CN106910180A - A kind of image quality measure method and device - Google Patents
A kind of image quality measure method and device Download PDFInfo
- Publication number
- CN106910180A CN106910180A CN201510970440.4A CN201510970440A CN106910180A CN 106910180 A CN106910180 A CN 106910180A CN 201510970440 A CN201510970440 A CN 201510970440A CN 106910180 A CN106910180 A CN 106910180A
- Authority
- CN
- China
- Prior art keywords
- image
- assessed
- area
- point set
- delaunay triangulation
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of image quality measure method and apparatus, by gathering characteristics of image point set, it is characterized point set and builds Delaunay triangular nets, and according to the uniqueness of Delaunay triangular nets, the related data of Delaunay triangular nets is calculated, carrys out evaluation image quality, whole evaluation procedure is simply efficient, and carried out from computer vision angle, it is adaptable to automated graphics are recognized and augmented reality field.
Description
Technical field
The present invention relates to image procossing calculating field, more particularly to a kind of image quality measure method and device.
Background technology
In the field of image recognition based on content, an image is characterized generally in the way of feature point set,
During image recognition, the retrieval result of image is determined according to the matching relationship between characteristic point pair, in extreme feelings
Under condition, there was only a small amount of matching characteristic point pair between retrieval result and target image, now, because feature is retouched
The limitation of algorithm is stated, if when searching database is set up, picture quality is poor (such as to meet condition
Characteristic point is little, or skewness), it will be difficult to correct result is retrieved under extreme conditions, can be disturbed
The normal work of image identification system, influences image recognition effect, and then influence Consumer's Experience.
Existing image quality evaluating method is broadly divided into：Reference image quality appraisement and non-reference picture quality
Evaluate.Reference image quality appraisement method due to needing to refer to figure, using when very flexible, be not suitable for image
Identification field.And it is existing without mainly having artificial evaluation method and according to human eye with reference to plot quality evaluation method
Subjective vision system founding mathematical models, and the quality of image, both mode are calculated by specific formula
All referring to human eye subjective vision, and automated graphics identification belongs to computer vision field, and these picture qualities are commented
Valency method can not all reach promising result.
The content of the invention
It is an object of the invention to provide a kind of image quality measure method and device, picture quality is commented
Estimate, solve, when image retrieval database is set up, because lacking image quality measure system, to be put into multimass
The poor problem of the image recognition effect that is caused compared with difference image, can improve the knowledge of image identification system from source
Other accuracy rate, lifts Consumer's Experience.
In order to realize foregoing invention purpose, the invention provides a kind of image quality measure method, including：
Feature extraction is carried out to image to be assessed, the feature point set data of image to be assessed, feature point set are obtained
Data include positional information of each characteristic point in image-region；
Feature point set to the image to be assessed carries out Delaunay Triangulation, obtains image to be assessed
Corresponding Delaunay triangulation network network；
The area area [i] of each triangle in the Delaunay triangulation network network is calculated, i is more than or equal to 1
Integer less than or equal to n, n is Delaunay triangulation network network intermediate cam figurate number amount；
Area occupied ratio of the Delaunay triangulation network network in image to be assessed is calculated according to area [i],
And quality evaluation is carried out to image to be assessed according to the area occupied ratio.
Preferably, methods described also includes：Calculate the distribution smoothness of area [i]；Calculated according to described
The distribution smoothness of area occupied ratio and area [i] of the Delaunay triangulation network network in image to be assessed is treated
Assessment image carries out quality evaluation.
Wherein, the distribution smoothness for calculating area [i] is specially：Calculate the average mean and variance of area [i]
variance；Calculate (area [i]-mean)2In maximum maxSub；The distribution smoothness of area [i]=
1-sqrt(variance/maxSub)。
Preferably, the feature point set to the image to be assessed carries out Delaunay Triangulation, tool
Body is：Spatial classification is carried out to each characteristic point in the feature point set of image to be assessed, is built according to ranking results
Delaunay triangular nets.
Preferably, the spatial classification refers to the feature in feature point set is clicked through according to the positional information of characteristic point
Row median-of-three sort, specially：By characteristic point in feature point set in x-axis and y-axis diameter maximum/minimum axle
As sequence axle；Two intermediate values of characteristic point for constituting the diameter are calculated, changing former feature point set makes spatially
Characteristic point on the left of intermediate value is located on the left of median point in data acquisition system, and it is right that right-hand point is located at median point
Side；Then the point set that the point set and right-hand point for being constituted to left-hand point are constituted carries out above-mentioned Recursion process, Zhi Daozhong
Value side characteristic point quantity is less than 2.
Accordingly, the present invention also proposes a kind of image quality measure device, including：
Characteristic extracting module, for carrying out feature extraction to image to be assessed, obtains the feature of image to be assessed
Point set data, feature point set data include positional information of each characteristic point in image-region；
Triangulation module, carries out Delaunay triangles and cuts open for the feature point set to the image to be assessed
Point, obtain the corresponding Delaunay triangulation network network of image to be assessed；
Computing module, the area area [i] for calculating each triangle in the Delaunay triangulation network network,
And area occupied ratio of the Delaunay triangulation network network in image to be assessed is calculated according to area [i], its
Middle i is the integer less than or equal to n more than or equal to 1, and n is Delaunay triangulation network network intermediate cam figurate number amount；
Evaluation module, the area occupied ratio calculated according to computing unit carries out matter to image to be assessed
Amount assessment.
Preferably, the computing unit is additionally operable to calculate the distribution smoothness of area [i]；The assessment unit root
The area occupied ratio and the distribution smoothness of area [i] calculated according to computing unit are entered to image to be assessed
Row quality evaluation.
Preferably, the triangle subdivision unit carries out Delaunay to the feature point set of the image to be assessed
Triangulation, specially：Spatial classification is carried out to each characteristic point in the feature point set of image to be assessed, according to
Ranking results build Delaunay triangular nets.
Compared with prior art, the present invention has the advantages that：
Image quality measure method and apparatus of the present invention, for automated graphics identification field is designed, lead to
Collection characteristics of image point set is crossed, point set is characterized and is built Delaunay triangular nets, according to Delaunay
The uniqueness of triangular net, calculates the related data of Delaunay triangular nets, carrys out evaluation image matter
Amount, whole evaluation procedure is simply efficient, and is carried out from computer vision angle, is more suitable for automated graphics
Identification field.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to implementing
Example or the accompanying drawing to be used needed for description of the prior art are briefly described, it should be apparent that, describe below
In accompanying drawing be only some embodiments of the present invention, for those of ordinary skill in the art, do not paying
On the premise of going out creative labor, other accompanying drawings can also be obtained according to these accompanying drawings：
Fig. 1 is embodiment of the present invention image quality measure method flow schematic diagram；
Fig. 2 is a feature point set schematic diagram in one embodiment of the invention；
Fig. 3 is embodiment of the present invention image quality measure apparatus structure schematic diagram.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clearly
Chu, it is fully described by, it is clear that described embodiment is only a part of embodiment of the invention, rather than
Whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creation
Property work under the premise of the every other embodiment that is obtained, belong to the scope of protection of the invention.
Computer vision has natural difference with human eye vision, and the image that some human eyes can be readily identified exists
Computer is difficult identification in the eyes, and such as characteristic point is few but gem-pure image.In field of image recognition,
And the augmented reality field of image recognition is used, image storage is the premise of image recognition, but by
Do not know computer vision is how to recognize image in most of user, at will place some storage images,
When being put in storage picture quality and being poor (such as meet the characteristic point of condition seldom, or skewness), retrieval is very
Easily failure.Therefore, it is necessary to an image quality measure method, informs that user should when image is put in storage
Whether figure is easily recognizable to the computer, and retrieval result is met the target of user.
The present invention utilizes the uniqueness of Delaunay triangulation network network, is the feature point set of each storage image
A Delaunay triangulation network network is built, according to the related data of Delaunay triangulation network network, assessment storage
Picture quality.
Because the present invention uses Delaunay triangle subdivisions, therefore before specific embodiment is introduced, first
Introduce Delaunay triangular nets.
Delaunay triangular nets are the networks for carrying out Delaunay Triangulation to point set and being formed,
Meet the definition of Delaunay Triangulation, it is necessary to meet two important criterions：
1) empty circle characteristic：Delaunay triangulation network is unique (any 4 points can not be concyclic), in Delaunay
Other points are not had in network of triangle in the range of the circumscribed circle of any triangle to exist；
2) minimum angle characteristic is maximized：In the triangulation that scatterplot collection is likely to form, Delaunay triangles
The minimum angle of the triangle that subdivision is formed is maximum.In this sense, Delaunay triangulation network is " most
Close to regularization " the triangulation network.In particular in two adjacent triangulars into convex quadrangle
Diagonal, after being exchanged with each other, the minimum angle of six interior angles no longer increases.
Delaunay triangulation network network possesses following excellent specific property：
1) it is closest：With nearest three-point shape triangularity, and each line segment (side of triangle) is all non-intersect；
2) uniqueness：No matter being built since the where of region, consistent result finally will be all obtained；
3) optimality：What if the diagonal of the convex quadrangle that any two adjacent triangle is formed can be exchanged
Words, then minimum angle will not become big in two triangles, six interior angles；
4) it is most regular：If the minimum angle of each triangle in the triangulation network is carried out into ascending order arrangement,
The numerical value that the arrangement of Delaunay triangulation network is obtained is maximum；
5) it is regional：The triangle for closing on can be only influenceed when newly-increased, deletion, mobile some summit；
6) shell with convex polygon：The outermost border of the triangulation network forms a shell for convex polygon.
Below, the specific embodiment of the invention is introduced with reference to accompanying drawing.
It is a kind of image quality measure method flow schematic diagram of the embodiment of the present invention referring to Fig. 1, including it is as follows
Step：
S101：Feature extraction is carried out to image to be assessed, the feature point set data of image to be assessed are obtained, it is special
Levying point set data includes positional information of each characteristic point in image-region, this steps characteristic extraction side
Method, can use the feature extracting method based on Scale invariant, such as ORB, SIFT, SURF etc., extract
Come characteristic point data in addition to positional information, can also including yardstick, direction, characterization information etc.,
Simply image quality measure of the present invention only uses positional information.
S102：Feature point set to the image to be assessed carries out Delaunay Triangulation, obtains to be evaluated
(triangular net has uniqueness, i.e., to same point set to estimate the corresponding Delaunay triangulation network network of image
Built, consistent result will be all obtained since which point, while to a same height concentrated
Collection carries out deletion action, and the triangulation network for obtaining is also consistent), this step Delaunay Triangulation is specific
Mode, is described below.
S103：The area area [i] of each triangle in the Delaunay triangulation network network is calculated, i is big
In the integer equal to 1 less than or equal to n, n is Delaunay triangulation network network intermediate cam figurate number amount, every calculating
During area area [i] of individual triangle, can be calculated using pixel unit；
S104：Occupancy face of the Delaunay triangulation network network in image to be assessed is calculated according to area [i]
Product ratio.The area of all triangles in Delaunay triangulation network network is added, Delaunay triangles are obtained
Network area；According to image to be assessed height wide, image area to be assessed is calculated；Delaunay triangulation network network face
Product obtains occupancy face of the Delaunay triangulation network network in image to be assessed the ratio between with image area to be assessed
Product ratio, the ratio is between 0.0~1.0, and this ratio is higher, represents image characteristic point distribution to be assessed
Wider, characteristic point distribution is wider, higher (in image retrieval, in the case where blocking to blocking adaptability
Complete retrieval success rate higher), thus we to can be understood as picture quality better.
S105:The distribution smoothness of area [i] is calculated, specific calculation includes the steps of following ABC tri-:
A:Calculate the average mean and variance variance of area [i]：
Mean=sum (area [i])/n；
Variance=sum (area [i] * area [i])/n-mean*mean；
B:Calculate the maximum maxSub in (area [i]-mean) 2:
C:Calculate：The distribution smoothness of area [i], smoothness is designated as smoothVal：
SmoothVal=1-sqrt (variance/maxSub), smoothness is bigger, represents characteristic point distribution
It is more uniform, it is higher to blocking adaptability, it is better that picture quality is interpreted as in the present invention.
In above-mentioned computing formula, sum represents summation, and Sqrt represents sqrt.
S106：According to area occupied of the Delaunay triangulation network network for calculating in image to be assessed
The distribution smoothness of ratio and area [i] carries out quality evaluation to image to be assessed, during assessment, according to reality
Demand freely defines the two weight, and such as the two weight can be respectively to account for 0.5, or 0.3,0.7,
Even 0.0,1.I.e. in present invention specific implementation, can be according to the Delaunay triangles for calculating
Any one parameter in the distribution smoothness of area occupied ratio and area [i] of the network in image to be assessed
Quality evaluation is carried out to image to be assessed, preferably the two combines to carry out image quality measure certainly.
In step S102, the feature point set to the image to be assessed carries out Delaunay Triangulation, tool
Body is：
Spatial classification is carried out to each characteristic point in the feature point set of image to be assessed, is built according to ranking results
Delaunay triangular nets, the spatial classification can be median-of-three sort, and the median-of-three sort refers to according to spy
The positional information levied a little carries out median-of-three sort to the characteristic point in feature point set, specially：By in feature point set
Characteristic point in x-axis and y-axis diameter maximum/minimum axle as sequence axle；Calculate constitute the diameter two
The intermediate value of characteristic point, changing former feature point set makes to be spatially positioned in characteristic point on the left of intermediate value in data acquisition system
Positioned at median point left side, right-hand point is located at median point right side；Then the point set and right-hand point for being constituted to left-hand point
The point set of composition carries out above-mentioned Recursion process, and until intermediate value side, characteristic point quantity is less than 2.Wherein x-axis is straight
Footpath refers in feature point set, the x coordinate of each characteristic point, the absolute value of the difference of maxima and minima；Y-axis is straight
Footpath refers in feature point set, the y-coordinate of each characteristic point, the absolute value of the difference of maxima and minima, referring to figure
2, it is a point set, including following 7 points：[(- 2,2) (2.5, -5) (2,1) (- 4, -1.5) (- 7.5,2.5)
(7,2) (1, -2.5)], this 7 x-axis a diameter of 14 of the point set of point composition, y-axis a diameter of 7.5,
Assuming that during median-of-three sort with xy axles week diameter in the greater for sort axle, then during the first minor sort, with x-axis
Used as sequence axle, intermediate value is 0, and (- 7.5,2.5), (- 2,2), (- 4, -1.5) three points are come into median point
Left side, other four points are placed on median point right side.Then Recursion process is carried out to left side point set and right side point set,
Left and right sides point set is found again and be relatively large in diameter in xy axles axle, calculate two characteristic points for constituting the diameter
Intermediate value, during the characteristic point that changing former feature point set makes to be spatially positioned on the left of intermediate value is located in data acquisition system
Value point left side, right-hand point is located at median point right side.
It is a kind of image quality measure device of the embodiment of the present invention referring to Fig. 3, including：
Characteristic extracting module 1, for carrying out feature extraction to image to be assessed, obtains the spy of image to be assessed
Point set data are levied, feature point set data include positional information of each characteristic point in image-region, feature
Extracting method, can use the feature extracting method based on Scale invariant, such as ORB, SIFT, SURF etc..
Triangulation module 2, carries out Delaunay triangles and cuts open for the feature point set to the image to be assessed
Point, the corresponding Delaunay triangulation network network of image to be assessed is obtained, Delaunay Triangulation method is referred to
To the detailed narration of S102 parts in image quality measure embodiment of the method, this place does not repeat.
Computing module 3, the area for calculating each triangle in the Delaunay triangulation network network
Area [i], and occupancy of the Delaunay triangulation network network in image to be assessed is calculated according to area [i]
Area ratio, wherein i are the integer less than or equal to n more than or equal to 1, and n is in Delaunay triangulation network network
Number of triangles；Each parameter calculation refers to previous embodiment step S103, S104 portion in computing module
Point.
Evaluation module 4, calculates the 3 area occupied ratios for going out and image to be assessed is entered according to computing unit
Row quality evaluation, the ratio is between 0.0~1.0, and this ratio is higher, represents image characteristic point to be assessed
Distribution is wider, and characteristic point distribution is wider, and higher to blocking adaptability, picture quality is better.
In another embodiment, the computing unit 3 is additionally operable to calculate the distribution smoothness of area [i],
Smoothness is bigger, represents that characteristic point distribution is more uniform, higher to blocking adaptability, is interpreted as in the present invention
Picture quality is better.The area occupied ratio that the assessment unit 4 is calculated according to computing unit 3 and
The distribution smoothness of area [i] carries out quality evaluation to image to be assessed, during assessment, according to the actual requirements certainly
By the two weight of definition, such as the two weight can be respectively to account for 0.5, or 0.3,0.7, or even 0.0,
1.I.e. in present invention specific implementation, can be according to the Delaunay triangulation network network for calculating to be evaluated
Any one parameter in the distribution smoothness of the area occupied ratio and area [i] in image is estimated to figure to be assessed
As carrying out quality evaluation, preferably the two combines to carry out image quality measure certainly.
Image quality measure method and apparatus of the present invention, it is adaptable to field of image recognition, and use image knowledge
Other augmented reality field, when image is put in storage, is estimated using the inventive method to storage image,
The prompting of assessment result can be given in some way, city user knows certain picture quality if appropriate for being used for
As the storage image of image retrieval, retrieval result is set to meet the target of user.
All features disclosed in this specification, or disclosed all methods or during the step of, except mutual
Beyond the feature and/or step mutually repelled, can combine by any way.
Any feature disclosed in this specification (including any accessory claim, summary and accompanying drawing), removes
Non-specifically describe, can alternative features equivalent by other or with similar purpose replaced.That is, unless
Especially narration, each feature is an example in a series of equivalent or similar characteristics.
The invention is not limited in foregoing specific embodiment.The present invention expand to it is any in this manual
The new feature of disclosure or any new combination, and disclose any new method or process the step of or it is any
New combination.
Claims (8)
1. a kind of image quality measure method, it is characterised in that including：
Feature extraction is carried out to image to be assessed, the feature point set data of image to be assessed are obtained, feature point set data include positional information of each characteristic point in image-region；
Feature point set to the image to be assessed carries out Delaunay Triangulation, obtains the corresponding Delaunay triangulation network network of image to be assessed；
The area area [i] of each triangle in the Delaunay triangulation network network is calculated, i is the integer less than or equal to n more than or equal to 1, and n is Delaunay triangulation network network intermediate cam figurate number amount；
Area occupied ratio of the Delaunay triangulation network network in image to be assessed is calculated according to area [i], and quality evaluation is carried out to image to be assessed according to the area occupied ratio.
2.Such as claimImage quality measure method described in 1, it is characterised in that methods described also includes：Calculate the distribution smoothness of area [i]；
The distribution smoothness of area occupied ratio and area [i] according to the Delaunay triangulation network network for calculating in image to be assessed carries out quality evaluation to image to be assessed.
3.Such as claimImage quality measure method described in 2, it is characterised in that the distribution smoothness for calculating area [i] is specially：
Calculate the average mean and variance variance of area [i]；
Calculate the maximum maxSub in (area [i]-mean) 2；
The distribution smoothness of area [i]=1-sqrt (variance/maxSub).
4.Such as claimImage quality measure method any one of 1 to 3, it is characterised in that the feature point set to the image to be assessed carries out Delaunay Triangulation, specially：
Spatial classification is carried out to each characteristic point in the feature point set of image to be assessed, Delaunay triangular nets are built according to ranking results.
5.Such as claimImage quality measure method described in 4, it is characterised in that the spatial classification to refer to and carry out median-of-three sort to the characteristic point in feature point set according to the positional information of characteristic point, specially：
Using characteristic point in feature point set in x-axis and y-axis diameter maximum/minimum axle as sequence axle；
Two intermediate values of characteristic point for constituting the diameter are calculated, the characteristic point that changing former feature point set makes to be spatially positioned on the left of intermediate value is located on the left of median point in data acquisition system, and right-hand point is located at median point right side；
Then the point set that the point set and right-hand point for being constituted to left-hand point are constituted carries out above-mentioned Recursion process, and until intermediate value side, characteristic point quantity is less than 2.
6. a kind of image quality measure device, it is characterised in that including：
Characteristic extracting module, for carrying out feature extraction to image to be assessed, obtains the feature point set data of image to be assessed, and feature point set data include positional information of each characteristic point in image-region；
Triangulation module, Delaunay Triangulation is carried out for the feature point set to the image to be assessed, obtains the corresponding Delaunay triangulation network network of image to be assessed；
Computing module, area area [i] for calculating each triangle in the Delaunay triangulation network network, and area occupied ratio of the Delaunay triangulation network network in image to be assessed is calculated according to area [i], wherein i is the integer less than or equal to n more than or equal to 1, and n is Delaunay triangulation network network intermediate cam figurate number amount；
Evaluation module, the area occupied ratio calculated according to computing unit carries out quality evaluation to image to be assessed.
7.Such as claimImage quality measure device described in 6, it is characterised in that the computing unit is additionally operable to calculate the distribution smoothness of area [i]；The area occupied ratio and the distribution smoothness of area [i] that the assessment unit is calculated according to computing unit carry out quality evaluation to image to be assessed.
8.Such as claimImage quality measure device described in 6 or 7, it is characterised in that the triangle subdivision unit carries out Delaunay Triangulation to the feature point set of the image to be assessed, specially：
Spatial classification is carried out to each characteristic point in the feature point set of image to be assessed, Delaunay triangular nets are built according to ranking results.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510970440.4A CN106910180B (en) | 2015-12-22 | 2015-12-22 | A kind of image quality measure method and device |
| PCT/CN2016/110367 WO2017107867A1 (en) | 2015-12-22 | 2016-12-16 | Image quality evaluation method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510970440.4A CN106910180B (en) | 2015-12-22 | 2015-12-22 | A kind of image quality measure method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106910180A true CN106910180A (en) | 2017-06-30 |
| CN106910180B CN106910180B (en) | 2019-08-20 |
Family
ID=59089140
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510970440.4A Active CN106910180B (en) | 2015-12-22 | 2015-12-22 | A kind of image quality measure method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106910180B (en) |
| WO (1) | WO2017107867A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107610110A (en) * | 2017-09-08 | 2018-01-19 | 北京工业大学 | Across the yardstick image quality evaluating method that a kind of global and local feature is combined |
| CN108829595A (en) * | 2018-06-11 | 2018-11-16 | Oppo（重庆）智能科技有限公司 | Test method, device, storage medium and electronic equipment |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107993230B (en) * | 2017-12-18 | 2021-11-19 | 辽宁师范大学 | Image tampering detection method based on triangular mesh comprehensive characteristics |
| CN108983233B (en) * | 2018-06-13 | 2022-06-17 | 四川大学 | PS point combination selection method in GB-InSAR data processing |
| CN112070739B (en) * | 2020-09-03 | 2024-10-18 | Oppo广东移动通信有限公司 | Image processing method, device, electronic equipment and storage medium |
| CN115249330B (en) * | 2022-09-05 | 2023-04-28 | 中国测绘科学研究院 | Method and system for evaluating vegetation connectivity |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012016242A2 (en) * | 2010-07-30 | 2012-02-02 | Aureon Biosciences, Inc. | Systems and methods for segmentation and processing of tissue images and feature extraction from same for treating, diagnosing, or predicting medical conditions |
| CN102760293A (en) * | 2012-06-14 | 2012-10-31 | 南京信息工程大学 | Image quality evaluation method based on distance matrix |
| CN103366378A (en) * | 2013-07-26 | 2013-10-23 | 深圳大学 | Reference-free type image quality evaluation method based on shape consistency of condition histogram |
| CN103996192A (en) * | 2014-05-12 | 2014-08-20 | 同济大学 | Non-reference image quality evaluation method based on high-quality natural image statistical magnitude model |
| CN104282019A (en) * | 2014-09-16 | 2015-01-14 | 电子科技大学 | Blind image quality evaluation method based on natural scene statistics and perceived quality propagation |
-
2015
- 2015-12-22 CN CN201510970440.4A patent/CN106910180B/en active Active
-
2016
- 2016-12-16 WO PCT/CN2016/110367 patent/WO2017107867A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012016242A2 (en) * | 2010-07-30 | 2012-02-02 | Aureon Biosciences, Inc. | Systems and methods for segmentation and processing of tissue images and feature extraction from same for treating, diagnosing, or predicting medical conditions |
| CN102760293A (en) * | 2012-06-14 | 2012-10-31 | 南京信息工程大学 | Image quality evaluation method based on distance matrix |
| CN103366378A (en) * | 2013-07-26 | 2013-10-23 | 深圳大学 | Reference-free type image quality evaluation method based on shape consistency of condition histogram |
| CN103996192A (en) * | 2014-05-12 | 2014-08-20 | 同济大学 | Non-reference image quality evaluation method based on high-quality natural image statistical magnitude model |
| CN104282019A (en) * | 2014-09-16 | 2015-01-14 | 电子科技大学 | Blind image quality evaluation method based on natural scene statistics and perceived quality propagation |
Non-Patent Citations (1)
| Title |
|---|
| YAO,Z等: "Quality Assessment of Fingerprints with Minutiae Delaunay", 《ICISSP 2015-1ST INTERNATIONAL CONFERENCE ON INFORMATION SYSTEMS SECURITY AND PRIVACY》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107610110A (en) * | 2017-09-08 | 2018-01-19 | 北京工业大学 | Across the yardstick image quality evaluating method that a kind of global and local feature is combined |
| CN107610110B (en) * | 2017-09-08 | 2020-09-25 | 北京工业大学 | Global and local feature combined cross-scale image quality evaluation method |
| CN108829595A (en) * | 2018-06-11 | 2018-11-16 | Oppo（重庆）智能科技有限公司 | Test method, device, storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106910180B (en) | 2019-08-20 |
| WO2017107867A1 (en) | 2017-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106910180A (en) | A kind of image quality measure method and device | |
| CN109409437B (en) | Point cloud segmentation method and device, computer readable storage medium and terminal | |
| CN107578418B (en) | Indoor scene contour detection method fusing color and depth information | |
| CN101216895B (en) | An automatic extracting method for ellipse image features in complex background images | |
| CN108961271B (en) | Color 3D point cloud hyper-voxel concave-convex segmentation algorithm | |
| CN103353987B (en) | A kind of superpixel segmentation method based on fuzzy theory | |
| CN104376556B (en) | A kind of rock CT images Target Segmentation method | |
| CN113077476B (en) | Height measurement method, terminal device and computer storage medium | |
| CN106991693B (en) | Binocular Stereo Matching Method Based on Fuzzy Support Weight | |
| CN109446889A (en) | Object tracking method and device based on twin matching network | |
| CN106709948A (en) | Quick binocular stereo matching method based on superpixel segmentation | |
| CN100545865C (en) | An Automatic Segmentation Method for Optimizing the Initial Segmentation Boundary of Image | |
| CN109711268B (en) | Face image screening method and device | |
| CN105118090A (en) | Adaptive point-cloud filtering method for complex terrain structure | |
| CN104778701A (en) | Local image describing method based on RGB-D sensor | |
| CN108573502A (en) | A Method of Automatically Measuring Cobb Angle | |
| CN110363170B (en) | Video face changing method and device | |
| CN111951196B (en) | Graph-based progressive point cloud downsampling method and device | |
| CN108230338A (en) | A kind of stereo-picture dividing method based on convolutional neural networks | |
| CN109377564B (en) | Method and device for virtual fitting based on monocular depth camera | |
| CN113763404B (en) | Foam image segmentation method based on optimization mark and edge constraint watershed algorithm | |
| CN111754618A (en) | An object-oriented method and system for multi-level interpretation of real 3D model | |
| CN115049925A (en) | Method for extracting field ridge, electronic device and storage medium | |
| CN107452058A (en) | The simplifying method generation of simulation model point and system that feature based retains | |
| CN106780479A (en) | A kind of high precision image fuzzy detection method based on deep learning |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |