WO2015176631A1 - 一种基于直方图均衡化的图像数据去雾方法 - Google Patents
一种基于直方图均衡化的图像数据去雾方法 Download PDFInfo
- Publication number
- WO2015176631A1 WO2015176631A1 PCT/CN2015/079189 CN2015079189W WO2015176631A1 WO 2015176631 A1 WO2015176631 A1 WO 2015176631A1 CN 2015079189 W CN2015079189 W CN 2015079189W WO 2015176631 A1 WO2015176631 A1 WO 2015176631A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- histogram
- image data
- channels
- value
- minpixels
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000001186 cumulative effect Effects 0.000 claims description 12
- 238000013507 mapping Methods 0.000 claims description 6
- 238000007493 shaping process Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
Images
Classifications
-
- 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
Definitions
- the present invention relates to an image defogging technique, and more particularly to an image data dehazing method based on histogram equalization.
- the image defogging processing method of the prior art is based on the image defogging technology of the dark primary color prior, but the method has complicated steps and low efficiency, and is not suitable for application on the smart mobile terminal device and the low-end configuration device.
- the object of the present invention is to overcome the deficiencies of the prior art, and to provide an image data dehazing method based on histogram equalization with a simple method and a simple effect.
- the image to be processed is a non-RGB image
- the image to be processed is first converted into an RGB image, and then RGB data of the image to be processed is acquired, and three channels of R, G, and B are separated. .
- step 2) is specifically:
- the minimum pixel value minPixels and the maximum pixel value maxPixels of the corresponding channel will be calculated.
- step 2.1) is specifically:
- the histogram statistics table is an integer array of 256 lengths in memory.
- step 2.3) calculates the minimum pixel value minPixels that the R, G, and B channels will be equalized, and the maximum pixel value maxPixels is specifically:
- step 2.3.1) is specifically:
- step 3 is specifically:
- step 3.2) is specifically:
- step 3.2.1 is as follows:
- Float scale (float)(maxTarget-minTarget)/(float)(maxPixels-minPixels);
- step 3.2.2 The implementation of step 3.2.2 is as follows:
- Int targetI (unsigned char)((i-minPixels)*scale+minTarget);
- targetI is the mapping result of any value i located in the range of minPixels and maxPixels.
- the image data defogging method according to the present invention is based on histogram equalization, and the overall steps of the method are simple and easy to implement.
- the method has a small program amount and high execution efficiency, and can be smoothly used even on a smart mobile terminal device with a low calculation rate and a device with a low-end configuration.
- Figure 1 is an image to be processed
- An image data dehazing method based on histogram equalization is used for defogging processing of an image to be processed (width 704, height 528) as shown in FIG. 1 , and the steps are as follows:
- the present invention first retrieves whether the image to be processed is an RGB image before performing the defogging process. For example, if the image to be processed is a non-RGB image, in step 1), the image to be processed is first converted into an RGB image, and RGB data of the image to be processed is acquired, and three channels of R, G, and B are separated. The image imported in this embodiment is RGB data, and no conversion is required.
- step 2) is specifically:
- minFlatten is 1858
- maxFlatten is 1858.
- the sum of minFlatten plus maxFlatten of the three channels R, G, and B cannot be greater than the size of the respective channel, otherwise it is regarded as an invalid value.
- Step 2.1) is specifically:
- the histogram statistics table is an integer array of 256 lengths in memory.
- a histogram information is stored with a 256-sized shaped array.
- the minimum pixel value minPixels and the maximum pixel value maxPixels of the corresponding channel will be calculated.
- the red channel has a minPixels of 94 and a maxPixels of 244;
- the green channel has a minPixels of 93 and a maxPixels of 240;
- the blue channel has a minPixels of 82 and a maxPixels of 236.
- Step 2.3) Calculate the minimum pixel value minPixels that the R, G, and B channels will be equalized, and the maximum pixel value maxPixels specifically:
- Step 2.3.1) is specifically:
- Step 3) is specifically:
- step 2.1.1) and step 2.1.2 Based on the results of step 2.1.1) and step 2.1.2), the values of the pixels of minPixels and maxPixels among the three channels R, G, and B are uniformly mapped to minTarget and maxTarget, respectively.
- Step 3.2) is specifically:
- step 3.2.1 Implemented in code, the implementation of step 3.2.1 is as follows:
- Float scale (float)(maxTarget-minTarget)/(float)(maxPixels-minPixels);
- step 3.2.2 The implementation of step 3.2.2 is as follows:
- Int targetI (unsigned char)((i-minPixels)*scale+minTarget);
- targetI is the mapping result of any value i located in the range of minPixels and maxPixels.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Facsimile Image Signal Circuits (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
Description
Claims (10)
- 一种基于直方图均衡化的图像数据去雾方法,其特征在于,步骤如下:1)获取待处理图像的RGB数据,并分离R、G、B三个通道;2)统计R、G、B三个通道的直方图信息;3)分别对R、G、B三个通道进行直方图均衡化;4)将均衡化的R、G、B三个通道的直方图信息加载到RGB图像数据中,将RGB图像数据转换为原始格式图像数据,输出去雾图像。
- 根据权利要求1所述的基于直方图均衡化的图像数据去雾方法,其特征在于,如果待处理图像为非RGB图像,则在步骤1)中,则先将待处理图像转成RGB图像,再进行获取待处理图像的RGB数据,并分离R、G、B三个通道。
- 根据权利要求1所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤2)具体为:2.1)分别预设R、G、B三个通道中最暗部分将被变平化的数量minFlatten,以及最亮部分将被变平化的数量maxFlatten;2.2)统计各自通道的直方图信息,建立直方图统计表。2.3)分别根据R、G、B三个通道的minFlatten和maxFlatten以及直方图统计表,计算得出相应通道将被均衡化的最小像素值minPixels,以及最大像素值maxPixels。
- 根据权利要求3所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤2.1)具体为:2.1.1)分别判断R、G、B三个通道中的值是否小于对应minPixels的像素,如果存在,则将该值赋值为对应minPixels的像素;2.1.2)分别判断R、G、B三个通道中的值是否大于对应maxPixels的像素,如果存在,则将该值赋值为对应的maxPixels。
- 根据权利要求3所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤2.2)中,直方图统计表为内存中的一个具有256个长度的整形数组。
- 根据权利要求3所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤2.3)计算R、G、B三个通道将被均衡化的最小像素值minPixels,以及最大像素值maxPixels 具体为:2.3.1)将直方图统计表,转换为累计直方图表;2.3.2)从直方图数组下标为0的位置开始向前遍历累计直方图表,搜索第一个大于minFlatten的索引值作为minPixles值;2.3.3)从直方图数组下标为255的位置开始向后遍历累计直方图表,搜索第一个小于(size–maxFlatten)的索引值作为maxPixels值,其中size为通道大小。
- 根据权利要求6所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤2.3.1)具体为:2.3.1.1)从直方图整形数组中下标为0的位置开始将该位置的值加到下标为1的位置;2.3.1.2)将下标为1的位置的值加到下标为2的位置;2.3.1.3)循环步骤2.3.1.1)、步骤2.3.1.2),直到下标值为255,完成直方图统计表转换为累计直方图表。
- 根据权利要求4所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤3)具体为:3.1)预设各自通道均衡化的最小目标值minTarget和最大目标值maxTarget;3.2)分别将R、G、B三个通道中,介于minPixels和maxPixels的像素的值均匀映射为minTarget和maxTarget。
- 根据权利要求8所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤3.2)具体为:3.2.1)遍历minPixels和maxPixels范围内的值;3.2.2)在minPixels和maxPixels范围内,逐个映射。
- 根据权利要求9所述的基于直方图均衡化的图像数据去雾方法,其特征在于,步骤3.2.1)的实现具体如下:Float scale=(float)(maxTarget-minTarget)/(float)(maxPixels-minPixels);步骤3.2.2)的实现具体如下:for(i=minPixels;i<maxPixels;i++)Int targetI=(unsigned char)((i-minPixels)*scale+minTarget);其中,targetI即是位于minPixels和maxPixels范围内的任意值i的映射结果。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410211137.1 | 2014-05-19 | ||
CN201410211137.1A CN105096282A (zh) | 2014-05-19 | 2014-05-19 | 一种基于直方图均衡化的图像数据去雾方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015176631A1 true WO2015176631A1 (zh) | 2015-11-26 |
Family
ID=54553416
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2015/079189 WO2015176631A1 (zh) | 2014-05-19 | 2015-05-18 | 一种基于直方图均衡化的图像数据去雾方法 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN105096282A (zh) |
WO (1) | WO2015176631A1 (zh) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109685735A (zh) * | 2018-12-21 | 2019-04-26 | 温州大学 | 基于雾层平滑先验的单张图片去雾方法 |
CN110322521A (zh) * | 2019-07-10 | 2019-10-11 | 南充职业技术学院 | 隐形图像的色度信息挖掘方法 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106328037B (zh) * | 2016-10-14 | 2019-10-11 | 深圳市华星光电技术有限公司 | 四色显示器的驱动方法 |
CN107689212A (zh) * | 2017-09-14 | 2018-02-13 | 遵义师范学院 | 一种低功耗高画质的电流限制算法 |
CN111738927A (zh) * | 2020-03-23 | 2020-10-02 | 阳光暖果(北京)科技发展有限公司 | 基于直方图均衡化的人脸识别特征增强与去噪方法及系统 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102202163A (zh) * | 2011-05-13 | 2011-09-28 | 成都西图科技有限公司 | 一种监控视频的自适应增强方法及其装置 |
CN102306384A (zh) * | 2011-08-26 | 2012-01-04 | 华南理工大学 | 一种基于单幅图像的颜色恒常性处理方法 |
KR101361644B1 (ko) * | 2012-08-01 | 2014-02-13 | 엠텍비젼 주식회사 | 영상 처리 장치 및 영상 내의 안개 제거 방법 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060274937A1 (en) * | 2005-06-07 | 2006-12-07 | Eric Jeffrey | Apparatus and method for adjusting colors of a digital image |
CN102930517A (zh) * | 2012-11-30 | 2013-02-13 | 江苏技术师范学院 | 直方图均衡化图像增强方法 |
-
2014
- 2014-05-19 CN CN201410211137.1A patent/CN105096282A/zh active Pending
-
2015
- 2015-05-18 WO PCT/CN2015/079189 patent/WO2015176631A1/zh active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102202163A (zh) * | 2011-05-13 | 2011-09-28 | 成都西图科技有限公司 | 一种监控视频的自适应增强方法及其装置 |
CN102306384A (zh) * | 2011-08-26 | 2012-01-04 | 华南理工大学 | 一种基于单幅图像的颜色恒常性处理方法 |
KR101361644B1 (ko) * | 2012-08-01 | 2014-02-13 | 엠텍비젼 주식회사 | 영상 처리 장치 및 영상 내의 안개 제거 방법 |
Non-Patent Citations (2)
Title |
---|
YIN, LI: "Research of Image Dehazing and Super-resolution Reconstruction", ELECTRONIC TECHNOLOGY & INFORMATION SCIENCE , CHINA MASTER'S THESES FULL-TEXT DATABASE, 15 October 2012 (2012-10-15), pages 1 7 * |
ZHONG, YILONG: "The Research of Single Image-Defog Processing Algorithm and Software Implement", ELECTRONIC TECHNOLOGY & INFORMATION SCIENCE , CHINA MASTER'S THESES FULL-TEXT DATABASE, 15 March 2013 (2013-03-15) * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109685735A (zh) * | 2018-12-21 | 2019-04-26 | 温州大学 | 基于雾层平滑先验的单张图片去雾方法 |
CN110322521A (zh) * | 2019-07-10 | 2019-10-11 | 南充职业技术学院 | 隐形图像的色度信息挖掘方法 |
CN110322521B (zh) * | 2019-07-10 | 2022-12-23 | 南充职业技术学院 | 隐形图像的色度信息挖掘方法 |
Also Published As
Publication number | Publication date |
---|---|
CN105096282A (zh) | 2015-11-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2015176631A1 (zh) | 一种基于直方图均衡化的图像数据去雾方法 | |
US9420247B2 (en) | White balance correcting apparatus and white balance correcting method for white balance correction in complex light condition | |
US11289049B2 (en) | Gamut mapping with blended scaling and clamping | |
CN105578165A (zh) | 一种图像白平衡处理方法、装置及摄像机 | |
CN103778900A (zh) | 一种图像处理方法及系统 | |
GB2556761A (en) | Image contrast enhancement method | |
CN103491357A (zh) | 一种图像传感器白平衡处理方法 | |
CN108337496A (zh) | 白平衡处理方法、处理装置、处理设备及存储介质 | |
CN105681775A (zh) | 一种白平衡方法和装置 | |
TWI591583B (zh) | 影像對比增強方法及其裝置 | |
CN111161163B (zh) | 一种直方图双重伽马校正的红外图像对比度增强方法 | |
CN104766276B (zh) | 一种基于颜色空间的偏色校正方法 | |
JP2013008346A5 (zh) | ||
CN103295206A (zh) | 一种基于Retinex的微光图像增强方法和装置 | |
CN104392425A (zh) | 一种基于人脸的自动调整对比度的图像增强方法 | |
CN104361566A (zh) | 一种优化暗部区域的图像处理方法 | |
CN108122217B (zh) | 一种自适应场景的实时红外图像增强方法 | |
TWI545964B (zh) | 影像白平衡方法及攝像裝置 | |
CN102752608A (zh) | 一种实现快速色彩校正的方法和装置 | |
CN103974052B (zh) | 一种图像饱和度的调整方法、装置和设备 | |
TWI523500B (zh) | 影像的動態範圍壓縮方法與影像處理裝置 | |
WO2010149575A3 (en) | Digital image processing method and device for lightening said image | |
WO2020168818A1 (zh) | 一种图像处理方法、系统、装置、存储介质及黑板装置 | |
CN110113590A (zh) | 一种图像色彩空间的转换方法及装置 | |
TWI424751B (zh) | 飽和度調整裝置及其方法 |
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: 15796653 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15796653 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15796653 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15796653 Country of ref document: EP Kind code of ref document: A1 |