CN112381820A - Evaluation method based on sharpness of group of photos in same scene - Google Patents
Evaluation method based on sharpness of group of photos in same scene Download PDFInfo
- Publication number
- CN112381820A CN112381820A CN202011427653.XA CN202011427653A CN112381820A CN 112381820 A CN112381820 A CN 112381820A CN 202011427653 A CN202011427653 A CN 202011427653A CN 112381820 A CN112381820 A CN 112381820A
- Authority
- CN
- China
- Prior art keywords
- photos
- same scene
- group
- sharpness
- distribution
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 claims abstract description 17
- 241001584785 Anavitrinella pampinaria Species 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 abstract description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 206010044565 Tremor Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an evaluation method based on the definition of a group of photos in the same scene, which comprises the following steps: s10, acquiring a group of multiple pictures with the same scene; s20, selecting an ROI (region of interest) of the photo with the same scene; s30, calculating gradient information of all color channels of the image in the ROI area to obtain a gradient map; s40, counting the histogram distribution of the gradient map according to a common gray map mode; s50, intercepting a section of distribution of the number of the first T% pixels with the highest intensity in each histogram distribution, wherein T is an adjustable parameter; s60, carrying out respective weighted average on the intercepted distribution to obtain a scalar value; and S70, judging the definition of the photo in the same scene according to the scalar value. The method can quickly and stably give the definition evaluation of a group of photos of the same scene with good noise resistance and basically no deviation from subjective evaluation, and therefore contrast and optimization of the group of photos are provided.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an evaluation method based on the definition of a group of photos in the same scene.
Background
At present, a plurality of photos with better definition are often continuously shot when people take photos, the existing image definition evaluation method is obviously interfered by noise, the evaluation result is easy to be inconsistent with the subjective evaluation result, the method is insensitive to local motion blur in a photo window, and the calculation mode of some methods is complicated and has larger calculation amount.
In view of the above, there is a need for a method for evaluating the sharpness of a group of photos of the same scene, which can solve the above-mentioned technical problems, is fast and stable, has good noise resistance, and is substantially unbiased from subjective evaluation, and thereby provides a comparison and selection of the group of photos to overcome the above-mentioned problems.
Disclosure of Invention
In view of the defects of the prior art, the invention aims to provide an evaluation method based on the definition of a group of photos in the same scene, and aims to solve the problems that the existing image definition evaluation method in the prior art is obviously interfered by noise, the evaluation result is easy to be inconsistent with the subjective evaluation result, the evaluation method is insensitive to local motion blur in a photo window, and the calculation amount is large.
In order to achieve the purpose, the invention adopts the following technical scheme:
an evaluation method based on the definition of a group of photos in the same scene is characterized by comprising the following steps:
s10, acquiring a group of multiple pictures with the same scene;
s20, selecting an ROI (region of interest) of the photo with the same scene;
s30, calculating gradient information of all color channels of the image in the ROI area to obtain a gradient map;
s40, counting the histogram distribution of the gradient map according to a common gray map mode;
s50, intercepting a section of distribution of the number of the first T% pixels with the highest intensity in each histogram distribution, wherein T is an adjustable parameter;
s60, carrying out respective weighted average on the intercepted distribution to obtain a scalar value;
and S70, judging the definition of the photo in the same scene according to the scalar value.
Preferably, the step S10 obtains a group of multiple photos of the same scene by taking the photos in the scene or extracting the photos from an album, wherein the taking in the scene is a single shot or an automatic continuous shot.
Preferably, in step S20, the ROI region of the same scene photograph is selected as the region of 1/4 total resolution of the image at the very center of the selected image.
Preferably, in the step S30, a Sobel calculation method is adopted in calculating gradient information of each color channel of the image in the ROI area to obtain a gradient map, where there is no special requirement for image color components of the gradient information, each color component can be calculated, and an obtained final result can also be used as an evaluation criterion of sharpness contrast on the component.
Preferably, the T value in a segment of the distribution in which the step S50 cuts the top T% pixel count with the highest intensity in each histogram distribution is selected as 10.
Preferably, the method further comprises step S80 of marking the picture with the highest definition and highlighting the picture with the highest definition after step S70.
Preferably, the step S70 is followed by the step S90 of sorting the photos of the same scene according to the degree of sharpness.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides an evaluation method based on the definition of a group of photos in the same scene, which comprises the following steps: s10, acquiring a group of multiple pictures with the same scene; s20, selecting an ROI (region of interest) of the photo with the same scene; s30, calculating gradient information of all color channels of the image in the ROI area to obtain a gradient map; s40, counting the histogram distribution of the gradient map according to a common gray map mode; s50, intercepting a section of distribution of the number of the first T% pixels with the highest intensity in each histogram distribution, wherein T is an adjustable parameter; s60, carrying out respective weighted average on the intercepted distribution to obtain a scalar value; and S70, judging the definition of the photo in the same scene according to the scalar value. The method can quickly and stably give the definition evaluation of a group of photos of the same scene with good noise resistance and basically no deviation from subjective evaluation, and therefore contrast and optimization of the group of photos are provided.
Drawings
FIG. 1 is a block flow diagram of a preferred embodiment of the present invention;
FIG. 2 is a block flow diagram of another preferred embodiment of the present invention;
fig. 3 is a flow chart of yet another preferred embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and effects of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1-3, the method for evaluating the sharpness of a group of photos in the same scene according to the present invention comprises the following steps:
s10, acquiring a group of multiple pictures with the same scene;
s20, selecting an ROI (region of interest) of the photo with the same scene;
s30, calculating gradient information of all color channels of the image in the ROI area to obtain a gradient map;
s40, counting the histogram distribution of the gradient map according to a common gray map mode;
s50, intercepting a section of distribution of the number of the first T% pixels with the highest intensity in each histogram distribution, wherein T is an adjustable parameter;
s60, carrying out respective weighted average on the intercepted distribution to obtain a scalar value;
and S70, judging the definition of the photo in the same scene according to the scalar value.
The method comprises the steps of obtaining a group of multiple pictures in the same scene, selecting an ROI (region of interest) area of the picture in the same scene, calculating gradient information of various color channels of the picture in the ROI area to obtain a gradient map, counting histogram distribution of the gradient map according to a common gray map mode, intercepting a section of distribution of the first T% of pixels with the highest intensity in each histogram distribution, wherein T is an adjustable parameter, selecting a value of T according to requirements, carrying out respective weighted average on the intercepted distribution to obtain a scalar value, using the scalar value as an index for evaluating and comparing the ROI definition of each frame of the picture, and judging the definition of the picture in the same scene according to the scalar value.
In some embodiments, the step S10 is to take a group of multiple photos of the same scene by taking the photos in the scene as a single shot or automatically taking the photos in a continuous shooting mode. Specifically, regarding the photo album extraction, the photos in the photo album are judged to be photos in the same scene through internal program setting, for example, the shooting time of the photos is analyzed, or the photo background is preliminarily judged through a photo judging module, and if the photos are similar backgrounds, the photos are judged to be photos in the same scene. Then, regarding the field shooting, a plurality of pictures can be manually shot in a single mode or the pictures can be automatically shot in a continuous mode, and pictures with the same scene and different definition can exist in the shooting process due to hand trembling and the like.
In some embodiments, the step S20 selects the ROI region of the same scene photograph as the region of 1/4 total resolution of the image at the very center of the selected image. The ROI is an interested area which is a key reference for evaluating the image definition, the definition value of the selected area is given with high weight, the definition values of the other areas are given with low weight or 0 weight, and finally the definition values of the whole image are obtained through addition. The selection mode includes not limited to a plurality of image block region compositions in a nine-grid (well-shaped), a central or boundary region composition behind a golden section, and a specific region composition set by human subjectivity, in the scheme, for compatibility and universality, based on the assumption that the scene of interest will be presented at the center of the picture in the common practice during photographing, an ROI (region of interest) is set in the region of the picture (set, the width w and the height h of the picture) and circled by four points of coordinates A [ w/4, h/4], B [ w/4,3 x h/4], C [3 x w/4,3 x h/4] and D [3 x w/4, h/4], and comprises most common scenes of interest in photographing.
In some embodiments, the step S30 adopts a Sobel calculation method in calculating gradient information of each color channel of the image in the ROI region to obtain a gradient map, where there is no special requirement for image color components for calculating gradient information, each color component can be calculated, and the obtained final result can also be used as an evaluation criterion of sharpness contrast on the component. For example, the processing is performed on R, G and B channels under an RGB three-channel 24-bit model, the processing based on a brightness channel V under an HSV color model, the processing based on a brightness channel Y under an YUV color model, and the obtained final result can also be used as an evaluation standard of the definition contrast on the component. The Sobel calculation method mentioned in this embodiment is to use the following convolution templates Gx and Gy to perform convolution on single channel data of an image
The convolution of the Gx template obtains an image matrix Cx, the convolution of the Gy template obtains an image matrix Cy, and the final sobel convolution result Cres is the square root of (Cx + Cy).
In some embodiments, the step S50 selects the T value in a segment of the distribution that intercepts the top T% pixels with the highest intensity in each histogram distribution to be 10, and considers the contribution of the noise component in the image gradient information to the gradient, and the balance of the contribution of the photo scene content to the gradient.
In some embodiments, the step S70 is followed by marking the picture with the highest definition and highlighting the picture with the highest definition. Specifically, by comparing the scalar values of the photos, the photo with the highest scalar value is the photo with the highest definition, and then the photo is highlighted in the album, so that the user can visually obtain the photo with the highest definition of the group of photos with the same scene.
In some embodiments, after the step S70, the method further includes sorting the photos of the same scene according to the degree of sharpness. Specifically, the group of photos in the same scene is sorted according to the scalar value by comparing the scalar value of each photo, so that the user can clearly know the definition condition.
In summary, the working principle of the invention is as follows:
the invention provides an evaluation method based on the definition of a group of photos in the same scene, which comprises the following steps: s10, acquiring a group of multiple pictures with the same scene; s20, selecting an ROI (region of interest) of the photo with the same scene; s30, calculating gradient information of all color channels of the image in the ROI area to obtain a gradient map; s40, counting the histogram distribution of the gradient map according to a common gray map mode; s50, intercepting a section of distribution of the number of the first T% pixels with the highest intensity in each histogram distribution, wherein T is an adjustable parameter; s60, carrying out respective weighted average on the intercepted distribution to obtain a scalar value; and S70, judging the definition of the photo in the same scene according to the scalar value. The method can quickly and stably give the definition evaluation of a group of photos of the same scene with good noise resistance and basically no deviation from subjective evaluation, and therefore contrast and optimization of the group of photos are provided.
It should be understood that equivalents and modifications of the technical solution and inventive concept thereof may occur to those skilled in the art, and all such modifications and alterations should fall within the scope of the appended claims.
Claims (7)
1. An evaluation method based on the definition of a group of photos in the same scene is characterized by comprising the following steps:
s10, acquiring a group of multiple pictures with the same scene;
s20, selecting an ROI (region of interest) of the photo with the same scene;
s30, calculating gradient information of all color channels of the image in the ROI area to obtain a gradient map;
s40, counting the histogram distribution of the gradient map according to a common gray map mode;
s50, intercepting a section of distribution of the number of the first T% pixels with the highest intensity in each histogram distribution, wherein T is an adjustable parameter;
s60, carrying out respective weighted average on the intercepted distribution to obtain a scalar value;
and S70, judging the definition of the photo in the same scene according to the scalar value.
2. The method for evaluating the sharpness of a group of photos with the same scene according to claim 1, wherein the step S10 is to take a group of multiple photos with the same scene by taking the photos in the scene or taking the photos from an album, wherein the taking in the scene is a single shot or an automatic continuous shot.
3. The method for claim 1, wherein the step S20 is performed to select the ROI area of the co-scene picture as the area of 1/4 total resolution of the image at the very center of the selected image.
4. The method for evaluating the sharpness of a group of photos of the same scene as in claim 1, wherein the step S30 includes calculating gradient information of each color channel of the image in the ROI area to obtain a gradient map by using a Sobel calculation method, wherein the image color components for calculating the gradient information have no special requirement, and can be calculated on each color component, and the final result can be used as an evaluation criterion for sharpness contrast on the component.
5. The method for evaluating the sharpness of a photograph based on a group of identical scenes as claimed in claim 1, wherein said step S50 selects the T value in a segment of the distribution in which the top T% pixel number with the highest intensity in each histogram distribution is cut as 10.
6. The method for evaluating the sharpness of a group of photos of the same scene as in claim 1, further comprising step S80 of marking the photos with the highest sharpness and highlighting the photos with the highest sharpness after step S70.
7. The method for evaluating clarity of photos with the same scene according to claim 1, further comprising, after the step S70, S90 of sorting photos with the same scene according to the clarity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011427653.XA CN112381820A (en) | 2020-12-07 | 2020-12-07 | Evaluation method based on sharpness of group of photos in same scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011427653.XA CN112381820A (en) | 2020-12-07 | 2020-12-07 | Evaluation method based on sharpness of group of photos in same scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112381820A true CN112381820A (en) | 2021-02-19 |
Family
ID=74589412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011427653.XA Pending CN112381820A (en) | 2020-12-07 | 2020-12-07 | Evaluation method based on sharpness of group of photos in same scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112381820A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AT505886A2 (en) * | 2007-01-12 | 2009-04-15 | Advanced Comp Vision Gmbh Acv | METHOD FOR DETERMINING A PICTORIAL CHARACTER |
| CN101950116A (en) * | 2010-09-14 | 2011-01-19 | 浙江工业大学 | Video automatic focusing method applied to multi-main-body scene |
| CN104182962A (en) * | 2013-05-28 | 2014-12-03 | 腾讯科技（深圳）有限公司 | Picture definition evaluation method and device |
| JP2017076240A (en) * | 2015-10-14 | 2017-04-20 | キヤノン株式会社 | Image processing apparatus, image processing method, and program |
| CN107105150A (en) * | 2016-02-23 | 2017-08-29 | 中兴通讯股份有限公司 | A kind of method, photographic method and its corresponding intrument of selection photo to be output |
| CN107507173A (en) * | 2017-08-15 | 2017-12-22 | 上海交通大学 | A kind of full slice image without refer to intelligibility evaluation method and system |
| CN108805033A (en) * | 2018-05-22 | 2018-11-13 | 中国科学院自动化研究所 | Shooting image preferred method and device based on partial gradient distribution |
| CN110335246A (en) * | 2019-05-29 | 2019-10-15 | 成都数之联科技有限公司 | A kind of license picture clarity evaluation method |
-
2020
- 2020-12-07 CN CN202011427653.XA patent/CN112381820A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AT505886A2 (en) * | 2007-01-12 | 2009-04-15 | Advanced Comp Vision Gmbh Acv | METHOD FOR DETERMINING A PICTORIAL CHARACTER |
| CN101950116A (en) * | 2010-09-14 | 2011-01-19 | 浙江工业大学 | Video automatic focusing method applied to multi-main-body scene |
| CN104182962A (en) * | 2013-05-28 | 2014-12-03 | 腾讯科技（深圳）有限公司 | Picture definition evaluation method and device |
| JP2017076240A (en) * | 2015-10-14 | 2017-04-20 | キヤノン株式会社 | Image processing apparatus, image processing method, and program |
| CN107105150A (en) * | 2016-02-23 | 2017-08-29 | 中兴通讯股份有限公司 | A kind of method, photographic method and its corresponding intrument of selection photo to be output |
| CN107507173A (en) * | 2017-08-15 | 2017-12-22 | 上海交通大学 | A kind of full slice image without refer to intelligibility evaluation method and system |
| CN108805033A (en) * | 2018-05-22 | 2018-11-13 | 中国科学院自动化研究所 | Shooting image preferred method and device based on partial gradient distribution |
| CN110335246A (en) * | 2019-05-29 | 2019-10-15 | 成都数之联科技有限公司 | A kind of license picture clarity evaluation method |
Non-Patent Citations (1)
| Title |
|---|
| 孟雪: "雾霾图像增强处理技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑(月刊)》, no. 2018, 15 March 2018 (2018-03-15), pages 18 - 19 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100667663B1 (en) | Image processing apparatus, image processing method and computer readable recording medium which records program therefore | |
| EP1918872B1 (en) | Image segmentation method and system | |
| US7133571B2 (en) | Automated cropping of electronic images | |
| DE102011078662B4 (en) | Acquiring and generating images with a high dynamic range | |
| KR100662830B1 (en) | Image processing apparatus, image processing method, and storage medium | |
| US8644638B2 (en) | Automatic localized adjustment of image shadows and highlights | |
| US10565742B1 (en) | Image processing method and apparatus | |
| KR20050094330A (en) | Image processing apparatus, image processing method and program product therefore | |
| CN110222555B (en) | Method and device for detecting skin color area | |
| JP2004310475A (en) | Image processor, cellular phone for performing image processing, and image processing program | |
| CN105512671B (en) | Photo management method based on fuzzy photo identification | |
| JP4087749B2 (en) | Image processing system and image processing method | |
| CN112381820A (en) | Evaluation method based on sharpness of group of photos in same scene | |
| RU2329535C2 (en) | Method of automatic photograph framing | |
| CN111083468B (en) | Short video quality evaluation method and system based on image gradient | |
| CN114219723A (en) | Image enhancement method, image enhancement device and computer readable storage medium | |
| CN111047669B (en) | Color taking method and device | |
| RU2383924C2 (en) | Method for adaptive increase of sharpness of digital photographs during printing | |
| CN111752506B (en) | Digital work display method, display device and computer readable medium | |
| CN112883989A (en) | Picture processing method and device | |
| CN113888657B (en) | Screen color acquisition method and device, camera equipment and storage medium | |
| CN107992853A (en) | Eye detection method, device, computer equipment and storage medium | |
| Saputra et al. | Local pixel analysis in color constancy to improve feature extraction process | |
| CN113487513A (en) | Picture brightness adjusting method and adjusting device and storage medium thereof | |
| Stentiford et al. | Attention-Based Colour Correction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |