CN107967478A - A kind of similar decision method of simplified DCT pixel grey scales picture - Google Patents
A kind of similar decision method of simplified DCT pixel grey scales picture Download PDFInfo
- Publication number
- CN107967478A CN107967478A CN201711296322.5A CN201711296322A CN107967478A CN 107967478 A CN107967478 A CN 107967478A CN 201711296322 A CN201711296322 A CN 201711296322A CN 107967478 A CN107967478 A CN 107967478A
- Authority
- CN
- China
- Prior art keywords
- dct
- picture
- values
- pixel
- similar
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/52—Scale-space analysis, e.g. wavelet analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The invention discloses a kind of similar decision method of simplified DCT pixel grey scales picture, picture is reduced to the pixel of pixel 32*32, by dct transform, reduces the calculating process of 8*8, obtains the DCT values of each pixel, by contrasting the average of DCT values, hashed.The present invention reduces the calculating process of 8*8, has filtered pure color information, further increase the statement of details, similar judgement rate is 40% or so, although sacrificing similar judgement rate, is reduction of calculation amount, saves portion of time cost by the conversion of DCT.
Description
Technical field
The present invention relates to image processing field, the similar decision method of more particularly to a kind of simplified DCT pixel grey scales picture.
Background technology
As intellectual property obtains the attention of society, IP dispute is also more and more, and picture is as intellectual property
Important component, abuse take place frequently, but picture obligee is difficult to find abuse, causes picture obligee's right-safeguarding to be stranded
It is difficult.At present, the whole network crawler technology can obtain the pictorial information of magnanimity, but how to screen and scheme in the pictorial information of magnanimity
The similar picture of piece obligee, is a problem to be solved.With current technology, it is difficult to computer method directly to figure
Piece carries out infringement judgement, but can screen mass picture, excludes the picture of dissmilarity, and a large amount of screening figures are saved for artificial judgment
Manual time's cost of piece.
To solve the above problems, applicant has applied for entitled " a kind of picture similar decision method ", Application No.
" 2017111545551 ", this method reduce the treating capacity of colour information by the way that two pictures to be carried out to the contrast of gray scale；It is right
Picture is divided into row matrix and gray scale same order, can meet same photograph, color is scaled, scalable in proportion to wait operation
The similar judgement of photo afterwards, but if being contrasted to average gray, the details of display is fewer, judges similarity degree
Success rate it is relatively low, probably 33% or so, finally need manually to make a decision, workload is slightly larger.
In order to overcome the above problem, applicant has done details DCT processing, has added details table in another patent
State, but virtually, add calculation amount.
The content of the invention
In order to overcome the above-mentioned deficiencies of the prior art, sentence the present invention provides a kind of simplified DCT pixel grey scales picture is similar
Determine method.
A kind of similar decision method of simplified DCT pixel grey scales picture provided by the invention, comprises the following steps：
The first step：Reduce dimension of picture
It is 32x32 sizes that picture is contracted to pixel, reduces DCT and calculates.
Second step：Picture is reduced to gray scale
Picture is reduced to 32 grades of gray scales, grey level is in the same size with dimension of picture, to eliminate the influence of scaling.
3rd step：Calculate DCT
By DCT algorithms, picture is divided into frequency and the set of scalar.
4th step：Reduce DCT
Remove upper left corner 8x8 data, the 8x8 data are frequency minimum in its representative graph.
5th step：Calculate the average value of DCT
Average DCT values are calculated, remove 8 × 8DCT low frequency values, not including Section 1, if because DC coefficients have with other values
Significant difference, and average value can be declined, the low two dimensional DCT coefficients since minimum second, eliminate first DC,
Completely flat image information, i.e. pure color can be excluded, reduces calculation amount.
6th step：Contrast the average value of DCT
Each DCT values and the average contrast of DCT, higher than for 1, less than 0.
7th step：Hash
DCT values are hashed in certain sequence, one integer by 0,1 composition of composition hashes.
8th step：Hash contrast
The hash of different pictures is contrasted, how many difference obtained；It is similar or identical photo less than 5.
Beneficial effect：The present invention reduces the calculating process of 8*8, has filtered pure color information, further by the conversion of DCT
The statement of details is improved, similar judgement rate, although sacrificing similar judgement rate, is reduction of calculation amount 40% or so,
Save portion of time cost.
Brief description of the drawings
Fig. 1 present invention simplifies the similar decision method compressed picture schematic diagram of DCT pixel grey scale pictures.
Fig. 2 present invention simplifies the similar decision method gradation conversion schematic diagram of DCT pixel grey scale pictures.
Embodiment
Embodiment：
As depicted in figs. 1 and 2, the first step：Reduce dimension of picture：It is 32x32 sizes that picture is contracted to pixel, is reduced
DCT is calculated.
Second step：Picture is reduced to gray scale：Picture is reduced to 32 grades of gray scales, grey level will be with dimension of picture size
Unanimously, with the influence of elimination scaling.
3rd step：Calculate DCT：By DCT algorithms, picture is divided into frequency and the set of scalar.
4th step：Reduce DCT：Remove upper left corner 8x8 data, the 8x8 data are frequency minimum in its representative graph.
5th step：Calculate the average value of DCT：Average DCT values are calculated, remove 8 × 8DCT low frequency values, not including Section 1,
If because DC coefficients have significant difference with other values, and average value can be declined, the low two dimension since minimum second
DCT coefficient, eliminates first DC, can exclude completely flat image information, i.e. pure color, reduces calculation amount.
6th step：Contrast the average value of DCT：Each DCT values and the average contrast of DCT, higher than for 1, less than 0.
7th step：Hash：DCT values are hashed in certain sequence, one integer by 0,1 composition of composition hashes.
8th step：Hash contrast：The hash of different pictures is contrasted, how many difference obtained；It is similar or phase less than 5
Same photo.
Claims (1)
1. the similar decision method of a kind of simplified DCT pixel grey scales picture, it is characterised in that comprise the following steps：
The first step：Reduce dimension of picture
It is 32x32 sizes that picture is contracted to pixel, reduces DCT and calculates；
Second step：Picture is reduced to gray scale
Picture is reduced to 32 grades of gray scales, grey level is in the same size with dimension of picture, to eliminate the influence of scaling；
3rd step：Calculate DCT
By DCT algorithms, picture is divided into frequency and the set of scalar；
4th step：Reduce DCT
Remove upper left corner 8x8 data, the 8x8 data are frequency minimum in its representative graph；
5th step：Calculate the average value of DCT
Calculate average DCT values, remove 8 × 8DCT low frequency values, not including Section 1, if because DC coefficients have with other values it is significant
Difference, and can decline average value, the low two dimensional DCT coefficients since minimum second, eliminate first DC, can be with
Completely flat image information, i.e. pure color are excluded, reduces calculation amount；
6th step：Contrast the average value of DCT
Each DCT values and the average contrast of DCT, higher than for 1, less than 0；
7th step：Hash
DCT values are hashed in certain sequence, one integer by 0,1 composition of composition hashes；
The hash contrast of 8th step
The hash of different pictures is contrasted, how many difference obtained；It is similar or identical photo less than 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711296322.5A CN107967478A (en) | 2017-12-08 | 2017-12-08 | A kind of similar decision method of simplified DCT pixel grey scales picture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711296322.5A CN107967478A (en) | 2017-12-08 | 2017-12-08 | A kind of similar decision method of simplified DCT pixel grey scales picture |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107967478A true CN107967478A (en) | 2018-04-27 |
Family
ID=61999549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711296322.5A Withdrawn CN107967478A (en) | 2017-12-08 | 2017-12-08 | A kind of similar decision method of simplified DCT pixel grey scales picture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107967478A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109493098A (en) * | 2018-11-28 | 2019-03-19 | 上海中信信息发展股份有限公司 | Birds product digital certificate method of calibration and device |
| CN113520172A (en) * | 2021-08-06 | 2021-10-22 | 广东福尔电子有限公司 | Pressure-resistant electric cooker temperature sensor detection device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103971113A (en) * | 2013-02-04 | 2014-08-06 | 纬创资通股份有限公司 | Image recognition method, electronic device and computer program product |
| WO2015034269A1 (en) * | 2013-09-03 | 2015-03-12 | 삼성전자 주식회사 | Image processing method and device |
| CN105407353A (en) * | 2014-09-11 | 2016-03-16 | 腾讯科技（深圳）有限公司 | Image compression method and apparatus |
| CN105912643A (en) * | 2016-04-08 | 2016-08-31 | 浙江理工大学 | Image retrieval method based on content improved Average Hash |
| CN106650829A (en) * | 2017-01-04 | 2017-05-10 | 华南理工大学 | Picture similarity calculation method |
| CN107239535A (en) * | 2017-05-31 | 2017-10-10 | 北京小米移动软件有限公司 | Similar pictures search method and device |
-
2017
- 2017-12-08 CN CN201711296322.5A patent/CN107967478A/en not_active Withdrawn
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103971113A (en) * | 2013-02-04 | 2014-08-06 | 纬创资通股份有限公司 | Image recognition method, electronic device and computer program product |
| WO2015034269A1 (en) * | 2013-09-03 | 2015-03-12 | 삼성전자 주식회사 | Image processing method and device |
| CN105407353A (en) * | 2014-09-11 | 2016-03-16 | 腾讯科技（深圳）有限公司 | Image compression method and apparatus |
| CN105912643A (en) * | 2016-04-08 | 2016-08-31 | 浙江理工大学 | Image retrieval method based on content improved Average Hash |
| CN106650829A (en) * | 2017-01-04 | 2017-05-10 | 华南理工大学 | Picture similarity calculation method |
| CN107239535A (en) * | 2017-05-31 | 2017-10-10 | 北京小米移动软件有限公司 | Similar pictures search method and device |
Non-Patent Citations (1)
| Title |
|---|
| ZOUXY09: "基于感知哈希算法的视觉目标跟踪", 《CSDN》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109493098A (en) * | 2018-11-28 | 2019-03-19 | 上海中信信息发展股份有限公司 | Birds product digital certificate method of calibration and device |
| CN113520172A (en) * | 2021-08-06 | 2021-10-22 | 广东福尔电子有限公司 | Pressure-resistant electric cooker temperature sensor detection device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Maini et al. | A comprehensive review of image enhancement techniques | |
| CN100380974C (en) | Method and device for video frequency deblocking | |
| US8335394B2 (en) | Image processing method for boundary resolution enhancement | |
| CN102930517A (en) | Image enhancement method implemented through histogram equalization | |
| CN104378636B (en) | A kind of video encoding method and device | |
| CN110211070B (en) | Low-illumination color image enhancement method based on local extreme value | |
| CN103702116B (en) | A kind of dynamic range compression method and apparatus of image | |
| CN109584185A (en) | Image processing method | |
| CN105809643A (en) | Image enhancing method based on self-adaptive block channel stretching | |
| CN107967478A (en) | A kind of similar decision method of simplified DCT pixel grey scales picture | |
| CN108280836B (en) | Image processing method and device | |
| JP2007067687A5 (en) | ||
| Ding et al. | Color image enhancement with a human visual system based adaptive filter | |
| CN105279742B (en) | A kind of image de-noising method quickly based on piecemeal estimation of noise energy | |
| CN1213600C (en) | N-dimensional filter and method for N-dimensionally filtering an original image | |
| CN108052969A (en) | A kind of similar determination method of DCT pixel grey scales picture | |
| Ding et al. | Video pre-processing with JND-based Gaussian filtering of superpixels | |
| CN108521575B (en) | Method and device for reducing noise of image noise | |
| CN114862706B (en) | Tone mapping method for keeping gradient direction of image | |
| CN108038436A (en) | A kind of similar decision method of local pixel gray scale picture | |
| CN102469238B (en) | Method and system for enhancing sharpness of image by using pixel region characteristics | |
| Pan et al. | Adapting iterative Retinex computation for high-dynamic-range tone mapping | |
| CN108052972A (en) | A kind of single pass picture similar determination method of Two bors d's oeuveres | |
| TW201505014A (en) | Method and system of enhancing a backlight-scaled image | |
| Asadi et al. | Improving dark channel prior for single image dehazing |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180427 |
|
| WW01 | Invention patent application withdrawn after publication |