CN110580729A - image color matching method and device and electronic equipment - Google Patents
image color matching method and device and electronic equipment Download PDFInfo
- Publication number
- CN110580729A CN110580729A CN201810595676.8A CN201810595676A CN110580729A CN 110580729 A CN110580729 A CN 110580729A CN 201810595676 A CN201810595676 A CN 201810595676A CN 110580729 A CN110580729 A CN 110580729A
- Authority
- CN
- China
- Prior art keywords
- color
- picture
- target
- content element
- style
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0276—Advertisement creation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/40—Filling a planar surface by adding surface attributes, e.g. colour or texture
Abstract
The embodiment of the invention provides a color matching method and device for pictures and electronic equipment, wherein the method comprises the following steps: acquiring a target style data set; performing probability density estimation based on color features for content element parts in the target style picture, wherein the color features comprise: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched; and forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be subjected to color matching. According to the embodiment of the invention, the color matching scheme is automatically formed by estimating the probability density of the target style picture based on the color characteristics and combining the difference between the content elements and the product in the picture to be color-matched, so that the creation cost of the commercial picture is reduced, the time spent by a user on content color matching is reduced, and the creation efficiency of the commercial picture is greatly improved.
Description
Technical Field
The invention relates to the technical field of internet, in particular to a color matching method and device for a picture and electronic equipment.
Background
With the rapid development of electronic commerce, the demand for the design of commercial pictures (particularly, advertisement pictures) is increasing. The flat advertisement (i.e. advertisement picture) can attract the eyes of consumers to a great extent by effectively using the color, simply and clearly transmit important information, and instantly catch the mind of the consumers, thereby becoming one of the main expression means of the advertisement. The creation of the flat advertisement design requires the concentration of expression means and symbolism, and an excellent flat advertisement design has a novel feeling full of times consciousness and has unique expression technique and feeling in design. Every year, along with activities such as various festivals, brand promotion and the like, a large number of advertisement pictures are created.
In the process of implementing the invention, the inventor finds that the prior art has at least the following problems: when creating business pictures, designers usually spend a long time and effort, but the created pictures are often used for only a few days or even hours, which causes a certain waste of resources.
Disclosure of Invention
the embodiment of the invention provides a color matching method and device for a picture and electronic equipment, aiming at solving the defect of low creation efficiency of commercial pictures in the prior art, automatically recoloring content elements and reducing creation cost.
in order to achieve the above object, an embodiment of the present invention provides a color matching method for a picture, where the picture includes a product portion and a content element portion, and the method includes:
Acquiring a target style data set, wherein the target style data set consists of a plurality of target style pictures;
Performing probability density estimation based on color features for content element parts in the target style picture, wherein the color features comprise: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched;
and forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be subjected to color matching.
The embodiment of the invention also provides a color matching device for pictures, wherein the pictures comprise a product part and a content element part, and the device comprises:
the target style acquisition module is used for acquiring a target style data set, and the target style data set consists of a plurality of target style pictures;
A probability density estimation module, configured to perform probability density estimation based on color features for a content element portion in the target style picture, where the color features include: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched;
And the color matching module is used for forming a target scheme according to the probability density estimation result of the target style picture and recoloring the content element part in the picture to be subjected to color matching.
An embodiment of the present invention further provides an electronic device, including:
A memory for storing a program;
A processor for executing the program stored in the memory for:
acquiring a target style data set, wherein the target style data set consists of a plurality of target style pictures;
Performing probability density estimation based on color features for content element parts in the target style picture, wherein the color features comprise: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched;
And forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be subjected to color matching.
According to the color matching method and device for the picture and the electronic equipment, the probability density estimation based on the color characteristics is carried out on the target style picture, the color matching scheme suggestion under the style is provided for the user by combining the color difference between the content elements and the product part in the picture to be color matched, the picture after being recoloring accords with the expected style of the user, and the method and device have high practicability and effectiveness, meanwhile, the creation cost of the commercial picture is reduced, the time spent by the user on content color matching is reduced, and the creation efficiency of the commercial picture is greatly improved.
the foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
Fig. 1a is a system block diagram of a service system according to an embodiment of the present invention;
fig. 1b is a schematic view of a scenario of a service system according to an embodiment of the present invention;
FIG. 2 is a flowchart of an embodiment of a color matching method for a picture according to the present invention;
FIG. 3 is a flowchart of another embodiment of a color matching method for a picture according to the present invention;
FIG. 4 is a schematic structural diagram of an embodiment of a color matching apparatus for pictures according to the present invention;
FIG. 5a is a schematic structural diagram of another embodiment of a color matching apparatus for pictures according to the present invention;
FIG. 5b is a second schematic structural diagram of another embodiment of the color matching apparatus for pictures according to the present invention;
Fig. 6 is a schematic structural diagram of an embodiment of an electronic device according to an embodiment of the present invention.
Description of reference numerals:
41-target style acquisition module, 42-probability density estimation module, 43-color matching module, 411-analysis submodule, 412-clustering submodule, 413-preference submodule, 431-acquisition submodule, 432-probability calculation submodule, 433-determination submodule, 434-attribute adjustment submodule, 435-color matching submodule, 4111-first calculation unit, 4112-second calculation unit, 4113-third calculation unit, 4114-fourth calculation unit, 4121-vector forming unit, 4122-dimension reduction unit.
Detailed Description
exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The business picture is taken as a means of business propaganda, and the use of the color of the business picture is often related to business needs, target crowds and the like, so the application provides a solution for color matching of the picture, and the main principle is as follows: collecting a mass of existing pictures which are put into use to form a sample data set, and selecting a target style data set expected by a user from the sample data set, wherein the target style data set generally consists of a plurality of target style pictures. And performing probability density estimation based on color features on the content element part in the target style picture, forming an optimal color matching scheme according to the probability density estimation result, and performing re-coloring on the content element part (background part and/or character part) in the picture to be subjected to color matching.
Regarding the color characteristics of the content element part in the target style picture, on one hand, the color characteristics refer to the average value of the chroma, the average value of the saturation or the average value of the brightness; on the other hand, the difference between the average chromaticity value and the average saturation value or the average brightness value of the product part of the picture to be matched with colors can be used. Under the condition of not changing the color of the product, the probability density estimation is carried out by combining the target style picture and the color difference of the product, so that the characteristic distribution of the product under the specific style can be obtained, the formed color matching scheme conforms to the expected style of the user, the visual effect of catching the eyes of the user is achieved, and the practicability and the effectiveness are high.
the method provided by the embodiment of the invention can be applied to any business system with color matching capability. Fig. 1a is a system block diagram of a service system provided in an embodiment of the present invention, and the structure shown in fig. 1a is only one example of a service system to which the technical solution of the present invention can be applied, and the service system can be used to execute the processing flows shown in fig. 2 and fig. 3 described below. Fig. 1b is a scene schematic diagram of a service system provided in the embodiment of the present invention. As shown in fig. 1a and 1b, the service system includes: the device comprises a target style acquisition module, a probability density estimation module and a color matching module. The target style acquisition module acquires a target style data set consisting of a plurality of target style pictures; the probability density estimation module performs probability density estimation based on color features for content element parts in the target style picture, and specifically, may obtain a probability density function of the color features through the probability density estimation based on the color features, where the color features include: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched; and then, the color matching module forms a target scheme according to the probability density estimation result of the target style picture, and performs re-coloring on the content element part in the picture to be subjected to color matching to form a picture after color matching.
Specifically, the target style obtaining module may include: an analysis sub-module, a clustering sub-module, and a preference sub-module. The color matching module may include: the system comprises an acquisition submodule, a probability calculation submodule, a determination submodule, an attribute adjustment submodule and a color matching submodule. The color matching scheme of the picture and the formed target color matching scheme accord with the expected style of the user, have high practicability and effectiveness, simultaneously reduce the creation cost of the commercial picture, reduce the time spent by the user on content color matching, and greatly improve the creation efficiency of the commercial picture.
The above embodiments are illustrations of technical principles and exemplary application frameworks of the embodiments of the present invention, and specific technical solutions of the embodiments of the present invention are further described in detail below through a plurality of embodiments.
Example one
Fig. 2 is a flowchart of an embodiment of a color matching method for a picture provided by the present invention, and an execution subject of the method may be the business system. As shown in fig. 2, the color matching method for the picture includes the following steps:
S201, acquiring a target style data set.
In the embodiment of the invention, each picture is composed of a product part and a content element part, wherein the content element part is a background part or a text part.
When the color matching of the pictures is carried out, a target style data set expected by a user is selected from the sample data set, and the target style data set is composed of a plurality of target style pictures.
S202, carrying out probability density estimation based on color features on the content element part in the target style picture.
And after the target style data set is acquired, carrying out probability density estimation based on color features on the content element part in each target style picture.
Specifically, the color characteristics may include: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be matched with colors. The content element features are the average chromaticity value, the average saturation value or the average brightness value of all pixel points in the content element part of the target style picture. The difference contrast characteristic is the difference of the average value of the chroma, the average value of the saturation or the average value of the brightness of the content element part of the target style picture and the product part of the picture to be matched with colors.
And S203, forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be matched with colors.
in the embodiment of the invention, according to the probability density estimation result of the target style picture, the characteristic distribution of the product under the specific style can be obtained, and the content element part in the picture to be color-matched is recolorized according to the formed target color matching scheme. The recolored picture conforms to the expected style of the user, achieves the visual effect of capturing the eyeball, and has high practicability and effectiveness.
According to the color matching method of the picture, the probability density estimation based on the color characteristics is carried out on the target style picture, the color matching scheme suggestion under the style is provided for the user by combining the content elements and the color difference of the product part in the picture to be color matched, the picture after being recoloring accords with the expected style of the user, the practicability and the effectiveness are high, meanwhile, the creation cost of the commercial picture is reduced, the time spent by the user on content color matching is reduced, and the creation efficiency of the commercial picture is greatly improved.
example two
fig. 3 is a flowchart of another embodiment of a color matching method for a picture according to the present invention. As shown in fig. 3, on the basis of the embodiment shown in fig. 2, the color matching method for a picture provided in this embodiment may further include the following steps:
S301, analyzing the color style of the existing advertisement pictures in the sample data set, and respectively obtaining the style characteristics of the existing advertisement pictures.
In various embodiments of the present invention, the color scheme for pictures may be widely applied to color matching of advertisement pictures. Therefore, in the embodiment of the present invention, the picture subjected to color matching may be an advertisement picture, the target style picture in the obtained target style data set may be a target style advertisement picture, and the picture to be color-matched may be an advertisement picture to be color-matched. The content element part contained in the picture can be an advertisement element part contained in the advertisement picture; the content characteristics of the content element portion may be the advertising characteristics of the advertising element portion.
in an embodiment of the present invention, the sample data set may be preprocessed to facilitate a user to select a target style data set therefrom. Firstly, the color style of the existing advertisement pictures in the sample data set can be analyzed, so that the style characteristics of the existing advertisement pictures can be obtained.
The style characteristics of the existing advertisement pictures in the embodiment of the invention can comprise: a chrominance value, a saturation value, a luminance value, a visual difference value, a color patch region value, a chrominance difference value, a saturation difference value, and a luminance difference value.
specifically, the following operations may be performed for each existing advertisement picture:
1. And calculating the average value of the chromaticity of all pixel points in the existing advertisement picture.
2. and calculating the average value of the saturation of all pixel points in the existing advertisement picture. The saturation degree represents the degree of color approaching the spectral color, and the higher the saturation degree, the higher the degree of color approaching the spectral color, and the darker the color.
3. And calculating the brightness values of all pixel points in the existing advertisement picture. The brightness indicates the brightness of the color.
4. And clustering the pixels of the existing advertisement picture according to RGB (Red Green Blue) color values, and calculating the visual difference value of the existing advertisement picture according to the Euclidean distance of each clustering center in the RGB space and the proportion of the number of the pixels in each cluster to the total number of the pixels in the existing advertisement picture.
specifically, the visual difference value Dif may be calculated according to the following formula (1)rgb：
Wherein d isijIs the Euclidean distance, omega, between the ith and jth cluster centers in RGB spaceiand ωjThe number of the pixel points in the ith cluster and the jth cluster is respectively the proportion of the total number of the pixel points in the existing advertisement picture.
The difference between two colors is defined as their Euclidean distance in RGB space, and a larger value of difference indicates that the two colors have a larger contrast. The more obvious the color contrast is when an advertisement gives a person a strong visual difference. Therefore, in the embodiment of the present invention, ten RGB color values of an advertisement may be extracted by using a k-mean clustering algorithm, and then the visual difference value of the whole advertisement picture is calculated according to the above formula (1).
Here, if the k-mean clustering algorithm is too small, the degree of color difference between patterns is not obvious enough, and if the k-mean clustering algorithm is too large, the calculation amount is large, and the processing speed is affected, so that taking k to 10 is a better choice obtained by an experience and experiment.
5. The method comprises the steps of carrying out color division on an existing advertisement picture according to RGB color values to form a Region Adjacency Graph (RAG), merging nodes with edge weights smaller than a preset edge weight threshold value in the RAG, and calculating color block Region values according to the number of the merged nodes.
The number of color block areas in an advertisement can be used to represent the complexity of colors in the advertisement picture, and the more the number is, the more complicated the advertisement picture is. Therefore, in the embodiment of the present invention, firstly, a k-mean clustering algorithm may be adopted to perform color division on an advertisement picture, so as to construct RAG, each node in the graph represents an adjacent pixel point set with similar colors, the edge weight between the nodes represents the similarity degree of adjacent color block regions, and the larger the edge weight is, the larger the color difference between adjacent regions is. And combining the nodes with the edge weights smaller than the preset edge weight threshold value, and taking the final residual node number as the color block area value of the advertisement. When only one node remains in one advertisement picture, the colors adopted by the whole advertisement are very close, the more concise the colors are, and the more complicated the colors are otherwise.
6. Clustering the pixel points of the existing advertisement pictures according to HSV (Hue, Saturation, brightness) channel values, and calculating the channel difference Value of the existing advertisement pictures according to the Euclidean distance between each clustering center and the main clustering center with the largest area and the proportion of the number of the pixel points in each cluster to the total number of the pixel points of the existing advertisement pictures, wherein the HSV channel values are chroma values, Saturation values and brightness values.
Specifically, the difference value Dif of the α channel can be calculated according to the following formula (2)α：
Wherein, ciIs the ith cluster center, cpThe main cluster center with the largest area, d (c)p,ci) Is ciAnd cpeuclidean distance of, ωiAnd ωpThe number of the pixel points in the ith cluster and the p-th cluster is respectively the proportion of the total number of the pixel points in the existing advertisement picture.
In the embodiment of the present invention, the relative difference values on the three channels (H channel, S channel, V channel) are calculated according to the above equation (2), respectively.
in the embodiment of the present invention, the operations related to obtaining the style characteristics of the existing advertisement pictures are not sequential, and may be performed simultaneously, or may be performed sequentially according to different orders. Since the objective of the color style analysis is to perform feature cluster division on the advertisement pieces in the sample data set subsequently, each style feature may be normalized to range from 0 to 1. Through the analysis of the color style of the advertisement, a sample data set with high convergence can be obtained, a basis is provided for the color matching of the advertisement, and the effectiveness and the practicability of the scheme are ensured.
And S302, performing feature clustering processing on the sample data set according to the style features of the existing advertisement pictures.
In the embodiment of the present invention, the feature clustering process is performed on the sample data set, and specifically, the style features of the existing advertisement pictures are combined into an eight-dimensional feature vector; then, the eight-dimensional feature vector is reduced to a two-dimensional feature vector by a dimension reduction algorithm.
Specifically, in the embodiment of the present invention, a t-distributed stored neighboring embedding (t-SNE) algorithm may be adopted to reduce data obtained in the color style analysis from an eight-dimensional feature vector to a two-dimensional feature vector, so as to implement data visualization in a two-dimensional planar space. In the two-dimensional plane space, the distance between the two instances (advertisement pictures) reflects the similarity degree of the color style between the two instances, and it can be seen that the similar advertisement pictures have the similar color style.
And S303, selecting a target style data set from the sample data set after the characteristic clustering processing according to the target style selection instruction.
in the embodiment of the invention, a user can select advertisement pictures with a desired style through a visual two-dimensional plane, for example, the advertisement pictures can be in a picture frame form, and a plurality of advertisement pictures at wide positions in the visual plane are used as a target style data set to provide training data for subsequent probability density estimation based on color features.
s304, aiming at the advertisement element part in the target style advertisement picture, probability density estimation based on color characteristics is carried out.
In the embodiment of the present invention, for the advertisement element part (background or text) in the target style advertisement picture, first, a certain color characteristic x thereof may be calculatedeObtaining a series of discrete data setsthese discrete data can then be converted into a continuous probability Density function according to a Kernel Density Estimation (KDE) algorithm.
Specifically, the color characteristic x may be calculated according to the following formula (3)ethe probability density function of (f), (x):
Wherein K is a kernel function, n is the number of target style advertisement pictures in the target style data set, h is a bandwidth parameter of the kernel function K,And the color characteristics of the advertisement part in the ith target style advertisement picture are obtained. In general, h is required to approach 0 as n approaches infinity, since a larger n ensures that there are enough sample points for calculating the probability density even within a smaller interval h. In embodiments of the present invention, cross-validation may be employed to find the most appropriateThe size of the bandwidth.
S305, acquiring a plurality of color schemes input by a user.
S306, calculating probability values of the color features in the color schemes in the probability density function f (x) respectively.
in the embodiment of the invention, the user can input some existing color schemes for the advertisement pictures to be color-matched. Probability values for each color feature in each color scheme are calculated by the probability density function.
And S307, determining the color scheme with the highest probability as the target scheme.
In the embodiment of the invention, the probability values of a certain color feature in all schemes can be compared, and then the color scheme with the highest probability is determined as the target scheme to be recommended to the user.
Of course, a preset probability threshold may also be set, and all color schemes with probabilities higher than the preset probability threshold are determined as the target schemes recommended to the user. Specifically, the target color attributes of the target scheme include: a target chrominance value, a target luminance value, and a target saturation value.
s308, clustering the pixel points of the advertisement element part in the advertisement picture to be matched with colors according to the color attributes, and adjusting the color attributes of the pixel points of the advertisement element part in the advertisement picture to be matched with colors according to the color attribute of the main clustering center with the largest area and the target color attribute.
in order to maintain the quality of the advertisement picture and avoid color distortion, in the embodiment of the present invention, it may be assumed that the pixel closest to the main attribute (chrominance, luminance, saturation) (having the smallest attribute value difference) has the largest adjustment value, and the pixel farthest from the main attribute (having the largest attribute value difference) has the smallest adjustment value.
Specifically, the color attribute of the primary cluster center having the largest area may be determined as the primary attribute Cmain(ii) a Then, adjusting the color attribute of each pixel point of the advertisement element part in the advertisement picture to be matched with color according to the following formula (4):
Wherein, PiIs the original color attribute value of the ith pixel point, when P isi≤Cmainwhen D isboundaryIs a main attribute Cmainabsolute difference from the minimum of the color attribute, when Pi＞Cmainwhen D isboundaryis a main attribute CmainAbsolute difference from maximum of color attribute, δ l being main attribute CmainAnd the difference between the target color attribute and the target color attribute,and adjusting the attribute value of the ith pixel point.
In the embodiment of the present invention, since all the parameters are normalized, the minimum value of the color attribute is 0, and the maximum value is 1.
And S309, recoloring the advertisement element part in the advertisement picture to be matched with colors according to the adjusted color attribute.
According to the color matching method for the pictures, which is provided by the embodiment of the invention, the color style analysis is carried out on the existing advertisement pictures in the sample data set, the similarity degree of the styles among different advertisements is measured, and the probability density estimation based on the color characteristics is carried out on the target style advertisement pictures based on the specific styles, so that the advertisement color matching suggestions under the specific styles or scenes are provided for the user, and the user can conveniently select the favorite advertisement styles to generate an ideal advertisement color matching scheme. Great convenience is brought to the user group without the plane design experience; meanwhile, the creation cost of commercial advertisements is reduced, the time spent by the user on advertisement color matching is reduced, and the advertisement creation efficiency is greatly improved.
EXAMPLE III
fig. 4 is a schematic structural diagram of an embodiment of a color matching apparatus for pictures according to the present invention, which may be used to perform the steps of the method shown in fig. 2. As shown in fig. 4, the apparatus may include: a target style acquisition module 41, a probability density estimation module 42, and a color matching module 43.
The target style acquisition module 41 is configured to acquire a target style data set, where the target style data set is composed of a plurality of target style pictures;
The probability density estimation module 42 is configured to perform probability density estimation based on color features for the content element part in the target style picture, where the color features include: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched; the color matching module 43 is configured to form a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element portion in the picture to be color-matched.
In an embodiment of the present invention, the various pictures are composed of a product part and a content element part, wherein the content element part is a background part or a text part. When performing color matching on pictures, the target style acquisition module 41 selects a target style data set desired by a user from the sample data set, where the target style data set is composed of a plurality of target style pictures.
After the target style acquisition module 41 acquires the target style data set, the probability density estimation module 42 performs probability density estimation based on color features for the content element part in each target style picture. The color characteristics include: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be matched with colors. The content element features are the average chromaticity value, the average saturation value or the average brightness value of all pixel points in the content element part of the target style picture. The difference contrast characteristic is the difference of the average value of the chroma, the average value of the saturation or the average value of the brightness of the content element part of the target style picture and the product part of the picture to be matched with colors.
then, the color matching module 43 obtains the feature distribution of the product in the specific style according to the calculated probability density estimation result of the target style picture, forms a target color matching scheme, and recoloring the content element part in the picture to be color matched. The recolored picture conforms to the expected style of the user, achieves the visual effect of capturing the eyeball, and has high practicability and effectiveness.
according to the color matching device for the picture, provided by the embodiment of the invention, the probability density estimation based on the color characteristics is carried out on the target style picture, the color matching scheme suggestion under the style is provided for the user by combining the content elements and the color difference of the product part in the picture to be color matched, the picture after being recoloring accords with the expected style of the user, the practicability and the effectiveness are very high, meanwhile, the creation cost of the commercial picture is reduced, the time spent by the user on content color matching is reduced, and the creation efficiency of the commercial picture is greatly improved.
Example four
fig. 5a is a schematic structural diagram of another embodiment of a color matching apparatus for pictures according to the present invention, which can be used to perform the steps of the method shown in fig. 3. As shown in fig. 5, on the basis of the embodiment shown in fig. 4, the target style obtaining module 41 may include: an analysis sub-module 411, a clustering sub-module 412, and a preference sub-module 413.
The analysis submodule 411 may be configured to perform color style analysis on existing advertisement pictures in the sample data set, and respectively obtain style characteristics of the existing advertisement pictures, where the style characteristics include: the color value, the saturation value, the brightness value, the visual difference value, the color block region value, the chromaticity difference value, the saturation difference value and the brightness difference value;
the clustering submodule 412 may be configured to perform feature clustering on the sample data set according to style features of existing advertisement pictures;
the preference submodule 413 is configured to select a target style data set from the sample data set after the feature clustering process according to the target style selection instruction.
In various embodiments of the present invention, the color scheme for pictures may be widely applied to color matching of advertisement pictures. Therefore, in the embodiment of the present invention, the picture subjected to color matching may be an advertisement picture, the target style picture in the obtained target style data set may be a target style advertisement picture, and the picture to be color-matched may be an advertisement picture to be color-matched. The content element part contained in the picture can be an advertisement element part contained in the advertisement picture; the content characteristics of the content element portion may be the advertising characteristics of the advertising element portion.
in an embodiment of the present invention, the sample data set may be preprocessed to facilitate a user to select a target style data set therefrom. First, the analysis submodule 411 may perform a color style analysis on the existing advertisement pictures in the sample data set, so as to obtain style characteristics of the existing advertisement pictures.
Fig. 5b is a second schematic structural diagram of another embodiment of the color matching apparatus for pictures provided by the present invention. As shown in fig. 5b, in the apparatus for matching color of a picture provided by the embodiment of the present invention, the analysis sub-module 411 may include: a first calculation unit 4111, a second calculation unit 4112, a third calculation unit 4113, and a fourth calculation unit 4114.
The first calculating unit 4111 may be configured to calculate a chrominance average value, a saturation average value, and a luminance average value of all pixel points in an existing advertisement picture. The saturation degree represents the degree of color approaching the spectral color, and the higher the saturation degree, the higher the degree of color approaching the spectral color, and the darker the color. The brightness indicates the brightness of the color.
The second calculating unit 4112 may be configured to cluster the pixel points of the existing advertisement picture according to RGB color values, and calculate the visual difference value of the existing advertisement picture according to the euclidean distance of each cluster center in the RGB space and the ratio of the number of the pixel points in each cluster to the total number of the pixel points in the existing advertisement picture.
Specifically, in the embodiment of the present invention, the second calculating unit 4112 may calculate the visual difference value Dif according to the following formula (5)rgb：
Wherein d isijIs the Euclidean distance, omega, between the ith and jth cluster centers in RGB spaceiAnd ωjthe number of the pixel points in the ith cluster and the jth cluster is respectively the proportion of the total number of the pixel points in the existing advertisement picture.
The difference between two colors is defined as their Euclidean distance in RGB space, and a larger value of difference indicates that the two colors have a larger contrast. The more obvious the color contrast is when an advertisement gives a person a strong visual difference. Therefore, in the embodiment of the present invention, the second calculating unit 4112 may adopt a k-mean clustering algorithm to extract ten RGB color values of the advertisement, and then calculate the visual difference value of the whole advertisement picture according to the above formula (5). Here, if the k-mean clustering algorithm is too small, the degree of color difference between patterns is not obvious enough, and if the k-mean clustering algorithm is too large, the calculation amount is large, and the processing speed is affected, so that taking k to 10 is a better choice obtained by an experience and experiment.
the third calculating unit 4113 may be configured to perform color division on an existing advertisement picture according to RGB color values to form RAG, merge nodes in the RAG whose edge weights are smaller than a preset edge weight threshold, and calculate a color block region value according to the number of the merged nodes.
The number of color block areas in an advertisement can be used to represent the complexity of colors in the advertisement picture, and the more the number is, the more complicated the advertisement picture is. Therefore, in the embodiment of the present invention, the third calculating unit 4113 may use a k-mean clustering algorithm to perform color division on the advertisement picture, so as to construct RAG, each node in the graph represents a set of adjacent pixel points with similar colors, the edge weight between nodes represents the similarity of adjacent color block regions, and the larger the edge weight is, the larger the color difference between adjacent regions is. And combining the nodes with the edge weights smaller than the preset edge weight threshold value, and taking the final residual node number as the color block area value of the advertisement. When only one node remains in one advertisement picture, the colors adopted by the whole advertisement are very close, the more concise the colors are, and the more complicated the colors are otherwise.
The fourth calculating unit 4114 may be configured to cluster pixel points of an existing advertisement picture according to HSV channel values, and calculate a channel difference value of the existing advertisement picture according to an euclidean distance between each cluster center and a main cluster center having a largest area and a ratio of the number of the pixel points in each cluster to the total number of the pixel points of the existing advertisement picture, where the HSV channel values are a chrominance value, a saturation value, and a luminance value.
In particular, in the embodiments of the present inventionIn (3), the fourth calculating unit 4114 may calculate the difference value Dif of the α -channel according to the following formula (6)α：
wherein, ciis the ith cluster center, cpthe main cluster center with the largest area, d (c)p,ci) Is ciand cpEuclidean distance of, ωiand ωpThe number of the pixel points in the ith cluster and the p-th cluster is respectively the proportion of the total number of the pixel points in the existing advertisement picture.
In the embodiment of the present invention, the fourth calculating unit 4114 calculates relative differences on the three channels (H channel, S channel, V channel) according to the above formula (6).
The operation related to obtaining the style characteristics of the existing advertisement pictures in the embodiment of the invention is not sequential, can be carried out simultaneously, and can also be carried out sequentially according to different sequences.
Since the objective of the color style analysis is to perform feature cluster division on the advertisement pieces in the sample data set, the analysis sub-module 411 may perform normalization processing on each style feature, so that the range of the style feature is between 0 and 1. Through the analysis of the color style of the advertisement, a sample data set with high convergence can be obtained, a basis is provided for the color matching of the advertisement, and the effectiveness and the practicability of the scheme are ensured.
In addition, in the color matching apparatus for pictures provided in the embodiment of the present invention, the clustering sub-module 412 may include: a vector forming unit 4121 and a dimensionality reduction unit 4122.
the vector forming unit 4121 may be configured to form the style features of the existing advertisement pictures into an eight-dimensional feature vector;
The dimension reduction unit 4122 may be configured to reduce the eight-dimensional feature vector formed by the vector formation unit 4121 into a two-dimensional feature vector by a dimension reduction algorithm.
Specifically, in the embodiment of the present invention, the dimension reduction unit 4122 may reduce the eight-dimensional feature vector formed by the vector forming unit 4121 to a two-dimensional feature vector by using a t-distributed neighboring embedding (t-SNE) algorithm, so as to implement data visualization in a two-dimensional plane space.
In the two-dimensional plane space, the distance between the two instances (advertisement pictures) reflects the similarity degree of the color style between the two instances, and it can be seen that the similar advertisement pictures have the similar color style. The user can select the advertisement pictures with the expected style through the visualized two-dimensional plane, for example, the advertisement pictures can be in the form of picture frames, and a plurality of advertisement pictures are taken as the target style data sets at wide positions in the visualized plane. The optimization submodule 413 selects a target style data set from the sample data set after the feature clustering processing according to the target style selection instruction, and provides training data for subsequent probability density estimation based on the color features.
Secondly, in the picture color matching apparatus provided by the embodiment of the present invention, the probability density estimation module 42 may also be configured to calculate a certain color feature x of an advertisement element part (background or text) in the target style advertisement picture according to the advertisement element parteObtaining a series of discrete data setsThese discrete data are then converted to a continuous probability density function according to the KDE algorithm.
Specifically, the probability density estimation module 42 may calculate the color feature x according to the following formula (7)ethe probability density function of (f), (x):
Wherein K is a kernel function, n is the number of target style advertisement pictures in the target style data set, h is a bandwidth parameter of the kernel function K,And the color characteristics of the advertisement part in the ith target style advertisement picture are obtained.
in general, h is required to approach 0 as n approaches infinity, since a larger n ensures that there are enough sample points for calculating the probability density even within a smaller interval h. In embodiments of the present invention, cross-validation may be employed to find the most appropriate bandwidth size.
Again, in the apparatus for matching colors of pictures provided in the embodiment of the present invention, the color matching module 43 may further include: an acquisition submodule 431, a probability calculation submodule 432, a determination submodule 433, an attribute adjustment submodule 434 and a color matching submodule 435.
The obtaining submodule 431 may be configured to obtain a plurality of color schemes input by a user;
The probability calculation submodule 432 may be configured to calculate probability values of the color features in the color schemes in the probability density function f (x), respectively;
The determination submodule 433 may be configured to determine the color scheme with the highest probability as the target scheme, where the target color attributes of the target scheme include: a target colorimetric value, a target luminance value, and a target saturation value;
The attribute adjusting submodule 434 may be configured to cluster the pixel points of the advertisement element portion in the advertisement picture to be matched with color according to the color attribute, and adjust the color attribute of each pixel point of the advertisement element portion in the advertisement picture to be matched with color according to the color attribute of the main cluster center having the largest area and the target color attribute;
The color matching sub-module 435 may be configured to recolor the advertisement element portion of the advertisement picture to be color matched according to the adjusted color attribute.
in the embodiment of the invention, the user can input some existing color schemes for the advertisement pictures to be color-matched. After the obtaining submodule 431 obtains a plurality of color schemes input by the user, the probability calculating submodule 432 calculates probability values about the color features in the color schemes through the probability density function. The determination submodule 433 may compare the probability values of the color schemes for a certain color feature, and then determine the color scheme with the highest probability as the target scheme to be recommended to the user.
Of course, a preset probability threshold may also be set in advance, and the determining sub-module 433 may determine all color schemes with probabilities higher than the preset probability threshold as the target schemes recommended to the user.
in order to maintain the quality of the advertisement picture and avoid color distortion, in the embodiment of the present invention, it may be assumed that the pixel closest to the main attribute (chrominance, luminance, saturation) (having the smallest attribute value difference) has the largest adjustment value, and the pixel farthest from the main attribute (having the largest attribute value difference) has the smallest adjustment value.
thus, the attribute adjustment submodule 434 may also be used to determine the color attribute of the primary cluster center with the largest area as the primary attribute CmainThen, adjusting the color attribute of each pixel point of the advertisement element part in the advertisement picture to be matched with color according to the following formula (8):
Wherein, PiIs the original color attribute value of the ith pixel point, when P isi≤CmainWhen D isboundaryis a main attribute CmainAbsolute difference from the minimum of the color attribute, when Pi＞Cmainwhen D isboundaryIs a main attribute Cmainabsolute difference from maximum of color attribute, δ l being main attribute CmainAnd the difference between the target color attribute and the target color attribute,And adjusting the attribute value of the ith pixel point.
in the embodiment of the present invention, since all the parameters are normalized, the minimum value of the color attribute is 0, and the maximum value is 1.
According to the color matching device for the pictures, which is provided by the embodiment of the invention, the color style analysis is carried out on the existing advertisement pictures in the sample data set, the similarity degree of the styles among different advertisements is measured, the probability density estimation based on the color characteristics is carried out on the target style advertisement pictures based on the specific styles, advertisement color matching suggestions under the specific styles or scenes are provided for the user, and the user can conveniently select the favorite advertisement styles to generate an ideal advertisement color matching scheme. Great convenience is brought to the user group without the plane design experience; meanwhile, the creation cost of commercial advertisements is reduced, the time spent by the user on advertisement color matching is reduced, and the advertisement creation efficiency is greatly improved.
EXAMPLE five
the internal functions and structure of the color matching apparatus for pictures, which can be implemented as an electronic device, are described above. Fig. 6 is a schematic structural diagram of an embodiment of an electronic device provided in the present invention. As shown in fig. 6, the electronic device includes a memory 61 and a processor 62.
And a memory 61 for storing programs. In addition to the above-described programs, the memory 61 may also be configured to store other various data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on the electronic device, contact data, phonebook data, messages, pictures, videos, and so forth.
The memory 61 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
A processor 62, coupled to the memory 61, that executes programs stored by the memory 61 for:
Acquiring a target style data set, wherein the target style data set consists of a plurality of target style pictures;
performing probability density estimation based on color features for a content element part in a target style picture, wherein the color features comprise: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched;
And forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be matched with colors.
Further, as shown in fig. 6, the electronic device may further include: communication components 63, power components 64, audio components 65, a display 66, and other components. Only some of the components are schematically shown in fig. 6, and the electronic device is not meant to include only the components shown in fig. 6.
The communication component 63 is configured to facilitate wired or wireless communication between the electronic device and other devices. The electronic device may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 63 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 63 further comprises a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
A power supply component 64 provides power to the various components of the electronic device. The power components 64 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for an electronic device.
The audio component 65 is configured to output and/or input an audio signal. For example, the audio assembly 65 includes a Microphone (MIC) configured to receive external audio signals when the electronic device is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may further be stored in the memory 61 or transmitted via the communication component 63. In some embodiments, audio assembly 65 also includes a speaker for outputting audio signals.
the display 66 includes a screen, which may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.
those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (12)
1. A method of color matching a picture, the picture comprising a product portion and a content element portion, the method comprising:
Acquiring a target style data set, wherein the target style data set consists of a plurality of target style pictures;
Performing probability density estimation based on color features for content element parts in the target style picture, wherein the color features comprise: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched;
And forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be subjected to color matching.
2. The method for matching colors in a picture according to claim 1, wherein the content element portion is a background portion or a text portion.
3. the method of claim 1, wherein the content element characteristic is a mean chromaticity value, a mean saturation value, or a mean luminance value of all pixel points in the content element portion of the target style picture.
4. the method of claim 1, wherein the contrast characteristic is a difference between a chrominance average, a saturation average, or a luminance average of the content element portion of the target style picture and the product portion of the picture to be matched.
5. the method for color matching of pictures according to any of claims 1 to 4, wherein said obtaining a target style data set comprises:
Analyzing the color style of the existing pictures in the sample data set, and respectively acquiring the style characteristics of the existing pictures, wherein the style characteristics comprise: the color value, the saturation value, the brightness value, the visual difference value, the color block region value, the chromaticity difference value, the saturation difference value and the brightness difference value;
Performing feature clustering processing on the sample data set according to the style features of the existing pictures;
And selecting a target style data set in the sample data set after the characteristic clustering processing according to the target style selection instruction.
6. The method for matching colors of pictures according to claim 5, wherein the analyzing the color style of the existing pictures in the sample data set to obtain the style characteristics of the existing pictures respectively comprises: for each of the existing pictures that has been taken,
calculating the average value of the chroma, the average value of the saturation and the average value of the brightness of all pixel points in the existing picture;
Clustering the pixel points of the existing picture according to RGB color values, and calculating the visual difference value of the existing picture according to Euclidean distance of each clustering center in RGB space and the proportion of the number of the pixel points in each cluster to the total number of the pixel points in the existing picture;
carrying out color division on the existing picture according to RGB color values to form a region adjacency graph, merging nodes with the edge weights smaller than a preset edge weight threshold value in the region adjacency graph, and calculating the color block region value according to the number of the merged nodes;
clustering the pixel points of the existing picture according to HSV channel values, and calculating the channel difference value of the existing picture according to the Euclidean distance between each clustering center and the main clustering center with the largest area and the proportion of the number of the pixel points in each clustering to the total number of the pixel points of the existing picture, wherein the HSV channel values are chromatic values, saturation values and brightness values.
7. The method for matching colors of pictures according to claim 5, wherein the performing of feature clustering processing on the sample data set according to the style features of the existing pictures comprises:
Forming the style characteristics of the sample existing pictures into an eight-dimensional characteristic vector;
And reducing the eight-dimensional feature vector into a two-dimensional feature vector through a dimension reduction algorithm.
8. the method for matching colors of a picture according to claim 1,
The estimating the probability density based on the color features for the content element part in the target style picture comprises the following steps:
Carrying out probability density estimation based on color features aiming at the content element part in the target style picture to obtain a probability density function of the color features;
forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be color-matched, wherein the method comprises the following steps:
Acquiring a plurality of color schemes input by a user;
Respectively calculating probability values of the color features in the color schemes in the probability density function;
determining the color scheme with the highest probability as the target scheme, wherein the target color attribute of the target scheme comprises the following steps: a target colorimetric value, a target luminance value, and a target saturation value;
Clustering the pixel points of the content element part in the picture to be color-matched according to the color attribute, and adjusting the color attribute of each pixel point of the content element part in the picture to be color-matched according to the color attribute of the main clustering center with the largest area and the target color attribute;
And according to the adjusted color attributes, re-coloring the content element part in the picture to be subjected to color matching.
9. A color matching apparatus for a picture, said picture including a product portion and a content element portion, said apparatus comprising:
The target style acquisition module is used for acquiring a target style data set, and the target style data set consists of a plurality of target style pictures;
A probability density estimation module, configured to perform probability density estimation based on color features for a content element portion in the target style picture, where the color features include: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched;
And the color matching module is used for forming a target scheme according to the probability density estimation result of the target style picture and recoloring the content element part in the picture to be subjected to color matching.
10. The color matching apparatus for pictures according to claim 9, wherein said target style obtaining module comprises:
the analysis submodule is used for carrying out color style analysis on the existing pictures in the sample data set and respectively obtaining style characteristics of the existing pictures, and the style characteristics comprise: the color value, the saturation value, the brightness value, the visual difference value, the color block region value, the chromaticity difference value, the saturation difference value and the brightness difference value;
The clustering submodule is used for carrying out characteristic clustering processing on the sample data set according to the style characteristics of the existing pictures;
and the optimization submodule is used for selecting a target style data set in the sample data set after the characteristic clustering processing according to the target style selection instruction.
11. The color matching apparatus for pictures according to claim 9, wherein the color matching module comprises:
The obtaining submodule is used for obtaining a plurality of color schemes input by a user;
The probability calculation submodule is used for calculating the probability values of the color features in the color schemes in the probability density function respectively;
A determining submodule, configured to determine a color scheme with a highest probability as the target scheme, where a target color attribute of the target scheme includes: a target colorimetric value, a target luminance value, and a target saturation value;
the attribute adjusting submodule is used for clustering the pixel points of the content element part in the picture to be matched according to the color attributes, and adjusting the color attributes of all the pixel points of the content element part in the picture to be matched according to the color attribute of the main clustering center with the largest area and the target color attribute;
and the color matching submodule is used for recoloring the content element part in the picture to be subjected to color matching according to the adjusted color attribute.
12. An electronic device, comprising:
A memory for storing a program;
A processor for executing the program stored in the memory for:
acquiring a target style data set, wherein the target style data set consists of a plurality of target style pictures;
Performing probability density estimation based on color features for content element parts in the target style picture, wherein the color features comprise: the content element characteristics of the content element part and the difference contrast characteristics of the content element part and the product part in the picture to be color-matched;
and forming a target scheme according to the probability density estimation result of the target style picture, and recoloring the content element part in the picture to be subjected to color matching.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810595676.8A CN110580729B (en) | 2018-06-11 | 2018-06-11 | Image color matching method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810595676.8A CN110580729B (en) | 2018-06-11 | 2018-06-11 | Image color matching method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110580729A true CN110580729A (en) | 2019-12-17 |
| CN110580729B CN110580729B (en) | 2022-12-09 |
Family
ID=68809898
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810595676.8A Active CN110580729B (en) | 2018-06-11 | 2018-06-11 | Image color matching method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110580729B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2523164A1 (en) * | 2011-05-09 | 2012-11-14 | Olympus Corporation | Image processing apparatus, image processing method, and image processing program |
| CN106355607A (en) * | 2016-08-12 | 2017-01-25 | 辽宁工程技术大学 | Wide-baseline color image template matching method |
| CN106874924A (en) * | 2015-12-14 | 2017-06-20 | 阿里巴巴集团控股有限公司 | A kind of recognition methods of picture style and device |
-
2018
- 2018-06-11 CN CN201810595676.8A patent/CN110580729B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2523164A1 (en) * | 2011-05-09 | 2012-11-14 | Olympus Corporation | Image processing apparatus, image processing method, and image processing program |
| CN106874924A (en) * | 2015-12-14 | 2017-06-20 | 阿里巴巴集团控股有限公司 | A kind of recognition methods of picture style and device |
| CN106355607A (en) * | 2016-08-12 | 2017-01-25 | 辽宁工程技术大学 | Wide-baseline color image template matching method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110580729B (en) | 2022-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11425454B2 (en) | Dynamic video overlays | |
| CN109618173B (en) | Video compression method, device and computer readable storage medium | |
| CN107507144B (en) | Skin color enhancement processing method and device and image processing device | |
| US10289939B2 (en) | Image classification method and image classification apparatus | |
| CN107948733B (en) | Video image processing method and device and electronic equipment | |
| KR20070090224A (en) | Method of electronic color image saturation processing | |
| JP7136956B2 (en) | Image processing method and device, terminal and storage medium | |
| CN105981360A (en) | Image processing apparatus, image processing system, image processing method and recording medium | |
| US20220036536A1 (en) | Video quality assessing method and apparatus | |
| Wang et al. | No-reference stereoscopic image quality assessment using quaternion wavelet transform and heterogeneous ensemble learning | |
| CN110580729B (en) | Image color matching method and device and electronic equipment | |
| CN112565887A (en) | Video processing method, device, terminal and storage medium | |
| CN111798525A (en) | Image data processing method, device, equipment and storage medium, and color chart recommendation method, device and equipment | |
| EP3046071A1 (en) | Methods and apparatus for groupwise contrast enhancement | |
| CN112218006B (en) | Multimedia data processing method and device, electronic equipment and storage medium | |
| Al-Otum | A novel set of image morphological operators using a modified vector distance measure with color pixel classification | |
| CN113497954B (en) | Video toning method, device and storage medium | |
| KR101571440B1 (en) | Method of perceptual color assessment on display, recording medium and device for performing the method | |
| KR20220117057A (en) | Method and apparatus for video quality assessment according to the presence and absence of audio | |
| Hu et al. | General regression neural network utilized for color transformation between images on RGB color space | |
| JP2008294969A (en) | Video image conversion apparatus, method, and program | |
| Bruni et al. | Perceptual-based color quantization | |
| CN113436086B (en) | Processing method of non-uniform illumination video, electronic equipment and storage medium | |
| CN117082222B (en) | Image and video optimization toning method for rebroadcasting vehicle | |
| JP5069599B2 (en) | Video conversion apparatus, video conversion method, and video conversion program |
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 |