WO2018121221A1 - 一种解决通道间亮度不一致的方法 - Google Patents
一种解决通道间亮度不一致的方法 Download PDFInfo
- Publication number
- WO2018121221A1 WO2018121221A1 PCT/CN2017/115207 CN2017115207W WO2018121221A1 WO 2018121221 A1 WO2018121221 A1 WO 2018121221A1 CN 2017115207 W CN2017115207 W CN 2017115207W WO 2018121221 A1 WO2018121221 A1 WO 2018121221A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- brightness
- model
- seam
- pixels
- optimal solution
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000005070 sampling Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- 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
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Definitions
- the present invention relates to the field of image sensor technologies, and in particular, to a method for solving brightness inconsistency between channels.
- the conventional method is to add a value to a channel. Make the brightness between the two channels consistent, or adjust the gain between the two channels, so that one channel is multiplied by a value to get the same brightness, or feathered at the stitching gap, so that it looks smooth and excessive, the brightness change does not look obvious .
- the first two methods are mainly for the case where the brightness of the two channels is consistent, and the effect of the brightness variation between the channels is inconsistent, that is, the effect may be different when the upper half and the lower half are different, and the last feathering effect is not obvious. Affecting the sharpness of the image, it is not feasible to face high definition.
- the object of the present invention is to remedy the above deficiencies of the prior art, and the present invention aims to provide a method for solving the inconsistency of brightness between channels, which can automatically identify the optimal solution of the value change between channels and improve the map. Like an effect.
- the present invention provides a method for solving brightness inconsistency between channels, which comprises the following steps:
- Step 01 providing two images, at a seam of the two images, setting a width range; the seam is surrounded by the set width;
- Step 02 Calculate the brightness of each line of pixels in the set width range
- Step 03 taking the coordinates of the pixel values of each row in the set width range as the abscissa and the ordinate, thereby obtaining a point set;
- Step 04 According to the method of obtaining the optimal solution in the Ransac algorithm, the point set is fitted to a straight line to obtain an optimal solution model of the point set;
- Step 05 Find a fitting value of each pixel in one side of the seam in the optimal solution model, and adjust the brightness of the pixel at the corresponding position on the side of the seam according to the fitting value.
- the brightness of each row of pixels is the sum of each pixel value in the corresponding row divided by the set width.
- the brightness of each row of pixels to the left of the seam is equal to the brightness of the pixels of the corresponding row to the right of the seam.
- the step 04 specifically includes:
- Step 041 Set a model with a minimum sampling set of n and a sample set P.
- the sample set P is a set of points with the pixel values in each row as the horizontal and vertical coordinates, and the number of samples in the sample set P is greater than n;
- n is The minimum number of samples required to initialize the model M i parameters, n ⁇ 1;
- Step 042 Randomly extract a subset S of the sample set P containing n samples from the sample set P, and simulate to obtain an initialization model M i .
- the corresponding initialization model (M i ) is the optimal solution model.
- the initialization model (M i ) is obtained by a least squares method.
- the step 05 specifically includes: matching all pixels on the left side or the right side of the patch to corresponding pixels in the optimal solution model, thereby completing brightness adjustment.
- the method for solving the inconsistency of brightness between channels can automatically compensate for the difference in brightness between channels and optimize the image effect.
- FIG. 1 is a schematic diagram of an image of a channel inconsistency according to a preferred embodiment of the present invention
- FIG. 2 is a schematic diagram of an optimal solution model according to a preferred embodiment of the present invention.
- FIG. 3 is a schematic flow chart of a method for solving brightness inconsistency between channels according to a preferred embodiment of the present invention.
- FIGS. 1 to 3 specific embodiments. It should be noted that the drawings are in a very simplified form, using a non-precise ratio, and are only used to facilitate the purpose of the present embodiment.
- a method for solving the brightness inconsistency between channels includes the following steps:
- Step 01 providing two images, at a seam of the two images, setting a width range; the seam is surrounded by the set width;
- FIG. 1 there is an image in which the channels are inconsistent, and a certain width is selected at the joint of the two images. As shown in FIG. 1, there is a seam between the gray image and the white image.
- the width set here is i_weight.
- Step 02 Calculate the brightness of each line of pixels in the set width range
- the brightness of each row of pixels is the sum of each pixel value in the corresponding row divided by the set width.
- Step 03 taking the coordinates of the pixel values of each row in the set width range as the abscissa and the ordinate, thereby obtaining a point set;
- a pixel I (i, j) has an abscissa of i, sitting vertically Marked as j, thus forming a point coordinate, and then obtaining a set of points within the set width range.
- Step 04 According to the method of obtaining the optimal solution in the Ransac algorithm, the point set is fitted to a straight line to obtain an optimal solution model of the point set;
- the brightness of each row of pixels on the left side of the patch is equal to the brightness of the pixel of the corresponding row on the right side of the patch.
- This step 04 specifically includes:
- Step 041 Set a model with a minimum sampling set of n and a sample set P.
- the sample set P is a set of points with the pixel values in each row as the horizontal and vertical coordinates, and the number of samples in the sample set P is greater than n; here, n is the minimum number of samples required to initialize the parameters of the model M, such as randomly taking 100 numbers, and then arbitrarily taking 2 different numbers among the 100 numbers; the sample set P is the horizontal and vertical values of the pixel values around the patchwork.
- the set of points of the coordinates, the number of samples of the sample set P is greater than n, n ⁇ 1.
- Step 042 Randomly extract a subset S of the sample set P containing n samples from the sample set P, and simulate to obtain an initialization model M i ; here, a subset of the sample set P containing n samples is randomly extracted from the sample set P
- the S simulation results in an initialization model M i , which is generally a least squares method.
- Step 043 The sample set of the remaining set of the sample set P and the initial model M i is smaller than the set threshold t, and the subset S constitutes a new set S*, and the set S* is an inner point set;
- the sample set with the shortest distance from the initialization model M being less than a certain set threshold t and the subset S constitute a new set S*, the new set S* is considered to be the inner point set, and the number of samples in the set (S*) N i , where i is a positive integer starting from 1.
- the corresponding initialization model (M i ) is the optimal solution model. Specifically, by setting the number of extractions in advance, the inner point set (S*) of the maximum number of samples is obtained, and the corresponding initial model (M i ) is required to obtain an optimal solution model, and the number of cycles is determined according to the required precision, preferably.
- the number of cycles is 1000 to 2000 times.
- Step 05 Find a fitting value of each pixel in one side of the seam in the optimal solution model, and adjust the brightness of the pixel at the corresponding position on the side of the seam according to the fitting value.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Television Receiver Circuits (AREA)
Abstract
Description
Claims (6)
- 一种解决通道间亮度不一致的方法,其特征在于,包括以下步骤:步骤01:提供两个图像,在两个图像的拼缝处,设定一宽度范围;拼缝被所设定的宽度包围;步骤02:计算出所设定的宽度范围内每行像素的亮度;步骤03:以所设定的宽度范围内的每行像素值的坐标作为横坐标和纵坐标,从而得到点集合;步骤04:根据Ransac算法中求取最优解的方法,将点集合拟合出一条直线,得到点集合的最优解模型;步骤05:在最优解模型中找出拼缝一侧中每个像素的拟合值,按照该拟合值来调整拼缝一侧相应位置的像素的亮度。
- 根据权利要求1所述的方法,其特征在于,所述步骤02中,每行像素的亮度为相应行中每个像素值的总和除以所设定的宽度。
- 根据权利要求1所述的方法,其特征在于,所述最优解模型中,拼缝左边的每行像素的亮度等于拼缝右边的相应行的像素的亮度。
- 根据权利要求1所述的方法,其特征在于,所述步骤04具体包括:步骤041:设定一个最小抽样集为n的模型和一个样本集(P),样本集(P)为以每行中的像素值作为横纵坐标的点集合,样本集(P)中样本个数大于n;n为初始化模型(Mi)参数所需的最小样本数,n≥1;步骤042:从样本集(P)中随机抽取包含n个样本的样本集P的子集(S),模拟得到初始化模型(Mi);步骤043:将样本集(P)的余集中与初始化模型(Mi)的最短距离小于设定阈值t的样本集,和子集(S)构成新的集合(S*),集合(S*)中样本的个数=Ni;其中,i为从1开始的正整数;步骤044:当i=1时,将N1赋值给Nmax;当i≥1时,比较Ni与Nmax,若Ni≥Nmax,则认为Ni所对应的集合(S*)所对应的初始化模型(Mi)正确,令Nmax=Ni;若Ni≤Nmax,则保持Nmax不变;步骤045:按照预设的循环次数重复步骤042至步骤044,每循环一次,令i=i+1,从而得到样本个数最大的集合(S*),并且样本个数最大的集合(S*)所对应的初始化模型(Mi)为最优解模型。
- 根据权利要求4所述的方法,其特征在于,所述步骤042中,采用最小二乘法得到初始化模型(Mi)。
- 根据权利要求1所述的方法,其特征在于,所述步骤05具体包括:将拼缝左侧或右侧的所有像素与最优解模型中的相应像素对应起来,从而完成亮度调整。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611224484.3 | 2016-12-27 | ||
CN201611224484.3A CN106780339B (zh) | 2016-12-27 | 2016-12-27 | 一种解决通道间亮度不一致的方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018121221A1 true WO2018121221A1 (zh) | 2018-07-05 |
Family
ID=58927337
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2017/115207 WO2018121221A1 (zh) | 2016-12-27 | 2017-12-08 | 一种解决通道间亮度不一致的方法 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106780339B (zh) |
WO (1) | WO2018121221A1 (zh) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106780339B (zh) * | 2016-12-27 | 2019-02-05 | 上海集成电路研发中心有限公司 | 一种解决通道间亮度不一致的方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6385349B1 (en) * | 1997-09-03 | 2002-05-07 | Mgi Software Corporation | Method and system for compositing images |
CN104318517A (zh) * | 2014-11-19 | 2015-01-28 | 北京奇虎科技有限公司 | 一种图像拼接处理方法、装置及客户端 |
CN104392416A (zh) * | 2014-11-21 | 2015-03-04 | 中国电子科技集团公司第二十八研究所 | 一种运动场景的视频拼接方法 |
CN104574421A (zh) * | 2015-01-29 | 2015-04-29 | 北方工业大学 | 一种大幅面小重合区域高精度多光谱图像配准方法及装置 |
CN106780339A (zh) * | 2016-12-27 | 2017-05-31 | 上海集成电路研发中心有限公司 | 一种解决通道间亮度不一致的方法 |
-
2016
- 2016-12-27 CN CN201611224484.3A patent/CN106780339B/zh active Active
-
2017
- 2017-12-08 WO PCT/CN2017/115207 patent/WO2018121221A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6385349B1 (en) * | 1997-09-03 | 2002-05-07 | Mgi Software Corporation | Method and system for compositing images |
CN104318517A (zh) * | 2014-11-19 | 2015-01-28 | 北京奇虎科技有限公司 | 一种图像拼接处理方法、装置及客户端 |
CN104392416A (zh) * | 2014-11-21 | 2015-03-04 | 中国电子科技集团公司第二十八研究所 | 一种运动场景的视频拼接方法 |
CN104574421A (zh) * | 2015-01-29 | 2015-04-29 | 北方工业大学 | 一种大幅面小重合区域高精度多光谱图像配准方法及装置 |
CN106780339A (zh) * | 2016-12-27 | 2017-05-31 | 上海集成电路研发中心有限公司 | 一种解决通道间亮度不一致的方法 |
Non-Patent Citations (1)
Title |
---|
SONG, WEIYAN ET AL.: "Electronic Technology & Information Science , China Master'sTheses Full-Text Database", RANSAC ALGORITHM AND ITS APPLICATION IN REMOTE SENSING IMAGES PROCESSING, 15 September 2011 (2011-09-15) * |
Also Published As
Publication number | Publication date |
---|---|
CN106780339B (zh) | 2019-02-05 |
CN106780339A (zh) | 2017-05-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019090794A1 (zh) | Gamma校正系统及其校正方法 | |
US11323676B2 (en) | Image white balance processing system and method | |
CN104732578B (zh) | 一种基于倾斜摄影技术的建筑物纹理优化方法 | |
CN107255521B (zh) | 一种红外图像非均匀性校正方法及系统 | |
WO2017035910A1 (zh) | Mura区域的亮度补偿方法及Mura像素点亮度的设计方法 | |
CN102892021B (zh) | 一种合成虚拟视点图像的新方法 | |
US20140267378A1 (en) | Global contrast correction | |
WO2018036096A1 (zh) | 一种四色显示器白平衡过程中颜色漂移的调整方法 | |
WO2016095318A1 (zh) | 显示面板缺陷的自动检测方法 | |
WO2017049703A1 (zh) | 图像对比度增强方法 | |
WO2016106852A1 (zh) | 一种显示面板的缺陷的灰阶补偿方法及灰阶补偿系统 | |
JP2018524896A5 (zh) | ||
CN104680492B (zh) | 基于样本结构一致性的图像修复方法 | |
CN106408596B (zh) | 基于边缘的局部立体匹配方法 | |
JP2020518879A5 (zh) | ||
CN108022223A (zh) | 一种基于对数映射函数分块处理融合的色调映射方法 | |
CN104021527B (zh) | 一种图像中雨雪的去除方法 | |
CN107705290A (zh) | 一种AMOLED显示屏Mura缺陷检测方法 | |
CN104702941A (zh) | 一种白点区域表示及判定方法 | |
JP2011258179A5 (zh) | ||
WO2018121221A1 (zh) | 一种解决通道间亮度不一致的方法 | |
US9571744B2 (en) | Video processing method and apparatus | |
CN108986765B (zh) | 一种涵盖视角色偏的显示屏白点校准方法 | |
WO2020093619A1 (zh) | 侦测显示装置的图像中高频成分的方法及装置 | |
Dai et al. | Improved digital image restoration algorithm based on criminisi |
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: 17886042 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: 17886042 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 07.11.2019) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 17886042 Country of ref document: EP Kind code of ref document: A1 |