.6A EP2806395B1 (en) 2013-04-19 2013-09-27 Color enhancement method and device
KR1020147007380A KR101552894B1 (ko) 2013-04-19 2013-09-27 색 강화 방법 및 장치
US14/331,544 US9196024B2 (en) 2013-04-19 2014-07-15 Method and apparatus for enhancing color

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310137163.X 2013-04-19
CN201310137163.XA CN103236040B (zh) 2013-04-19 2013-04-19 一种色彩增强方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/331,544 Continuation US9196024B2 (en) 2013-04-19 2014-07-15 Method and apparatus for enhancing color

Publications (1)

Publication Number Publication Date
WO2014169579A1 true WO2014169579A1 (zh) 2014-10-23

Family

ID=48884079

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/084377 WO2014169579A1 (zh) 2013-04-19 2013-09-27 一种色彩增强方法及装置

Country Status (6)

Country Link
US (1) US9196024B2 (zh)
EP (1) EP2806395B1 (zh)
JP (1) JP5864817B2 (zh)
KR (1) KR101552894B1 (zh)
CN (1) CN103236040B (zh)
WO (1) WO2014169579A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113160096A (zh) * 2021-05-27 2021-07-23 山东中医药大学 一种基于视网膜模型的低光图像增强方法
CN113781338A (zh) * 2021-08-31 2021-12-10 咪咕文化科技有限公司 图像增强方法、装置、设备及介质

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI566205B (zh) * 2012-11-02 2017-01-11 輝達公司 圖形驅動程式在顯像圖框中近似動態模糊的方法
CN103236040B (zh) * 2013-04-19 2016-03-30 华为技术有限公司 一种色彩增强方法及装置
CN105096278B (zh) * 2015-09-22 2016-08-17 南阳理工学院 基于光照调整的图像增强方法和设备
CN105635525A (zh) * 2015-12-23 2016-06-01 努比亚技术有限公司 一种图像细节处理方法和装置
WO2018038340A2 (en) 2016-08-24 2018-03-01 Samsung Electronics Co., Ltd. Electronic device including light-emitting elements and method of operating electronic device
CN106531125B (zh) * 2017-01-04 2019-02-01 深圳创维-Rgb电子有限公司 图像显示亮度的调整方法、装置及电视机
KR102015708B1 (ko) * 2017-11-24 2019-08-28 주식회사 아이오로라 광원 보정을 수행하는 영상 보정 장치 및 방법
CN108230272B (zh) * 2018-01-04 2022-04-15 京东方科技集团股份有限公司 一种图像增强方法和装置
CN109712097B (zh) * 2019-01-04 2021-04-30 Oppo广东移动通信有限公司 图像处理方法、装置、存储介质及电子设备
CN109872285B (zh) * 2019-01-21 2022-06-10 电子科技大学 一种基于变分约束的Retinex低照度彩色图像增强方法
CN109859144B (zh) * 2019-02-22 2021-03-12 上海商汤智能科技有限公司 图像处理方法及装置、电子设备和存储介质
CN112581400B (zh) * 2020-12-22 2024-05-31 安徽圭目机器人有限公司 基于高斯标准差和对比度调谐图像增强方法
CN112580672A (zh) * 2020-12-28 2021-03-30 安徽创世科技股份有限公司 一种适用于暗环境的车牌识别预处理方法、装置及存储介质
CN112837388B (zh) * 2021-02-01 2023-04-28 清华大学深圳国际研究生院 多光源图片生成方法
CN113313639A (zh) * 2021-03-03 2021-08-27 辽宁工程技术大学 基于Retinex多层次分解的图像增强方法
CN113362294B (zh) * 2021-05-27 2022-09-02 同济大学 含穿刺针超声血管图像的穿刺针识别方法、系统和设备
CN113808036B (zh) * 2021-08-31 2023-02-24 西安理工大学 基于Retinex模型的低照度图像增强与去噪方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100771158B1 (ko) * 2005-10-07 2007-10-29 삼성전자주식회사 칼라 영상의 화질 향상을 위한 방법 및 시스템
US20080107333A1 (en) * 2006-11-08 2008-05-08 Amir Mazinani Method and apparatus for color image correction
CN101303766A (zh) * 2008-07-09 2008-11-12 北京航空航天大学 一种基于Retinex理论的快速彩色图像增强方法
CN102044070A (zh) * 2011-01-10 2011-05-04 北京师范大学 一种基于Retinex的非线性彩色图像增强方法
CN102129673A (zh) * 2011-04-19 2011-07-20 大连理工大学 一种随意光照下彩色数字图像增强和去噪方法
CN103236040A (zh) * 2013-04-19 2013-08-07 华为技术有限公司 一种色彩增强方法及装置

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5991456A (en) * 1996-05-29 1999-11-23 Science And Technology Corporation Method of improving a digital image
US6834125B2 (en) * 2001-06-25 2004-12-21 Science And Technology Corp. Method of improving a digital image as a function of its dynamic range
US6842543B2 (en) * 2001-06-25 2005-01-11 Science And Technology Corporation Method of improving a digital image having white zones
KR100879536B1 (ko) 2006-10-30 2009-01-22 삼성전자주식회사 영상의 화질 개선을 위한 방법 및 시스템
US8165418B2 (en) * 2007-03-30 2012-04-24 Brother Kogyo Kabushiki Kaisha Image processor
US8111943B2 (en) * 2009-04-15 2012-02-07 The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration Smart image enhancement process
JP5392560B2 (ja) * 2009-10-13 2014-01-22 株式会社Jvcケンウッド 画像処理装置および画像処理方法
JP2012249256A (ja) * 2011-05-31 2012-12-13 Sony Corp 画像処理装置、画像処理方法、プログラム
KR101241347B1 (ko) 2011-08-19 2013-03-11 전남대학교산학협력단 휘도 및 조명마스크의 밝기에 따른 가중치를 이용한 영상의 콘트라스트 보정 장치 및 방법
CN102436640A (zh) 2011-09-21 2012-05-02 北京航空航天大学 一种基于HIS空间的多尺度Retinex模型的雾天图像清晰化方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100771158B1 (ko) * 2005-10-07 2007-10-29 삼성전자주식회사 칼라 영상의 화질 향상을 위한 방법 및 시스템
US20080107333A1 (en) * 2006-11-08 2008-05-08 Amir Mazinani Method and apparatus for color image correction
CN101303766A (zh) * 2008-07-09 2008-11-12 北京航空航天大学 一种基于Retinex理论的快速彩色图像增强方法
CN102044070A (zh) * 2011-01-10 2011-05-04 北京师范大学 一种基于Retinex的非线性彩色图像增强方法
CN102129673A (zh) * 2011-04-19 2011-07-20 大连理工大学 一种随意光照下彩色数字图像增强和去噪方法
CN103236040A (zh) * 2013-04-19 2013-08-07 华为技术有限公司 一种色彩增强方法及装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113160096A (zh) * 2021-05-27 2021-07-23 山东中医药大学 一种基于视网膜模型的低光图像增强方法
CN113160096B (zh) * 2021-05-27 2023-12-08 山东中医药大学 一种基于视网膜模型的低光图像增强方法
CN113781338A (zh) * 2021-08-31 2021-12-10 咪咕文化科技有限公司 图像增强方法、装置、设备及介质

Also Published As

Publication number Publication date
US9196024B2 (en) 2015-11-24
JP2015517167A (ja) 2015-06-18
CN103236040B (zh) 2016-03-30
EP2806395A4 (en) 2014-12-17
CN103236040A (zh) 2013-08-07
JP5864817B2 (ja) 2016-02-17
EP2806395A1 (en) 2014-11-26
KR101552894B1 (ko) 2015-09-14
US20140328539A1 (en) 2014-11-06
KR20140145109A (ko) 2014-12-22
EP2806395B1 (en) 2020-01-01

Similar Documents

Publication Publication Date Title
WO2014169579A1 (zh) 一种色彩增强方法及装置
US20200273154A1 (en) Image enhancement method and system
JP6461165B2 (ja) 画像の逆トーンマッピングの方法
WO2016206087A1 (zh) 一种低照度图像处理方法和装置
US9639956B2 (en) Image adjustment using texture mask
US9406148B2 (en) Image processing method and apparatus, and shooting terminal
US9514525B2 (en) Temporal filtering for image data using spatial filtering and noise history
JP5767064B2 (ja) イメージのエッジ向上方法
TW201120809A (en) System and method for processing an image edge
CN111353955A (zh) 一种图像处理方法、装置、设备和存储介质
CN107564085B (zh) 图像扭曲处理方法、装置、计算设备及计算机存储介质
Mu et al. Low and non-uniform illumination color image enhancement using weighted guided image filtering
CN110689486A (zh) 图像的处理方法、装置、设备及计算机可存储介质
JP2024037722A (ja) コンテンツに基づく画像処理
CN117252773A (zh) 基于自适应颜色校正和导向滤波的图像增强方法及系统
WO2020113824A1 (zh) 图像处理方法
JP6631076B2 (ja) 画像処理装置、プログラム及び画像処理方法
JP7365206B2 (ja) 画像処理装置、画像処理方法、及びプログラム
Liu et al. An adaptive tone mapping algorithm based on gaussian filter
Kawasaki et al. A multiscale retinex with low computational cost
CN113379631B (zh) 一种图像去雾的方法及装置
Xue et al. P‐5.6: Synergistic Low‐Light Image Enhancement: A Fusion of Dark Channel Dehazing and K‐means Clustering
US20080013856A1 (en) Method for obtaining high snr image
JP2014178742A (ja) 画像処理装置、画像処理方法及び画像処理プログラム
Li et al. Realistic Scenes Reproduction Based on Total Variation

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 2013834368

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 20147007380

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2015511930

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE