WO2010015140A1 - 一种视频增强的方法及其装置 - Google Patents
一种视频增强的方法及其装置 Download PDFInfo
- Publication number
- WO2010015140A1 WO2010015140A1 PCT/CN2009/000843 CN2009000843W WO2010015140A1 WO 2010015140 A1 WO2010015140 A1 WO 2010015140A1 CN 2009000843 W CN2009000843 W CN 2009000843W WO 2010015140 A1 WO2010015140 A1 WO 2010015140A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- brightness
- value
- mapping table
- luminance
- image
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000002708 enhancing effect Effects 0.000 title abstract description 4
- 238000013507 mapping Methods 0.000 claims abstract description 157
- 238000012935 Averaging Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 78
- 230000011218 segmentation Effects 0.000 claims description 25
- 238000001514 detection method Methods 0.000 claims description 17
- 230000000694 effects Effects 0.000 abstract description 7
- 238000005286 illumination Methods 0.000 description 16
- 238000012545 processing Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000000007 visual effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/14—Picture signal circuitry for video frequency region
- H04N5/20—Circuitry for controlling amplitude response
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N9/00—Details of colour television systems
- H04N9/64—Circuits for processing colour signals
- H04N9/68—Circuits for processing colour signals for controlling the amplitude of colour signals, e.g. automatic chroma control circuits
- H04N9/69—Circuits for processing colour signals for controlling the amplitude of colour signals, e.g. automatic chroma control circuits for modifying the colour signals by gamma correction
Definitions
- Video enhancement method and device thereof Video enhancement method and device thereof
- the present invention relates to video enhancement techniques, and more particularly to a video enhancement method and apparatus therefor.
- Video enhancement is a basic method of video pre- and post-processing.
- video that can be played normally in a theater or TV will have two when playing on some portable devices such as mobile phones.
- the present invention enhances the detail and saturation of the image based on the improved partial histogram method, processes the data of the decompressed buffer, enhances the detail of the video image, and improves the video image on the mobile phone.
- the display effect is not a case where the present invention is not a case where the present invention is not a video image.
- the present invention enhances the detail and saturation of the image based on the improved partial histogram method, processes the data of the decompressed buffer, enhances the detail of the video image, and improves the video image on the mobile phone. The display effect.
- the technical problem to be solved by the present invention is to provide a video enhancement method and apparatus thereof for improving the visual effect of video images under low illumination and normal illumination conditions.
- the present invention provides a video enhancement method, the method comprising:
- the histogram of the frame image is calculated based on the luminance data thereof. And mapping the table, and then performing weighted averaging on the mapping values in the nonlinear stretch curve mapping table and the histogram equalization mapping table for each brightness value in the brightness value set of the pixel points of the frame image, and using the brightness
- the mapping table records the correspondence between the luminance values and the weighted averaged mapping values. Then, according to the luminance mapping table, the luminance values of the pixel points of the frame image are updated to the weighted average mapping values corresponding to the luminance values. , complete brightness enhancement.
- the step of performing weighted averaging on each of the luminance value sets of the pixel points of the frame image in the nonlinear stretch curve mapping table and the histogram equalization mapping table includes: ⁇ Use the following formula to perform a weighted average:
- MapTablAvg[k] (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M where k is a luminance value in the luminance value set of the pixel of the frame image; The number of brightness levels of the image;
- ⁇ 73 ⁇ 4 ⁇ , ⁇ ] is the corresponding mapping value of the luminance value k in the luminance mapping table, and the luminance mapping table is formed according to the above formula ⁇ 73 ⁇ 4 ⁇ ⁇ t];
- ImgClass is a brightness level of the frame image obtained by performing brightness detection on the frame image
- MapTabEq k] is a corresponding mapping value of the luminance value k in the histogram equalization mapping table
- MapTabGm[k] is the corresponding mapping value of the luminance value k in the nonlinear tensile curve mapping table. Further, the steps of calculating a histogram equalization mapping table of the frame image according to the luminance data thereof include:
- Step 2 Calculate the intermediate point of each histogram interval obtained by the segmentation, and then calculate the dynamic segmentation point c n of the histogram interval according to the following formula;
- ⁇ adjuent ⁇ is the first and end points of a histogram interval; “for pre-set The control parameter of the histogram equalization degree, 0 ⁇ « ⁇ 1; L* means rounding down, "* means rounding down, o ⁇ c voyage ⁇ (J -i);
- the number L of intervals into which the histogram is divided is 256, and the number of iterations is 8 times.
- the control parameter of the histogram equalization degree ranges from 0.4 to 0.6.
- each of the same sections is numbered and the number of the same section is consecutive; the luminance values in the same section correspond to a plurality of numbers, and the luminance values thereof are agreed upon by the mapping. Map to one of the numbers.
- the method further includes the step of performing saturation enhancement on the frame image by: setting a brightness degree level of the image and setting a saturation coefficient for each brightness degree level;
- saturation enhancement When the frame image is subjected to saturation enhancement, the brightness level of the frame image is first detected, and then the chrominance value of the pixel of the frame image is adjusted using the corresponding saturation coefficient.
- the brightness level of the frame image is determined according to the following manner: the configuration information of the brightness level of the frame image is saved, including the brightness level and the brightness value interval corresponding to each level; After the detection, the brightness value of the frame image is obtained, and the brightness level corresponding to the interval to which the brightness value belongs is the brightness degree level ImgC/o ⁇ of the frame image.
- the configuration information of the brightness level of the frame image is determined according to the following manner: dividing the brightness degree of the frame image into levels, and dividing the entire value range of the brightness value into corresponding levels In the interval, the interval with the lowest brightness value corresponds to a brightness degree level of 0, and the interval with the second lowest brightness value corresponds to a brightness degree level of 1, and the higher the brightness value corresponds to a higher brightness level, and will be equal to and All brightness values greater than the set normal brightness value threshold are grouped into an interval, and the corresponding brightness level is ⁇ /-1.
- the step of performing saturation enhancement on the frame image follows the formula below:
- Vout(x, y) S [m]x (Vin(x, y) - - ⁇ ) + - ⁇
- Uin(x, y) and Vin c, y) are the U-components of the pixel before saturation enhancement
- the value of the V component, f/oMt(x, and P3 ⁇ 4 M t(x, is the value of the U and V components of the pixel after saturation enhancement, and is the saturation coefficient corresponding to the degree of light and darkness in the array Su
- the saturation coefficient is 1, where L is the number of intervals into which the entire histogram is divided.
- the number L of intervals into which the entire histogram is divided is 256.
- a brightness enhancement module includes a nonlinear curve stretching table storage unit, a histogram equalization mapping table calculation unit, a brightness mapping table calculation unit, and an image brightness enhancement mapping unit, wherein:
- the non-linear curve stretching table storage unit is configured to hold a non-linear tensile curve map to be used
- the histogram equalization mapping table calculation unit is configured to calculate a corresponding histogram equalization mapping table according to the brightness data of each frame image to be processed;
- the brightness mapping table calculation unit is configured to perform weighted averaging on each of the brightness value sets of the pixel values of the pixel points of the frame image in the nonlinear stretch curve mapping table and the histogram equalization mapping table, and use one
- the brightness mapping table records a correspondence between each brightness value and a weighted averaged mapping value
- the image brightness enhancement mapping unit is configured to update the brightness value of the pixel of the image to be processed to the weighted averaged mapping value corresponding to the brightness value according to the brightness mapping table to complete the brightness enhancement.
- the brightness map calculation unit is configured to set a mapping value of each of the brightness value sets of the pixel points of the frame image in the nonlinear stretch curve mapping table and the histogram equalization mapping table according to the following formula.
- MapTablAvg[k] (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M where k is a luminance value in the set of luminance values of the pixel of the frame image;
- ⁇ 73 ⁇ 4 ⁇ , ⁇ ] is the corresponding mapping value of the luminance value k in the luminance mapping table, and the luminance mapping table is formed according to the above formula ⁇ 73 ⁇ 4 ⁇ ⁇ t];
- ImgClass is a brightness level of the frame image obtained by performing brightness detection on the frame image
- MapTabEq k] is a corresponding mapping value of the luminance value k in the histogram equalization mapping table
- MapTabGm[k] is the corresponding mapping value of the luminance value k in the nonlinear tensile curve mapping table.
- the intermediate point calculation unit is configured to calculate the intermediate point ⁇ ⁇ of each histogram section obtained by the division according to the histogram value of each luminance value k, and output Dynamic segmentation point calculation unit;
- the dynamic segmentation point calculation unit is configured to calculate a dynamic segmentation point of the histogram interval according to the following formula
- ⁇ а ⁇ is the first and end points of a histogram interval respectively; “The control parameter for the preset histogram equalization degree, 0 ⁇ « ⁇ 1; L*” means rounding down, “* means Round down,
- the control unit is configured to control the intermediate point calculation unit and the dynamic segmentation point calculation unit to perform an operation, and by n iterations, the entire histogram is divided into L intervals, and the generated L intervals and the brightness thereof are included.
- the value is output to the mapping table generating unit;
- control unit is arranged to divide the histogram into 256 intervals, and the number of iterations is eight.
- the present invention also provides an enhancement device comprising the brightness enhancement module.
- the video enhancement device further includes an image brightness detection module.
- the image brightness detection module includes a configuration information storage unit and a brightness level calculation unit, wherein:
- the configuration information storage unit is configured to save configuration information of a brightness level of the frame image, including a brightness level and a brightness value interval corresponding to each level;
- the brightness level calculation unit is configured to calculate a brightness value of the frame image according to the received brightness data of the image to be processed, and determine, according to the configuration information, a brightness level corresponding to the interval to which the brightness value belongs. And outputting the brightness level as the brightness map level of the frame image to the brightness map calculation unit in the brightness enhancement module.
- the brightness level has M levels, and the entire value range of the brightness value is divided into corresponding intervals corresponding to the level, and the brightness level corresponding to the lowest brightness level is 0, brightness
- the interval of the lower value corresponds to a brightness degree level of 1, the higher the brightness value, the higher the brightness level corresponding to the interval, and all the brightness values equal to and greater than the set normal brightness value threshold are classified into one interval.
- the interval corresponding to this interval is ⁇ /-1.
- the video enhancement device further includes a saturation enhancement module, where the saturation enhancement module includes a configuration information storage unit and a saturation enhancement calculation unit, where:
- the brightness level calculation unit in the image brightness detection module is further configured to output a brightness level level of the image to be processed to the saturation enhancement calculation unit in the saturation enhancement module;
- the configuration information storage unit is configured to save a brightness degree level of the set image, and a saturation coefficient corresponding to each brightness level level;
- the saturation enhancement calculation unit is configured to perform saturation enhancement on a frame of image to be processed At the time, the chromaticity value of the pixel of the frame image is adjusted according to the received brightness degree level of the frame image using the corresponding saturation coefficient.
- the gray range of the pixel value is appropriately extended while avoiding gray scale
- the levels are excessively combined to solve the image distortion problem caused by the traditional histogram equalization method, and the visual effect of the image is obviously improved, and a higher quality output image is obtained.
- the processed video image is clearer than the source image under low illumination, and is more prominent than the source image under normal illumination, and the effect is better, which is in line with the human eye.
- the method has small calculation amount and can meet the requirement of real-time performance, and has strong practical value.
- the method of the present invention can be set not only to an ATM, a wireless environment, but also to the processing of video decoded images in Internet video communication. Especially suitable for video enhancement of mobile TV.
- FIG. 1 is a schematic diagram of a position of a video enhancement device in a system according to an embodiment of the present invention
- FIG. 2 is a block diagram of the video enhancement device of Figure 1;
- FIG. 3 is a block diagram of the unit of the brightness enhancement module of Figure 2;
- FIG. 4 is a unit structure diagram of the saturation enhancement module of FIG. 2;
- FIG. 5 is a flowchart of an algorithm of the histogram equalization mapping table calculation unit of FIG. 3;
- Figure 6 is a schematic illustration of the tensile curve used in the examples of the present invention.
- the video enhancement method of the present invention is not limited to video enhancement of mobile TV, and may be set to other occasions where video enhancement is required.
- the video enhancement device 104 performs video enhancement processing on the source image data output from the input data buffer 102, and then processes the processed target.
- the target image data is output to the output data buffer 106.
- the source image processed in this embodiment is an image in the YUV format, but the present invention can also process an image in the RGB format.
- FIG. 2 shows a block diagram of video enhancement device 104 of FIG. 1, video enhancement device 104 including image brightness detection module 202, brightness enhancement module 204, and saturation enhancement module 206.
- the luminance data of the source image data (e.g., the value of the Y component) is input to the image brightness detection module 202 and the brightness enhancement module 204
- the source image chrominance data (e.g., the values of the U and V components) is input to the saturation enhancement module 206.
- the image brightness detection module 202 outputs the brightness level of the detected image to the brightness enhancement module 204 and the saturation enhancement module 206
- the brightness enhancement module 204 enhances the brightness value of the pixel of the image according to the brightness level.
- the enhanced brightness value is output, and the saturation enhancement module 206 enhances the chrominance value of the pixel of the image according to the brightness level and outputs the enhanced chromaticity value.
- the image light and dark detection module 202 further includes a configuration information storage unit and a light and darkness level calculation unit, wherein:
- the configuration information storage unit is configured to save the configuration information of the brightness level, and the configuration information of the brightness level includes the brightness level and the brightness value interval corresponding to each level.
- the brightness and darkness of the image are divided into levels, and the entire range of the brightness value is divided into intervals corresponding to the M levels, and the brightness level corresponding to the lowest brightness level is 0, and the brightness is
- the lower value interval corresponds to a brightness level of 1, ..., and so on, and all brightness values equal to and greater than the set normal brightness value threshold are classified into an interval corresponding to the interval.
- the degree of light and darkness is ⁇ ;
- the brightness level calculation unit is configured to calculate a brightness value of the source image according to the brightness data of the received source image, and determine, according to the configuration information, a brightness level corresponding to the interval to which the brightness value belongs, and the brightness degree The level is output as the brightness level of the source image to the brightness map calculation unit 306 in the brightness enhancement module 204, and the saturation enhancement calculation unit 404 in the saturation enhancement module 206.
- the luminance map calculation unit 306 and the saturation enhancement calculation unit 404 will be described below.
- 3 is a unit configuration diagram of the brightness enhancement module 204 of FIG. 2.
- the brightness enhancement module 204 includes a non-linear curve stretching table storage unit 302, a histogram equalization mapping table calculation unit 304, a brightness mapping table calculation unit 306, and an image brightness enhancement map.
- Unit 308 wherein:
- the nonlinear curve stretching table storage unit 302 is arranged to hold a non-linear tensile curve mapping table to be used.
- the histogram equalization mapping table calculation unit 304 is configured to calculate a corresponding histogram equalization mapping table according to the input source image luminance data, and the specific method will be described below.
- the brightness mapping table calculation unit 306 is configured to perform at least each of the brightness value sets of the frame image pixel points, and perform the brightness value on the non-linear stretch curve mapping table and the histogram equalization mapping table.
- the weighted average is used, and a correspondence relationship between each luminance value and the weighted averaged mapping value is recorded by a luminance mapping table.
- the specific calculation formula is shown below.
- the image brightness enhancement mapping unit 308 is configured to update the brightness value of the pixel of the input source image to the weighted averaged mapping value corresponding to the brightness value according to the brightness mapping table.
- the saturation enhancement module 206 includes a configuration information storage unit 402 and a saturation enhancement calculation unit 404, where:
- the configuration information holding unit 402 is configured to save the brightness degree level of the set image, and the saturation coefficient corresponding to each brightness level level, or the chromaticity value of the source image pixel point and the saturation enhancement. Correspondence of chromaticity values of pixel points of the target image.
- the saturation enhancement calculation unit 404 is configured to adjust the chrominance value of the pixel of the source image using the corresponding saturation coefficient according to the brightness level of the received source image when the saturation enhancement is performed on the source image, or The chromaticity map is searched according to the input source image chromaticity data, that is, the chromaticity value of the pixel and the brightness degree level of the detected image, to obtain the chromaticity value of the target image pixel after the saturation enhancement.
- the histogram equalization map calculation unit 304 further includes a luminance histogram calculation unit, an intermediate point calculation unit, a dynamic division point calculation unit, a control unit, and a mapping table generation unit (not shown), wherein:
- the intermediate point calculating unit is configured to calculate the intermediate point ⁇ ⁇ of each histogram interval obtained by the segmentation according to the histogram value of each brightness value, and output to the dynamic dividing point calculating unit;
- the dynamic segmentation point calculation unit is configured to calculate the dynamic segmentation point of the histogram interval according to the following formula: Where: ⁇ vine ⁇ is the first and end points of a histogram interval respectively; “The control parameter for the preset histogram equalization degree, 0 ⁇ « ⁇ 1; L*” means rounding down, “* means Round down, 0 ⁇ C Conduct ⁇ 255;
- the control unit is configured to control the intermediate point calculation unit and the dynamic division point calculation unit to perform an operation, and by calling 8 iterations, dividing the entire histogram into 256 intervals, and generating the generated 256 intervals and the brightness values thereof Output to the mapping table generation unit;
- the mapping table generating unit is configured to sequentially number each interval from 255 to 255 in order of the first brightness value included in each interval, and to assign all the brightness values included in each interval to the mapping equal to the interval number.
- the value after replacing each luminance value in the interval [0, 255] with the corresponding mapping value, obtains the histogram equalization mapping table.
- Step 10 Calculate an image brightness histogram
- Step 30 Calculate an intermediate point of each histogram interval obtained by the current segmentation
- Step 40 Calculate a dynamic segmentation point of each histogram interval
- Step 50 it is determined whether 8 splits have been performed, and if so, step 70 is performed, otherwise, step 60 is performed;
- Step 60 Reset the first end point and the end point of the segment after the segmentation
- Step 70 Set a histogram equalization mapping table according to the 256 histogram intervals obtained after the segmentation.
- the above intermediate points, the calculation of the dynamic segmentation points, and the specific method of setting the histogram equalization mapping table according to the histogram interval are disclosed in detail below, and are not described herein again.
- the method of this embodiment includes the following steps: In the first step, first determine the nonlinear tensile curve mapping table to be used ⁇ 3 ⁇ 4 ;
- the nonlinear tensile curve (or piecewise linear) is characterized by a slope of more than one per point at low illumination.
- the curve in Figure 6 is an example of such a curve.
- a gamma curve with a gamma coefficient of less than 1 is also a similar curve.
- Using the nonlinear stretch curve mapping table 3 73 ⁇ 4 (3 ⁇ 4 can be used to increase the brightness of low-light images. According to the mapping table, each discrete luminance value of a [0, 255] interval can be mapped to the luminance value of another [0, 255] interval. Finally, a 256-point one-dimensional mapping table is generated.
- the nonlinear stretching curve mapping table may use an existing mapping table, or may be selected according to actual effects by processing a sufficient number of images.
- the third step is to calculate an intermediate point of each histogram interval obtained by the segmentation
- the middle point of the histogram interval is such that A, 0 ⁇ ⁇ ⁇ ⁇ 255 of the following formula:
- ⁇ h[k] ⁇ h[k] into which: / ] is the histogram value of the luminance value ⁇ , that is, the number of pixels having the luminance value; "ander, respectively the first endpoint and the end point of a histogram interval, Indicates the first iteration calculation.
- the first and last points of the histogram interval are 0 and 255, respectively.
- the number of pixels in the histogram of the interval is discrete, when writing the function, first calculate the total number of pixels in the interval, and then accumulate the number of pixels from the first endpoint, if the accumulated number of pixels is greater than or equal to the total number of pixels Half of the point is the point in the histogram found.
- the dynamic segmentation point of the histogram interval is calculated according to the following formula: c:
- the dynamic segmentation point c garbage divides the histogram interval with the first end and the end A into two intervals, and c n can be attributed to one of the two intervals when segmentation. Since the interval here contains discrete brightness values, after division The first end point of the interval may be the same, that is, only one luminance value is included, and when the interval is divided again, the same two intervals including the luminance value are obtained.
- the third step and the fourth step are repeated.
- the entire histogram can be divided into 256 intervals;
- Some of the 256 sections divided may include multiple luminance values, or there may be intervals where the first end points are the same.
- the divided 256 intervals are numbered from 0 to 255 according to the first brightness value of the first brightness value, and then all the brightness values included in each interval are equal to the interval number.
- the mapping value after replacing each luminance value in the interval [0, 255] with the corresponding mapping value, obtains a histogram equalization mapping table ⁇ ⁇ ⁇ ⁇ ⁇ .
- the histogram equalization map M ⁇ 73 ⁇ 4 3 ⁇ 4 is also a one-dimensional map in the interval [0, 255].
- the formula in the fourth step is a common method for obtaining a dynamic segmentation point when the histogram is equalized.
- the resulting histogram equalization map is also the same as that obtained by the known method.
- the histogram equalization mapping table obtained by adding the control parameter parameters can prevent the gray level from being excessively merged while appropriately expanding the dynamic range of the pixel values.
- the brightness of the frame image is detected to determine which brightness level the detected brightness value belongs to;
- the brightness and darkness of the image can be divided into levels, and the entire range of the brightness value is divided into intervals corresponding to the level.
- the corresponding The brightness level is 0, the brightness value is the second lowest interval, the corresponding brightness level is 1 , ..., and so on, the image brightness value is normal or above, and the corresponding brightness level is For M- ⁇ .
- a normal brightness threshold (which can be set between 110 and 135) can be set, and the brightness value greater than or equal to the normal brightness threshold is classified into the brightness value interval corresponding to the brightness level level ⁇ .
- the frame image is detected to obtain a brightness value of the frame image, and the brightness level corresponding to the interval to which the brightness value belongs is the brightness level of the frame image.
- MapTablAvg[k] (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M where:
- MapTablAvg[k] is the mapping value corresponding to the luminance value k in the luminance mapping table Map73 ⁇ 4Mvg, and MapTablAv ⁇ k] calculated according to the above formula constitutes MapTablAvg.
- ImgC/o ⁇ is the level of brightness and darkness of the detected image of the frame.
- MapTabEq k] is the mapping value corresponding to the luminance value in the histogram equalization mapping table ⁇ .
- MapTabGm[k] is the corresponding mapping value of the luminance value in the nonlinear stretch curve mapping table Map73 ⁇ 4G.
- the brightness value k may be each brightness value in [0, 255], or the brightness value set of the pixel of the image may be detected first, and only the brightness value in the brightness value set is calculated. This can reduce the amount of calculations.
- the weighted average of the histogram equalization map and the nonlinear stretch curve map is actually performed.
- the weights are adaptively adjusted according to the brightness of the image.
- the weights used are not limited.
- the algorithm of the above formula can also directly define a set of weights corresponding to the level of brightness and darkness for calculation. Because the nonlinear stretch curve mapping can improve the brightness of low illumination images. However, the details of the image cannot be enhanced. Therefore, the present embodiment can achieve the optimized brightness enhancement effect by combining the two methods and using different weights according to the brightness of the detected image. Even if a histogram equalization map obtained by a known method is used, it is superior to the existing image enhancement processing.
- the brightness value of the pixel of the frame image is mapped according to the generated brightness mapping table to achieve the brightness enhancement effect;
- the frame image is subjected to saturation enhancement according to the saturation coefficient.
- Y represents the luminance component, which is enhanced by the methods from the first step to the ninth step, (U-128)/(V-128) represents the hue, ( ⁇ -128) 2 + ( -128) 2 means saturation, the principle of adjusting the saturation is to keep (U-128) / (V-128) unchanged, increase the saturation of the image,
- brightness enhancement and saturation enhancement can be performed simultaneously.
- Su saturation coefficient corresponding to each brightness level
- w the brightness level of the set image
- m 0,1 , . . . M-1
- M the brightness of the set image
- the coefficients in the array are the saturation coefficients at different levels of brightness and darkness. This value can be selected and adjusted according to the experiment.
- the set of coefficients given above is only an example. When the brightness level is M-1, that is, under normal illumination or above, the saturation enhancement is not performed, so the coefficient is 1.
- the saturation enhancement can be solved by using the method of calculating the table in advance, which can speed up the calculation; it can also be calculated when the enhancement is performed.
- the formula is:
- Vout(x, y) Su[m]x (Vm(x, y) - 128) + 128 Uin(x, y) and Vin c, y) are the U and V components of the pixel before saturation enhancement
- the value, f/oMt(x, and P3 ⁇ 4 M t(x, is the value of the U and V components of the pixel after saturation enhancement.
- luminance and chrominance values of the image of the present invention may also use the luminance value and the chrominance value of the RGB space, as for how the luminance value and the chrominance value of the image and the pixel point in the RGB space are determined, in the prior art. There are many discussions in the middle, and no further explanation is given here.
- the invention detects the image of different illumination conditions, and enhances the video or image under the low illumination image and the normal illumination according to the detection result, and appropriately expands the dynamic range of the pixel value while avoiding the gray level being excessively
- the combination solves the problem of image distortion caused by the traditional histogram equalization method, and obviously improves the visual effect of the image, and obtains a higher quality output image.
- the processed video image is clearer than the source image under low illumination, and is more prominent than the source image under normal illumination, and the effect is better, which is in line with the human eye.
- the method of the invention has small calculation amount and can meet the requirement of real-time performance, and can be set not only into ATM, wireless environment, but also can be set as processing of video decoding image in Internet video communication, especially suitable for video enhancement of mobile TV. Therefore, it has industrial applicability.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
- Picture Signal Circuits (AREA)
Description
一种视频增强的方法及其装置
技术领域
本发明涉及视频增强技术, 尤其是指一种视频增强的方法及其装置。
背景技术
视频增强是视频前后处理的一种基本手段, 在一般情况下, 由于手机终 端成本和能耗的限制, 在影院或电视中能正常播放的视频, 在一些便携式设 备如手机中播放时会有两个缺点:
1 )在黑夜或昏暗场景下的视频, 在手机上显示会更暗, 以至于在室内外 正常或过亮环境下看不清画面;
2 )正常光照条件下的视频在影院或电视中细节等能显示的很好, 但在手 机上显示时, 由于屏幕分辨率和颜色分辨率的限制, 场景会有些模糊, 丟失 了部分细节, 特别是经过压缩解压后的视频, 这种现象会更严重;
在第一种情况下, 要么是增加显示屏的亮度, 要么通过图像处理的方法 进行增强, 前者会快速增大手机的能耗, 以至于有时候一部视频都不能播放 完, 所以在本发明中釆用图像处理的方法, 能使得暗场景视频既能显示清楚, 又不至于使手机功耗过大。 在第二种情况下, 本发明釆用一种基于改进的部 分直方图方法增强图像的细节和饱和度, 处理解压后緩冲区的数据, 能够增 强视频图像的细节, 改善视频图像在手机上的显示效果。
发明内容
本发明所要解决的技术问题是提供一种视频增强的方法及其装置, 以改 善低照度和正常照度条件下视频图像的视觉效果。
为了解决上述技术问题, 本发明提供了一种视频增强的方法, 该方法包 括:
设置非线性拉伸曲线映射表;
对待处理的每一帧图像, 根据其亮度数据计算得到该帧图像的直方图均
衡映射表, 然后将该帧图像的像素点的亮度值集合中的每一亮度值在所述非 线性拉伸曲线映射表和所述直方图均衡映射表中的映射值进行加权平均, 并 用亮度映射表记录各亮度值与加权平均后的映射值之间的对应关系; 然后根 据所述亮度映射表, 将该帧图像的像素点的亮度值更新为该亮度值对应的加 权平均后的映射值, 完成亮度增强。
进一步地, 将该帧图像的像素点的亮度值集合中的每一亮度值在所述非 线性拉伸曲线映射表和所述直方图均衡映射表中的映射值进行加权平均的所 述步骤包括釆用下式进行加权平均:
MapTablAvg[k] = (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M 式中, k为该帧图像的像素点的亮度值集合中的一亮度值; 为设定的 图像亮暗程度等级的数目;
Μ 7¾Μ、^^]为亮度值 k在所述亮度映射表中对应的映射值, 根据上式 计算出来的 Μ 7¾Μ ^t]构成了所述亮度映射表;
ImgClass为对该帧图像进行亮度检测后得到的该帧图像的亮暗程度等 级;
MapTabEq k]为亮度值 k在所述直方图均衡映射表中对应的映射值;
MapTabGm[k]为亮度值 k在所述非线性拉伸曲线映射表中对应的映射值。 进一步地, 根据其亮度数据计算得到该帧图像的直方图均衡映射表的所 述步骤包括:
步骤一, 对待处理的一帧图像, 先计算该帧图像的亮度直方图, 得到每 一个亮度值 的直方图数值即具有该亮度值的像素点数, A=0,l,2,..丄-1 , 其 中, L为将整个直方图分成的区间数;
步骤二, 计算分割得到的每一个直方图区间的中间点 然后根据下式 计算直方图区间的动态分割点 cn;
其中: α„ Α分别为一个直方图区间的首端点和末端点; "为预先设定的
直方图均衡程度的控制参数, 0 < « < 1; L*」表示向下取整, 「*表示向下取整, o < c„ < (J -i);
步骤三, 重复执行步骤二, 通过 n次迭代地调用, 将整个直方图分成 L 个区间,并按各区间包含的首个亮度值按从小到大的顺序依次对各区间编号, 从 0开始依次对各区间编号直到 L -1 , 然后将每个区间包含的所有亮度值均 对应到等于该区间编号的映射值, 将区间 [0, L -1]上的每一亮度值替换为对应 的映射值后, 即得到所述直方图均衡映射表, 其中, 2n= L。
较佳地, 将直方图分成的区间数 L为 256, 则所述迭代次数为 8次。 较佳地, 所述直方图均衡程度的控制参数"的取值范围是 0.4~0.6。
进一步地, 所分成的区间存在相同区间时, 对相同区间中的每一个区间 分别编号且相同区间的编号连续; 相同区间中的亮度值对应多个编号, 在映 射时按约定将其中的亮度值映射到其中的一个编号。
进一步地, 完成亮度增强后, 该方法还包括通过以下处理对该帧图像进 行饱和度增强的步骤: 设定图像的亮暗程度等级并为每一个亮暗程度等级对 应设置一个饱和度系数; 对该帧图像进行饱和度增强时, 先检测出该帧图像 的亮暗程度等级, 然后使用对应的饱和度系数对该帧图像的像素点的色度值 进行调整。
较佳地, 该帧图像的亮暗程度等级根据以下方式确定: 保存该帧图像的 亮暗程度等级的配置信息,包括亮暗程度等级及每一等级对应的亮度值区间; 对该帧图像进行检测后得到该帧图像的亮度值, 该亮度值所属区间对应的亮 暗程度等级即为该帧图像的亮暗程度等级 ImgC/o^。
进一步地, 该帧图像的亮暗程度等级的配置信息按照以下方式确定: 将 该帧图像的亮暗程度分成 个等级,将亮度值的整个取值范围划分为与该 个等级——对应的 个区间, 亮度值最低的区间对应的亮暗程度等级为 0, 亮度值次低的区间对应的亮暗程度等级为 1 , 亮度值越高的区间对应的亮暗 程度等级越高, 且将等于和大于设定的正常亮度值门限的所有亮度值归入一 个区间, 该区间对应的亮暗程度等级为 Λ/-1。
进一步地,帧图像是 YUV格式时,对该帧图像进行饱和度增强的所述步
骤按以下公式进行:
Uo t(x,y) = S [m] x (Uin(x, ) --^) + -^
Vout(x, y) = S [m]x (Vin(x, y) - -^) + -^ 其中, Uin(x, y)和 Vin c, y)是进行饱和度增强之前像素点 U分量和 V分量 的值, f/oMt(x, 和 P¾Mt(x, 是进行饱和度增强之后像素点 U分量和 V分量的 值, 为数组 Su中与亮暗程度等级对应的饱和度系数, 数组 Su中记录有 每一个亮暗程度等级对应的饱和度系数, m = 0,1 , . . .M-1 , 而 0,1 , . . .M-1为设 定的图像亮暗程度等级, 且亮暗程度等级为 M-1时, 其饱和度系数为 1 , 其 中, L为将整个直方图分成的区间数。
较佳地, 将整个直方图分成的区间数 L为 256。
一种亮度增强模块, 该亮度增强模块包括非线性曲线拉伸表存储单元、 直方图均衡映射表计算单元、亮度映射表计算单元和图像亮度增强映射单元, 其中:
所述非线性曲线拉伸表存储单元设置成保存要使用的非线性拉伸曲线映 射表;
所述直方图均衡映射表计算单元设置成对待处理的每一帧图像, 根据其 亮度数据计算得到相应的直方图均衡映射表;
所述亮度映射表计算单元设置成将该帧图像的像素点的亮度值集合中的 每一亮度值在非线性拉伸曲线映射表和直方图均衡映射表中的映射值进行加 权平均, 并用一亮度映射表记录各亮度值与加权平均后的映射值之间的对应 关系;
所述图像亮度增强映射单元设置成根据所述亮度映射表, 将待处理的一 帧图像的像素点的亮度值更新为该亮度值对应的加权平均后的映射值, 完成 亮度增强。
进一步地, 所述亮度映射表计算单元设置成按下式将该帧图像的像素点 的亮度值集合中的每一亮度值在非线性拉伸曲线映射表和直方图均衡映射表 中的映射值进行加权平均:
MapTablAvg[k] = (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M 式中, k为该该帧图像的像素点的亮度值集合中的一亮度值;
为设定的图像亮暗程度等级的数目;
Μ 7¾Μ、^^]为亮度值 k在所述亮度映射表中对应的映射值, 根据上式 计算出来的 Μ 7¾Μ ^t]构成了所述亮度映射表;
ImgClass为对该帧图像进行亮度检测后得到的该帧图像的亮暗程度等 级;
MapTabEq k]为亮度值 k在所述直方图均衡映射表中对应的映射值;
MapTabGm[k]为亮度值 k在所述非线性拉伸曲线映射表中对应的映射值。 进一步地, 所述直方图均衡映射表计算单元包括亮度直方图计算单元、 中间点计算单元、 动态分割点计算单元、 控制单元和映射表生成单元, 其中: 所述亮度直方图计算单元设置成计算出待处理的一帧图像的亮度直方 图, 得到每一个亮度值 的直方图数值即具有该亮度值的像素点数, 输出到 中间点计算单元, =0,1,2,... L -1 , 其中, L为将整个直方图分成的区间数; 所述中间点计算单元设置成根据每一个亮度值 k的直方图数值, 计算分 割得到的每一个直方图区间的中间点 βη , 输出到动态分割点计算单元;
所述动态分割点计算单元设置成根据下式计算直方图区间的动态分割点
0 < C„ < (J -1);
所述控制单元设置成控制所述中间点计算单元和动态分割点计算单元进 行运算, 通过 n次迭代地调用, 将整个直方图分成 L个区间, 并将生成的 L 个区间及其包含的亮度值输出到映射表生成单元;
所述映射表生成单元设置成按各区间包含的首个亮度值从小到大的顺序
从 0开始依次对各区间编号直到 L-1 , 将每个区间包含的所有亮度值均对应 到等于该区间编号的映射值, 将区间 [0, L-1]上的每一亮度值替换为对应的映 射值后, 即得到所述直方图均衡映射表, 其中, 2n= L。
较佳地, 所述控制单元设置成将直方图分成 256个区间, 则所述迭代次 数为 8次。
本发明还提供了一种增强装置, 其包括所述亮度增强模块。
进一步地, 所述视频增强装置还包括一图像亮暗检测模块,
该图像亮暗检测模块包括配置信息存储单元和亮暗程度等级计算单元, 其中:
所述配置信息存储单元设置成保存该帧图像的亮暗程度等级的配置信 息, 包括亮暗程度等级及每一等级对应的亮度值区间;
所述亮暗程度等级计算单元设置成根据接收到的待处理的一帧图像的亮 度数据计算出该帧图像的亮度值, 根据所述配置信息判断出该亮度值所属区 间对应的亮暗程度等级, 将该亮暗程度等级作为该帧图像的亮暗程度等级输 出到所述亮度增强模块中的亮度映射表计算单元。
较佳地, 所述亮暗程度等级有 M个等级, 亮度值的整个取值范围划分为 与该 个等级——对应的 个区间 ,亮度值最低的区间对应的亮暗程度等级 为 0, 亮度值次低的区间对应的亮暗程度等级为 1 , 亮度值越高的区间对应的 亮暗程度等级越高, 且将等于和大于设定的正常亮度值门限的所有亮度值归 入一个区间, 该区间对应的亮暗程度等级为 Λ/-1。
较佳地, 所述视频增强装置进一步包括一饱和度增强模块, 该饱和度增 强模块包括配置信息存储单元和饱和度增强计算单元, 其中:
所述图像亮暗检测模块中的亮暗程度等级计算单元还设置成将待处理的 一帧图像的亮暗程度等级输出到所述饱和度增强模块中的饱和度增强计算单 元;
所述配置信息存储单元设置成保存设定的图像的亮暗程度等级, 以及为 每一个亮暗程度等级对应设置的饱和度系数;
所述饱和度增强计算单元设置成对待处理的一帧图像进行饱和度增强
时, 根据接收到的该帧图像的亮暗程度等级, 使用对应的饱和度系数对该帧 图像的像素点的色度值进行调整。
釆用本发明的方法, 通过对不同光照条件图像进行检测、 并根据检测结 果对低照度图像和正常照度下的视频或图像进行增强, 在适当扩展像素取值 的动态范围的同时尽量避免灰度级被过多地合并, 解决传统直方图均衡化方 法带来的图像失真问题, 明显地改善了图像的视觉效果, 得到较高品质的输 出图像。 使得处理后的视频图像在低照度下比源图像更清晰, 在正常照度下 比源图像细节突出, 效果更好, 符合人眼的视觉。
该方法计算量小, 可以满足实时性的需求, 具有较强的实用价值。 本发 明的方法不但可以设置成 ATM、 无线环境, 而且也可设置成 Internet视频通 信中视频解码图像的处理。 尤其适设置成手机电视的视频增强。
附图概述
图 1是本发明实施例视频增强装置在***中的位置的示意图;
图 2是图 1中视频增强装置的模块图;
图 3是图 2中亮度增强模块的单元结构图;
图 4是图 2中饱和度增强模块的单元结构图;
图 5是图 3中直方图均衡映射表计算单元的算法流程图;
图 6是本发明实施例所用的拉伸曲线的示意图。
本发明的较佳实施方式
以下以手机电视的视频增强为例, 结合附图对本发明的具体实施方式进 行说明。 不过本发明的视频增强方法并不局限于手机电视的视频增强, 同样 可以应设置成其它需要进行视频增强的场合。
图 1示出了整个视频增强装置在***中的位置, 该视频增强装置 104对 输入数据緩冲 102输出的源图像数据进行视频增强处理, 然后将处理后的目
标图像数据输出到输出数据緩冲 106。本实施例处理的源图像是 YUV格式的 图像, 不过本发明也可以对 RGB格式的图像进行处理。
图 2示出了图 1中视频增强装置 104的模块图, 视频增强装置 104包括 图像亮暗检测模块 202、 亮度增强模块 204和饱和度增强模块 206。 源图像数 据的亮度数据 (如 Y分量的值 )输入到图像亮暗检测模块 202和亮度增强模 块 204, 源图像色度数据(如 U、 V分量的值)输入到饱和度增强模块 206。 图像亮暗检测模块 202将检测出的图像的亮暗程度等级输出到亮度增强模块 204和饱和度增强模块 206 ,亮度增强模块 204根据该亮暗程度等级对图像的 像素点的亮度值进行增强并输出增强后的亮度值, 饱和度增强模块 206根据 该亮暗程度等级对图像的像素点的色度值进行增强并输出增强后的色度值。
图像亮暗检测模块 202进一步包括配置信息存储单元和亮暗程度等级计 算单元, 其中:
配置信息存储单元, 设置成保存亮暗程度等级的配置信息, 亮暗程度等 级的配置信息包括亮暗程度等级及每一等级对应的亮度值区间。
本实施例是将图像的亮暗程度分成 个等级, 亮度值的整个取值范围划 分为与该 M个等级——对应的 个区间,亮度值最低的区间对应的亮暗程度 等级为 0, 亮度值次低的区间对应的亮暗程度等级为 1 , ... ... , 以此类推, 且 将等于和大于设定的正常亮度值门限的所有亮度值归入一个区间, 该区间对 应的亮暗程度等级为 ΜΛ;
亮暗程度等级计算单元, 设置成根据接收到的源图像的亮度数据计算出 源图像的亮度值, 根据所述配置信息判断出该亮度值所属区间对应的亮暗程 度等级, 将该亮暗程度等级作为源图像的亮暗程度等级输出到亮度增强模块 204中的亮度映射表计算单元 306,以及饱和度增强模块 206中的饱和度增强 计算单元 404。
亮度映射表计算单元 306和饱和度增强计算单元 404将在下面介绍。 图 3是图 2中亮度增强模块 204的单元结构图, 亮度增强模块 204包括 非线性曲线拉伸表存储单元 302、 直方图均衡映射表计算单元 304、 亮度映射 表计算单元 306和图像亮度增强映射单元 308, 其中:
非线性曲线拉伸表存储单元 302, 设置成保存要使用的非线性拉伸曲线 映射表。
直方图均衡映射表计算单元 304, 设置成根据输入的源图像亮度数据计 算得到相应的直方图均衡映射表, 具体方法将在下文中描述。
亮度映射表计算单元 306 , 设置成至少对该帧图像像素点的亮度值集合 中的每一亮度值, 将该亮度值在非线性拉伸曲线映射表和直方图均衡映射表 中的映射值进行加权平均, 并用一亮度映射表记录各亮度值与加权平均后的 映射值之间的对应关系。 具体的计算公式参见下文。
图像亮度增强映射单元 308 , 设置成根据所述亮度映射表, 将输入的源 图像的像素点的亮度值更新为该亮度值对应的加权平均后的映射值。
图 4是图 2中饱和度增强模块 206的单元结构图, 饱和度增强模块 206 包括配置信息存储单元 402和饱和度增强计算单元 404, 其中:
配置信息保存单元 402 , 设置成保存设定的图像的亮暗程度等级, 以及 为每一个亮暗程度等级对应设置的饱和度系数, 或者源图像像素点的色度值 与进行饱和度增强后的目标图像像素点的色度值的对应关系。
饱和度增强计算单元 404 , 设置成对源图像进行饱和度增强时, 根据接 收到的源图像的亮暗程度等级, 使用对应的饱和度系数对源图像的像素点的 色度值进行调整, 或者根据输入的源图像色度数据即像素点的色度值以及检 测到的图像的亮暗程度等级, 查找色度映射表, 得到饱和度增强后的目标图 像像素点的色度值。
所述直方图均衡映射表计算单元 304进一步包括亮度直方图计算单元、 中间点计算单元、动态分割点计算单元、控制单元和映射表生成单元(图略), 其中:
亮度直方图计算单元, 设置成计算出源图像的亮度直方图, 得到每一个 亮度值 k的直方图数值即具有该亮度值的像素点数,输出到中间点计算单元, =0,1,2,...255;
中间点计算单元, 设置成根据每一个亮度值 的直方图数值, 计算分割 得到的每一个直方图区间的中间点 βη , 输出到动态分割点计算单元;
动态分割点计算单元,设置成根据下式计算直方图区间的动态分割点 c„
其中: α„ Α分别为一个直方图区间的首端点和末端点; "为预先设定的 直方图均衡程度的控制参数, 0 < « < 1; L*」表示向下取整, 「*表示向下取整, 0 < C„ < 255;
控制单元, 设置成控制所述中间点计算单元和动态分割点计算单元进行 运算, 通过 8次迭代地调用, 将整个直方图分成 256个区间, 并将生成的 256 个区间及其包含的亮度值输出到映射表生成单元;
映射表生成单元, 设置成按各区间包含的首个亮度值从小到大的顺序从 0开始依次对各区间编号直到 255 ,将每个区间包含的所有亮度值均对应到等 于该区间编号的映射值, 将区间 [0,255]上的每一亮度值替换为对应的映射值 后, 即得到所述直方图均衡映射表。
图 5中给出直方图均衡映射表计算单元 304的处理流程, 包括以下步骤: 步骤 10 , 计算图像亮度直方图;
步骤 20 , 初始化直方图区间的首末端点《。 = 0 , = 255;
步骤 30 , 计算当前分割得到的每一个直方图区间的中间点;
步骤 40 , 计算每一个直方图区间的动态分割点;
步骤 50 , 判断是否已进行 8次分割, 如是, 执行步骤 70 , 否则, 执行步 骤 60;
步骤 60 , 重设分割后区间的首端点和末端点;
步骤 70 , 按分割后得到的 256个直方图区间设置直方图均衡映射表。 上述中间点、 动态分割点的计算, 以及根据直方图区间设置直方图均衡 映射表的具体方法在下文中有详细的公开, 这里不再赘述。
基于以上装置, 本实施例方法包括以下步骤:
第一步, 首先确定要使用的非线性拉伸曲线映射表 Μ ¾ ;
该非线性拉伸曲线 (或者是分段线性)的特征是在低照度下每点的斜率要 大于 1。 如图 6中的曲线就是该类曲线的一个示例。 gamma系数小于 1的伽 马曲线也是类似的曲线。利用非线性拉伸曲线映射表 Μ 7¾(¾可以用来提高 低照度图像的亮度。 根据该映射表可以把一个 [0,255]区间的每个离散亮度值 映射到另一个 [0,255]区间的亮度值, 最终生成一个 256点的一维映射表。 该 非线性拉伸曲线映射表可以釆用已有的映射表, 也可以通过对足够数量的图 像进行处理后根据实际的效果来选择。
第二步, 对待处理的一帧图像, 先计算该帧图像的亮度直方图,得到每一 个亮度值^的直方图数值即具有该亮度值的像素点数, =0,1,2,...255;
第三步, 计算分割得到的每一个直方图区间的中间点;
直方图区间的中间点即使得下式成立的 A , 0 < βη < 255:
k=Pn k=bn
∑h[k] =∑h[k] 成 其中: / ]为亮度值^的直方图数值即具有该亮度值的像素点数;《„ , 分 别为一个直方图区间的首端点和末端点, 《表示第《次迭代计算, 在第一次迭 代计算时, 直方图区间的首端点和末端点分别为 0和 255。
由于在该区间的直方图中像素的个数为离散的, 在编写函数时, 先计算 该区间总的像素点数, 然后再从首端点累加像素点数, 如果累加的像素点数 大于或等于总像素点数的一半, 即为找到的直方图中点 。
第四步, 根据下式计算直方图区间的动态分割点 c„:
其中: , 分别为一个直方图区间的首端点和末端点; A为第二步计算 出的直方图区间中间点; "为预先设定的直方图均衡程度的控制参数, 取 值为 0~1 , 较佳为 0.40 0.6, 值可以根据经验值来选择, 并可根据帧图像不 同的亮度值来设定不同的 取值; L*」表示向下取整() , 且。 ^^ ? 。 公
式中的向下取整也可以改为用向上取整。
动态分割点 c„将首、末端为 A的直方图区间分为两个区间 ,分割时 cn可 以归属于该两个区间其中的一个。 由于这里的区间包含的是离散的亮度值, 划分后的区间首末端点可能是相同即只包含一个亮度值, 对该区间再次划分 时, 得到的是相同的包含该亮度值的两个区间。
第五步, 重复第三步和第四步, 通过 8次迭代地调用, 可以将整个直方 图分成 256个区间;
划分成的 256个区间有些可能包括多个亮度值, 也可能存在首末端点相 同的区间。
第六步, 把划分出来的 256个区间, 按其包含的首个亮度值从小到大的 顺序从 0开始依次编号到 255 , 然后将每个区间包含的所有亮度值均对应到 等于该区间编号的映射值, 将区间 [0,255]上的每一亮度值替换为对应的映射 值后, 即得到直方图均衡映射表 Μ ΓαΜ^。
因为编号是 256个, 因此对相同区间中的每一个区间要分别编号且相同 区间的编号是连续的。 在映射时, 相同区间中的亮度值对应有多个编号, 可 以按约定映射到其中的一个编号, 具体是哪个编号不作限定。 如可以相同区 间中第一个区间或最后一个区间的编号, 或对第一个区间和最后一个区间编 号取均值后再取整得到的编号等等。 直方图均衡映射表 M^7¾ ¾也是一个一 维的在区间 [0,255]的映射表。
可以看出, 当 取值为 1 时, 第四步中的公式就是普通的对直方图均衡 化时求取动态分割点的方法。 相应地得到的直方图均衡映射表也与已知方法 得到的相同。 而本实施例加入了控制参数参数 得到的直方图均衡映射表 可以在适当扩展像素取值的动态范围的同时, 可以尽量避免灰度级被过多地 合并。
第七步, 对该帧图像的亮暗程度进行检测, 确定检测到的亮度值属于哪 一个亮暗程度等级;
可以将图像的亮暗程度分成 个等级, 将亮度值的整个取值范围划分为 与该 个等级——对应的 个区间。 亮度值最低的区间即图像最暗时,对应
的亮暗程度等级为 0 , 亮度值次低的区间, 对应的亮暗程度等级为 1 , ... ... , 以此类推, 图像亮度值正常或以上的区间, 对应的亮暗程度等级为 M-\。 可 以设置一个正常亮度门限(可以在 110~135之间取值) , 将大于等于该正常 亮度门限的亮度值都归入亮暗程度等级为 ΜΛ时对应的亮度值区间。
对该帧图像进行检测后得到该帧图像的亮度值, 该亮度值所属区间对应 的亮暗程度等级即为该帧图像的亮暗程度等级。
第八步, 根据确定的亮暗程度等级以及生成的非线性拉伸曲线映射表 MapTabGm和直方图均衡映射表 ΜαρΓαΜ^ , 4艮据下式生成一个综合的亮度映射 表 MapTcMAvg , 该映射表仍是一维的, 区间为 [0,255] :
MapTablAvg[k] = (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M 式中:
MapTablAvg[k]为亮度值 k在亮度映射表 Map7¾Mvg中对应的映射值,根据 上式计算出来的 MapTablAv^k]组成了 MapTablAvg。
ImgC/o^为检测出来的该帧图像的亮暗程度等级。
MapTabEq k]为亮度值 在直方图均衡映射表 ΜαρΓαΜ^中对应的映射值。
MapTabGm[k]为亮度值 在非线性拉伸曲线映射表 Map7¾G 中对应的映 射值。
上述亮度值 k可以是 [0,255]中的每一个亮度值, 也可以先检测一下图像 的像素点的亮度值集合, 只对该亮度值集合中的亮度值进行上述计算。 这样 可以减少运算量。
这里实际上是对直方图均衡映射和非线性拉伸曲线映射这两种方式进行 了一下加权平均, 其中的权值根据图像的亮暗程度来自适应地调整, 当然釆 用的权值并不局限于上式的算法, 也可以直接定义与亮暗程度等级对应的一 组权值来进行计算。 因为非线性拉伸曲线映射可以提高低照度图像的亮度。 但不能增强图像的细节, 因此本实施例综合釆用这两种方式并根据检测到的 图像的亮暗程度釆用不同的权值, 可以达到优化的亮度增强的效果。 即使釆 用的是用已知方法得到的直方图均衡映射表, 也是优于现有的图像增强处理 的效果的。
第九步, 根据生成的亮度映射表对该帧图像像素点的亮度值进行映射, 以实现亮度增强的效果;
第十步, 根据饱和度系数对该帧图像进行饱和度增强。
在亮度调整以后, 特别是低照度下, 通过亮度增强后, 整个图像会变得 发白, 图像的饱和度下降, 这就有必要进行饱和度的增强。 在 YUV色彩空间 中, Y 表示亮度分量, 用上述第一步到第九步的方法进行增强, ( U-128)/(V-128)表示色调, (^ -128)2 +( -128)2 表示饱和度, 调节饱和度的 原则就是要保持(U-128)/(V-128)不变, 增大图像的饱和度,
应当说明的是, 亮度增强和饱和度增强可以同时进行。
在饱和度增强时, 可以先定义一个饱和度系数数组 Su, Su[ ]为该数据 中的元素即每一亮暗程度等级对应的饱和度系数,如 Su= { 1.5, 1.48, 1.46, 1.44 1.42, 1.40, 1.38, 1.36, 1.34, 1.32, 1.30,1 } ; 其中 w为设定的图像的亮暗程度等 级, m = 0,1 , . . .M-1 , M为设定的图像的亮暗程度等级的数目。 数组中的系数 为不同亮暗程度等级下的饱和度系数, 该值可以根据试验进行选择和调节, 上面给出的一组系数仅为一种示例。 亮暗程度等级为 M-1时, 即正常光照或 以上的情况下, 不进行饱和度增强, 所以其系数为 1。
饱和度增强可以利用事先计算出来表格用查表的方法来求解, 能加快计 算速度; 也可以在增强的时候再进行计算, 计算的公式为:
Uo t(x,y) = Su[m] x (Um(x,y)— 128) + 128
Vout(x, y) = Su[m]x (Vm(x,y)— 128) + 128 Uin(x, y)和 Vin c, y)是进行饱和度增强之前像素点 U分量和 V分量的值, f/oMt(x, 和 P¾Mt(x, 是进行饱和度增强之后像素点 U分量和 V分量的值。
应当说明的是,本发明图像的上述亮度和色度值也可以使用 RGB空间的 亮度值和色度值, 至于 RGB 空间中图像和像素点的亮度值和色度值如何确 定, 在现有技术中多有论述, 这里不再进行说明。
工业实用性
本发明通过对不同光照条件图像进行检测、 并根据检测结果对低照度图 像和正常照度下的视频或图像进行增强, 在适当扩展像素取值的动态范围的 同时尽量避免灰度级被过多地合并, 解决传统直方图均衡化方法带来的图像 失真问题, 明显地改善了图像的视觉效果, 得到较高品质的输出图像。 使得 处理后的视频图像在低照度下比源图像更清晰, 在正常照度下比源图像细节 突出, 效果更好, 符合人眼的视觉。 且本发明的方法计算量小, 可以满足实 时性的需求, 不但可以设置成 ATM、 无线环境, 而且也可设置成 Internet视 频通信中视频解码图像的处理, 尤其适设置成手机电视的视频增强, 因此具 有^ 虽的工业实用性。
Claims
1、 一种视频增强的方法, 该方法包括:
设置非线性拉伸曲线映射表;
对待处理的每一帧图像, 根据其亮度数据计算得到该帧图像的直方图均 衡映射表, 然后将该帧图像的像素点的亮度值集合中的每一亮度值在所述非 线性拉伸曲线映射表和所述直方图均衡映射表中的映射值进行加权平均, 并 用亮度映射表记录各亮度值与加权平均后的映射值之间的对应关系; 然后根 据所述亮度映射表, 将该帧图像的像素点的亮度值更新为该亮度值对应的加 权平均后的映射值, 完成亮度增强。
2、 如权利要求 1所述的方法, 其中:
将该帧图像的像素点的亮度值集合中的每一亮度值在所述非线性拉伸曲 线映射表和所述直方图均衡映射表中的映射值进行加权平均的所述步骤包括 釆用下式进行加权平均:
MapTablAvg[k] = (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M 式中, k为该帧图像的像素点的亮度值集合中的一亮度值; 为设定的 图像亮暗程度等级的数目;
Μ 7¾Μ、^^]为亮度值 k在所述亮度映射表中对应的映射值, 根据上式 计算出来的 Μ 7¾Μ ^t]构成了所述亮度映射表;
ImgClass为对该帧图像进行亮度检测后得到的该帧图像的亮暗程度等 级;
MapTabEq k]为亮度值 k在所述直方图均衡映射表中对应的映射值; MapTabGm[k]为亮度值 k在所述非线性拉伸曲线映射表中对应的映射值。
3、 如权利要求 1或 2所述的方法, 其中, 根据其亮度数据计算得到该帧 图像的直方图均衡映射表的所述步骤包括:
步骤一, 对待处理的一帧图像, 先计算该帧图像的亮度直方图, 得到每 一个亮度值 的直方图数值即具有该亮度值的像素点数, A=0,l,2,..丄-1 , 其
中, L为将整个直方图分成的区间数;
步骤二, 计算分割得到的每一个直方图区间的中间点 然后根据下式 计算直方图区间的动态分割点 cn;
其中: α„ Α分别为一个直方图区间的首端点和末端点; "为预先设定的 直方图均衡程度的控制参数, 0 < « < 1; L*」表示向下取整, 「*表示向下取整,
0 < C„ < (J - 1);
步骤三, 重复执行步骤二, 通过 n次迭代地调用, 将整个直方图分成 L 个区间,并按各区间包含的首个亮度值按从小到大的顺序依次对各区间编号, 从 0开始依次对各区间编号直到 L -1 , 然后将每个区间包含的所有亮度值均 对应到等于该区间编号的映射值, 将区间 [0, L -1]上的每一亮度值替换为对应 的映射值后, 即得到所述直方图均衡映射表, 其中, 2n= L。
4、 如权利要求 3所述的方法, 其中, 将直方图分成的区间数 L为 256, 则所述迭代次数为 8次。
5、 如权利要求 4所述的方法, 其中, 所述直方图均衡程度的控制参数《 的取值范围是 0.4~0.6。
6、 如权利要求 3所述的方法, 其中:
所分成的区间存在相同区间时, 对相同区间中的每一个区间分别编号且 相同区间的编号连续; 相同区间中的亮度值对应多个编号, 在映射时按约定 将其中的亮度值映射到其中的一个编号。
7、 如权利要求 1所述的方法, 其中, 完成亮度增强后, 该方法还包括通 过以下处理对该帧图像进行饱和度增强的步骤:
设定图像的亮暗程度等级并为每一个亮暗程度等级对应设置一个饱和度 系数; 对该帧图像进行饱和度增强时, 先检测出该帧图像的亮暗程度等级, 然后使用对应的饱和度系数对该帧图像的像素点的色度值进行调整。
8、 如权利要求 2或 7所述的方法, 其中, 该帧图像的亮暗程度等级根据 以下方式确定:
保存该帧图像的亮暗程度等级的配置信息, 包括亮暗程度等级及每一等 级对应的亮度值区间;
对该帧图像进行检测后得到该帧图像的亮度值, 该亮度值所属区间对应 的亮暗程度等级即为该帧图像的亮暗程度等级 ImgC/o^。
9、 如权利要求 8所述的方法, 其中, 该帧图像的亮暗程度等级的配置信 息按照以下方式确定: 将该帧图像的亮暗程度分成 个等级, 将亮度值的整个取值范围划分为 与该 个等级——对应的 个区间,亮度值最低的区间对应的亮暗程度等级 为 0, 亮度值次低的区间对应的亮暗程度等级为 1, 亮度值越高的区间对应的 亮暗程度等级越高, 且将等于和大于设定的正常亮度值门限的所有亮度值归 入一个区间, 该区间对应的亮暗程度等级为 ΜΛ。
10、 如权利要求 9所述的方法, 当间接引用权利要求 7时, 其中: 该帧 图像是 YUV格式时, 对该帧图像进行饱和度增强的所述步骤按以下公式进 行:
Uo t(x,y) = S [m] x (Uin(x, )--^) + -^ Vout(x, y) = S [m]x (Vin(x, y)--^) + -^ 其中, Uin(x, y)和 Vin c, y)是进行饱和度增强之前像素点 U分量和 V分量 的值, f/oMt(x, 和 P¾Mt(x, 是进行饱和度增强之后像素点 U分量和 V分量的 值, 为数组 Su中与亮暗程度等级对应的饱和度系数, 数组 Su中记录有 每一个亮暗程度等级对应的饱和度系数, m = 0,1, ...M-1, 而 0,1, ...M-1为设 定的图像亮暗程度等级, 且亮暗程度等级为 M-1时, 其饱和度系数为 1, 其 中, L为将整个直方图分成的区间数。
11、 如权利要求 10所述的方法, 其中, 将整个直方图分成的区间数 L为 256。
12、 一种亮度增强模块, 该亮度增强模块包括非线性曲线拉伸表存储单 元、 直方图均衡映射表计算单元、 亮度映射表计算单元和图像亮度增强映射 单元, 其中:
所述非线性曲线拉伸表存储单元设置成保存要使用的非线性拉伸曲线映 射表;
所述直方图均衡映射表计算单元设置成对待处理的每一帧图像, 根据其 亮度数据计算得到相应的直方图均衡映射表;
所述亮度映射表计算单元设置成将该帧图像的像素点的亮度值集合中的 每一亮度值在非线性拉伸曲线映射表和直方图均衡映射表中的映射值进行加 权平均, 并用一亮度映射表记录各亮度值与加权平均后的映射值之间的对应 关系;
所述图像亮度增强映射单元设置成根据所述亮度映射表, 将待处理的一 帧图像的像素点的亮度值更新为该亮度值对应的加权平均后的映射值, 完成 亮度增强。
13、 如权利要求 12所述的视频增强装置, 其中:
所述亮度映射表计算单元设置成按下式将该帧图像的像素点的亮度值集 合中的每一亮度值在非线性拉伸曲线映射表和直方图均衡映射表中的映射值 进行力。权平均:
MapTablAvg[k] = (Im gClass xMapTabEq[k] + (M - Im gClass) x MapTabGm[k]) / M 式中, k为该该帧图像的像素点的亮度值集合中的一亮度值;
为设定的图像亮暗程度等级的数目;
Μ 7¾Μ、^^]为亮度值 k在所述亮度映射表中对应的映射值, 根据上式 计算出来的 Μ 7¾Μ ^t]构成了所述亮度映射表;
ImgClass为对该帧图像进行亮度检测后得到的该帧图像的亮暗程度等 级;
MapTabEq k]为亮度值 k在所述直方图均衡映射表中对应的映射值; MapTabGm[k]为亮度值 k在所述非线性拉伸曲线映射表中对应的映射值。
14、 如权利要求 12或 13所述的视频增强装置, 其中, 所述直方图均衡 映射表计算单元包括亮度直方图计算单元、 中间点计算单元、 动态分割点计 算单元、 控制单元和映射表生成单元, 其中:
所述亮度直方图计算单元设置成计算出待处理的一帧图像的亮度直方 图, 得到每一个亮度值 的直方图数值即具有该亮度值的像素点数, 输出到 中间点计算单元, =0,1,2,... L -1 , 其中, L为将整个直方图分成的区间数; 所述中间点计算单元设置成根据每一个亮度值 k的直方图数值, 计算分 割得到的每一个直方图区间的中间点 βη , 输出到动态分割点计算单元;
所述动态分割点计算单元设置成根据下式计算直方图区间的动态分割点
0 < C„ < (J -1);
所述控制单元设置成控制所述中间点计算单元和动态分割点计算单元进 行运算, 通过 n次迭代地调用, 将整个直方图分成 L个区间, 并将生成的 L 个区间及其包含的亮度值输出到映射表生成单元;
所述映射表生成单元设置成按各区间包含的首个亮度值从小到大的顺序 从 0开始依次对各区间编号直到 L-1 , 将每个区间包含的所有亮度值均对应 到等于该区间编号的映射值, 将区间 [0, L-1]上的每一亮度值替换为对应的映 射值后, 即得到所述直方图均衡映射表, 其中, 2n= L。
15、 如权利要求 14所述的方法, 其中, 所述控制单元设置成将直方图分 成 256个区间, 则所述迭代次数为 8次。
16、 一种视频增强装置, 其包括如权利要求 12-15 中任一项所述亮度增 强模块。
17、 如权利要求 16所述的视频增强装置, 其中: 所述视频增强装置还包
括一图像亮暗检测模块, 该图像亮暗检测模块包括配置信息存储单元和亮暗程度等级计算单元, 其中:
所述配置信息存储单元设置成保存该帧图像的亮暗程度等级的配置信 息, 包括亮暗程度等级及每一等级对应的亮度值区间;
所述亮暗程度等级计算单元设置成根据接收到的待处理的一帧图像的亮 度数据计算出该帧图像的亮度值, 根据所述配置信息判断出该亮度值所属区 间对应的亮暗程度等级, 将该亮暗程度等级作为该帧图像的亮暗程度等级输 出到所述亮度增强模块中的亮度映射表计算单元。
18、如权利要求 17所述的视频增强装置, 其中: 所述亮暗程度等级有 M 个等级, 亮度值的整个取值范围划分为与该 个等级——对应的 个区间 , 亮度值最低的区间对应的亮暗程度等级为 0, 亮度值次低的区间对应的亮暗 程度等级为 1 , 亮度值越高的区间对应的亮暗程度等级越高, 且将等于和大 于设定的正常亮度值门限的所有亮度值归入一个区间, 该区间对应的亮暗程 度等级为 Λ/-1。
19、 如权利要求 17-18 中任一项所述的视频增强装置, 其中: 所述视频 增强装置进一步包括一饱和度增强模块, 该饱和度增强模块包括配置信息存 储单元和饱和度增强计算单元, 其中:
所述图像亮暗检测模块中的亮暗程度等级计算单元还设置成将待处理的 一帧图像的亮暗程度等级输出到所述饱和度增强模块中的饱和度增强计算单 元;
所述配置信息存储单元设置成保存设定的图像的亮暗程度等级, 以及为 每一个亮暗程度等级对应设置的饱和度系数;
所述饱和度增强计算单元设置成对待处理的一帧图像进行饱和度增强 时, 根据接收到的该帧图像的亮暗程度等级, 使用对应的饱和度系数对该帧 图像的像素点的色度值进行调整。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP09804440.7A EP2323373B1 (en) | 2008-08-07 | 2009-07-28 | Video enhancing method and device thereof |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810146055.8 | 2008-08-07 | ||
CN2008101460558A CN101340510B (zh) | 2008-08-07 | 2008-08-07 | 一种视频增强的方法及其装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010015140A1 true WO2010015140A1 (zh) | 2010-02-11 |
Family
ID=40214454
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2009/000843 WO2010015140A1 (zh) | 2008-08-07 | 2009-07-28 | 一种视频增强的方法及其装置 |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP2323373B1 (zh) |
CN (1) | CN101340510B (zh) |
WO (1) | WO2010015140A1 (zh) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102223547A (zh) * | 2011-06-16 | 2011-10-19 | 王洪剑 | 图像色彩增强装置和方法 |
CN107132031A (zh) * | 2017-01-19 | 2017-09-05 | 中国科学院上海技术物理研究所 | 一种led最大辐亮度测量装置 |
CN107451979A (zh) * | 2017-08-08 | 2017-12-08 | 腾讯科技(深圳)有限公司 | 一种图像处理方法、装置和存储介质 |
CN109035182A (zh) * | 2018-07-17 | 2018-12-18 | 华侨大学 | 一种自适应动态双直方图均衡方法 |
CN109918971A (zh) * | 2017-12-12 | 2019-06-21 | 深圳光启合众科技有限公司 | 监控视频中人数检测方法及装置 |
CN111028181A (zh) * | 2019-12-25 | 2020-04-17 | 歌尔股份有限公司 | 一种图像增强处理方法、装置、设备及存储介质 |
CN113393397A (zh) * | 2021-06-21 | 2021-09-14 | 杭州微帧信息科技有限公司 | 一种图像对比度增强的方法和*** |
CN113766273A (zh) * | 2021-01-05 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | 一种处理视频数据的方法和装置 |
Families Citing this family (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101340510B (zh) * | 2008-08-07 | 2010-06-23 | 中兴通讯股份有限公司 | 一种视频增强的方法及其装置 |
CN101626454B (zh) * | 2009-04-10 | 2011-01-05 | 黄宝华 | 一种增强视频可视性的方法 |
CN101620819B (zh) * | 2009-06-25 | 2013-10-16 | 北京中星微电子有限公司 | 显示图像背光亮度的动态调整方法、装置及移动显示设备 |
CN101873503B (zh) * | 2009-12-09 | 2013-02-13 | 杭州海康威视数字技术股份有限公司 | 一种调节图像属性的方法和装置 |
CN101916431B (zh) * | 2010-07-23 | 2012-06-27 | 北京工业大学 | 一种低照度图像数据处理方法及*** |
CN102054266B (zh) * | 2010-11-26 | 2012-06-13 | 西安理工大学 | 一种夜间低照度图像噪声的抑制方法 |
BR112013028782A2 (pt) | 2011-05-11 | 2017-01-31 | I-Cubed Res Center Inc | aparelho de processamento de imagem, método de processamento de imagem, e meio de armazenamento onde um programa é armazenado |
CN102780889B (zh) * | 2011-05-13 | 2015-04-01 | 中兴通讯股份有限公司 | 视频图像处理方法、装置及设备 |
JP5829758B2 (ja) | 2011-10-20 | 2015-12-09 | ドルビー ラボラトリーズ ライセンシング コーポレイション | ビデオ等化のための方法およびシステム |
CN102567966A (zh) * | 2011-12-20 | 2012-07-11 | 四川大学 | 针对图像非线性增强的小误差查表方法 |
CN102611851A (zh) * | 2012-03-01 | 2012-07-25 | 林青 | 一种视频图像的自动照度补偿方法及*** |
GB2502591B (en) * | 2012-05-31 | 2014-04-30 | Sony Comp Entertainment Europe | Apparatus and method for augmenting a video image |
CN102750684B (zh) * | 2012-06-07 | 2015-11-25 | 王跃进 | 一种图像增强方法 |
CN105321153B (zh) * | 2014-07-09 | 2019-11-19 | 三亚中兴软件有限责任公司 | 视频监控低照度图像色彩还原方法和装置 |
CN105303586B (zh) * | 2014-07-16 | 2018-05-11 | 深圳Tcl新技术有限公司 | 获取直方图的方法、亮度动态调整的方法及图像处理装置 |
CN104202604B (zh) | 2014-08-14 | 2017-09-22 | 深圳市腾讯计算机***有限公司 | 视频增强的方法和装置 |
CN105744118B (zh) * | 2016-02-01 | 2018-11-30 | 杭州当虹科技有限公司 | 一种基于视频帧自适应的视频增强方法及视频增强*** |
US10129511B2 (en) * | 2016-08-01 | 2018-11-13 | Ricoh Company, Ltd. | Image processing apparatus, image projection apparatus, and image processing method |
JP6805968B2 (ja) * | 2016-08-01 | 2020-12-23 | 株式会社リコー | 画像処理装置、画像投影装置、及び画像処理方法 |
CN108460730B (zh) | 2017-02-17 | 2020-06-26 | 京东方科技集团股份有限公司 | 一种图像处理方法及其装置 |
CN108337495B (zh) * | 2018-02-08 | 2020-06-02 | 北京慧摩尔科技有限公司 | 一种视频图像动态对比度调整方法及其调整装置 |
CN110246086B (zh) * | 2018-03-07 | 2021-09-03 | 舜宇光学(浙江)研究院有限公司 | 彩色图像增强的图像处理方法及其图像处理*** |
CN108711140B (zh) * | 2018-05-16 | 2021-09-10 | 广东欧谱曼迪科技有限公司 | 一种基于类间方差描述的图像亮度均匀性实时恢复方法 |
CN108845779B (zh) * | 2018-06-19 | 2021-08-20 | 环鸿电子(昆山)有限公司 | 基于影像灰阶调整亮度的电子装置及其调整亮度的方法 |
CN110070508B (zh) * | 2019-04-23 | 2020-11-10 | 西安交通大学 | 一种基于门限值和线性映射的反锐化增强方法 |
CN112055131A (zh) * | 2019-06-05 | 2020-12-08 | 杭州吉沁文化创意有限公司 | 一种视频处理***及方法 |
CN110473492B (zh) * | 2019-08-28 | 2021-01-26 | 上海灵信视觉技术股份有限公司 | 一种led全彩显示屏的动态非线性显示调整方法、***和装置 |
CN110555799A (zh) * | 2019-09-26 | 2019-12-10 | 北京百度网讯科技有限公司 | 用于处理视频的方法和装置 |
CN110636331B (zh) * | 2019-09-26 | 2022-08-09 | 北京百度网讯科技有限公司 | 用于处理视频的方法和装置 |
CN111050211B (zh) * | 2019-12-13 | 2021-10-26 | 广州酷狗计算机科技有限公司 | 视频处理方法、装置及存储介质 |
CN112261438B (zh) * | 2020-10-16 | 2022-04-15 | 腾讯科技(深圳)有限公司 | 视频增强方法、装置、设备以及存储介质 |
CN114998135A (zh) * | 2022-05-31 | 2022-09-02 | 北京义礼科技有限公司 | 图像增强方法、装置、现场可编程逻辑门阵列以及设备 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08317250A (ja) * | 1995-05-19 | 1996-11-29 | Fujitsu General Ltd | ダイナミック映像制御回路 |
CN1750043A (zh) * | 2005-08-22 | 2006-03-22 | 上海广电(集团)有限公司中央研究院 | 一种控制平均亮度的直方图均衡方法 |
JP2007013666A (ja) * | 2005-06-30 | 2007-01-18 | Victor Co Of Japan Ltd | 画像処理装置 |
CN1925558A (zh) * | 2005-08-30 | 2007-03-07 | 株式会社东芝 | 视频信号处理设备及视频信号处理方法 |
US20070171183A1 (en) * | 2006-01-26 | 2007-07-26 | Asustek Computer Inc. | Image processing method for display device |
CN101179650A (zh) * | 2007-11-23 | 2008-05-14 | 西安交通大学 | 一种视频图像对比度增强方法及装置 |
CN101340510A (zh) * | 2008-08-07 | 2009-01-07 | 中兴通讯股份有限公司 | 一种视频增强的方法及其装置 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0963111A1 (en) * | 1998-06-02 | 1999-12-08 | Deutsche Thomson-Brandt Gmbh | Method and apparatus for dynamic contrast improvement in video pictures |
CN1266950C (zh) * | 2003-11-10 | 2006-07-26 | 华亚微电子(上海)有限公司 | 一种视频图像的品质增强***与方法 |
JP4277773B2 (ja) * | 2004-09-21 | 2009-06-10 | 株式会社日立製作所 | 映像表示装置 |
JP4648071B2 (ja) * | 2005-04-28 | 2011-03-09 | 株式会社日立製作所 | 映像表示装置及び映像信号の色飽和度制御方法 |
US7684639B2 (en) * | 2005-10-20 | 2010-03-23 | Sharp Laboratories Of America, Inc. | Methods and systems for automatic digital image enhancement |
-
2008
- 2008-08-07 CN CN2008101460558A patent/CN101340510B/zh not_active Expired - Fee Related
-
2009
- 2009-07-28 WO PCT/CN2009/000843 patent/WO2010015140A1/zh active Application Filing
- 2009-07-28 EP EP09804440.7A patent/EP2323373B1/en not_active Not-in-force
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08317250A (ja) * | 1995-05-19 | 1996-11-29 | Fujitsu General Ltd | ダイナミック映像制御回路 |
JP2007013666A (ja) * | 2005-06-30 | 2007-01-18 | Victor Co Of Japan Ltd | 画像処理装置 |
CN1750043A (zh) * | 2005-08-22 | 2006-03-22 | 上海广电(集团)有限公司中央研究院 | 一种控制平均亮度的直方图均衡方法 |
CN1925558A (zh) * | 2005-08-30 | 2007-03-07 | 株式会社东芝 | 视频信号处理设备及视频信号处理方法 |
US20070171183A1 (en) * | 2006-01-26 | 2007-07-26 | Asustek Computer Inc. | Image processing method for display device |
CN101179650A (zh) * | 2007-11-23 | 2008-05-14 | 西安交通大学 | 一种视频图像对比度增强方法及装置 |
CN101340510A (zh) * | 2008-08-07 | 2009-01-07 | 中兴通讯股份有限公司 | 一种视频增强的方法及其装置 |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102223547A (zh) * | 2011-06-16 | 2011-10-19 | 王洪剑 | 图像色彩增强装置和方法 |
CN107132031A (zh) * | 2017-01-19 | 2017-09-05 | 中国科学院上海技术物理研究所 | 一种led最大辐亮度测量装置 |
CN107132031B (zh) * | 2017-01-19 | 2023-05-05 | 中国科学院上海技术物理研究所 | 一种led最大辐亮度测量装置 |
CN107451979B (zh) * | 2017-08-08 | 2022-11-01 | 腾讯科技(深圳)有限公司 | 一种图像处理方法、装置和存储介质 |
CN107451979A (zh) * | 2017-08-08 | 2017-12-08 | 腾讯科技(深圳)有限公司 | 一种图像处理方法、装置和存储介质 |
CN109918971A (zh) * | 2017-12-12 | 2019-06-21 | 深圳光启合众科技有限公司 | 监控视频中人数检测方法及装置 |
CN109918971B (zh) * | 2017-12-12 | 2024-01-05 | 深圳光启合众科技有限公司 | 监控视频中人数检测方法及装置 |
CN109035182A (zh) * | 2018-07-17 | 2018-12-18 | 华侨大学 | 一种自适应动态双直方图均衡方法 |
CN109035182B (zh) * | 2018-07-17 | 2023-06-20 | 华侨大学 | 一种自适应动态双直方图均衡方法 |
CN111028181B (zh) * | 2019-12-25 | 2023-07-14 | 歌尔股份有限公司 | 一种图像增强处理方法、装置、设备及存储介质 |
CN111028181A (zh) * | 2019-12-25 | 2020-04-17 | 歌尔股份有限公司 | 一种图像增强处理方法、装置、设备及存储介质 |
CN113766273A (zh) * | 2021-01-05 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | 一种处理视频数据的方法和装置 |
CN113393397A (zh) * | 2021-06-21 | 2021-09-14 | 杭州微帧信息科技有限公司 | 一种图像对比度增强的方法和*** |
Also Published As
Publication number | Publication date |
---|---|
CN101340510B (zh) | 2010-06-23 |
EP2323373A1 (en) | 2011-05-18 |
EP2323373A4 (en) | 2013-10-02 |
EP2323373B1 (en) | 2018-09-05 |
CN101340510A (zh) | 2009-01-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2010015140A1 (zh) | 一种视频增强的方法及其装置 | |
JP4616749B2 (ja) | 適応コントラスト向上 | |
KR102234092B1 (ko) | 이미지의 역 톤 매핑을 위한 방법 | |
KR100782845B1 (ko) | 비로그 도메인 조도 수정을 이용한 디지털 영상 개선방법과 시스템 | |
KR102317613B1 (ko) | 로컬라이징된 콘트라스트 향상을 위한 시스템들 및 방법들 | |
US8965120B2 (en) | Image processing apparatus and method of controlling the same | |
CN105407296B (zh) | 实时视频增强方法和装置 | |
US20080252791A1 (en) | Image processing device and method, and program | |
CN111885312B (zh) | Hdr图像的成像方法、***、电子设备及存储介质 | |
CN103916669A (zh) | 高动态范围图像压缩方法及装置 | |
CN110706172B (zh) | 基于自适应混沌粒子群优化的低照度彩色图像增强方法 | |
CN103353982A (zh) | 一种基于直方图均衡化的色调映射方法 | |
CN108537758B (zh) | 一种基于显示器与人眼视觉特性的图像对比度增强方法 | |
US20130039577A1 (en) | Method for improving image quality | |
WO2023098251A1 (zh) | 图像处理方法、设备及可读存储介质 | |
CN107862672B (zh) | 图像去雾的方法及装置 | |
CN111127340B (zh) | 一种图像去雾方法 | |
TW202046698A (zh) | 影像調整方法以及相關的影像處理電路 | |
US20130287299A1 (en) | Image processing apparatus | |
CN107358592B (zh) | 一种迭代式全局自适应图像增强方法 | |
KR100765760B1 (ko) | 이미지의 전역 정보를 이용하여 어두운 부분의 세부들을개선하는 방법 및 장치 | |
KR102370400B1 (ko) | 영상 처리 장치 | |
TW200913730A (en) | Method and apparatus for processing image and electronic device using the same | |
CN105631812B (zh) | 一种对显示图像进行色彩增强的控制方法及控制装置 | |
JP4719200B2 (ja) | 画像補正装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09804440 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2009804440 Country of ref document: EP |