CN114140694A - Aesthetic composition method for coupling individual aesthetics with photographic aesthetics - Google Patents
Aesthetic composition method for coupling individual aesthetics with photographic aesthetics Download PDFInfo
- Publication number
- CN114140694A CN114140694A CN202111484987.5A CN202111484987A CN114140694A CN 114140694 A CN114140694 A CN 114140694A CN 202111484987 A CN202111484987 A CN 202111484987A CN 114140694 A CN114140694 A CN 114140694A
- Authority
- CN
- China
- Prior art keywords
- image
- aesthetic
- composition
- photographic
- aesthetics
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000008878 coupling Effects 0.000 title claims abstract description 13
- 238000010168 coupling process Methods 0.000 title claims abstract description 13
- 238000005859 coupling reaction Methods 0.000 title claims abstract description 13
- 238000013135 deep learning Methods 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 230000006870 function Effects 0.000 claims description 16
- 230000004913 activation Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims 1
- 238000004458 analytical method Methods 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000000059 patterning Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- 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
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Studio Devices (AREA)
Abstract
The invention discloses an aesthetic composition method for coupling individual aesthetics and photographic aesthetics, which comprises the following steps: the aesthetic composition method of the shooting terminal comprises the following steps: step 1: when the shooting terminal enters a shooting mode, acquiring image information in a current view-finding frame; step 2: determining the composition type of the current shot object according to the image information, and selecting a shooting composition mode according to the composition type; and step 3: intelligently marking in a composition mode of a target image to ensure accurate scene selection of the image; and 4, step 4: the individual characteristics and the group characteristics of the shot objects are determined through the image characteristic identification module. According to the method, the analysis and calculation capabilities of complex data and hidden characteristics of a deep learning network are utilized from the aspects of automatic big data marking and image target detection, the big data sample image library is analyzed according to image characteristics, images are distinguished, and the accuracy and stability of image identification are guaranteed.
Description
Technical Field
The invention relates to the technical field of photography, in particular to an aesthetic composition method for coupling individual aesthetics with photography aesthetics.
Background
People have higher and higher living standard, work leisure, relatives and friends can travel, photos are always taken during the travel as a memorial idea, and people can remember the once good times by looking over the photos later. The picture to be taken is not only required to be adjusted in the settings of the focal length, the aperture and the like of the camera, but also the picture composition is important, but most people can take a good picture without knowing how to take the picture composition, and no equipment capable of assisting in taking the picture composition exists in the market at present. With the development of electronic technology, camera terminals are more and more commonly owned and used by people, people can conveniently use integrated camera terminals, such as mobile phones integrated with camera functions to take pictures or record video images, and with the frequent taking of pictures and the popularity of social networks, the aesthetic feeling of pictures is more and more emphasized by people.
The effect and quality of photos shot by the conventional photographing mobile terminal can hardly reach professional level, and even if some users have composition concepts, the users can only perform composition according to photographing experience, so that the composition of the photos can not be quickly and effectively photographed, and the photos with more reasonable layout can not be quickly and effectively photographed. Therefore, a new technical solution needs to be provided.
Disclosure of Invention
The invention aims to provide an aesthetic composition method for coupling individual aesthetics with photography aesthetics, and solves the problems that the effect and quality of photos shot by the conventional photographing mobile terminal are difficult to reach professional level, and even if some users have composition concepts, the composition can be only carried out according to the photographing experience, so that the composition for photographing can not be carried out quickly and effectively, and the quick and effective photographing is not facilitated to obtain the photos with more reasonable layout.
In order to achieve the purpose, the invention provides the following technical scheme: an aesthetic composition method for coupling between individual aesthetics and photographic aesthetics, comprising: the aesthetic composition method of the shooting terminal comprises the following steps:
step 1: when the shooting terminal enters a shooting mode, acquiring image information in a current view-finding frame;
step 2: determining the composition type of the current shot object according to the image information, and selecting a shooting composition mode according to the composition type;
and step 3: intelligently marking in a composition mode of a target image to ensure accurate scene selection of the image;
and 4, step 4: determining individual characteristics and group characteristics of the shot object through an image characteristic identification module;
and 5: and confirming the use habit of the user through the deep learning network, and outputting an image result after confirming the characteristic recognition result of the shot object so as to complete composition.
In a preferred embodiment of the present invention, the image feature recognition module in step 4 includes an individual feature and a group feature, the individual feature includes a physical feature unique to the photographic subject, and the group feature includes a physical feature common to the photographic subject.
In a preferred embodiment of the present invention, the deep learning network of step 5 includes a sample library, and at least 10 ten thousand frames of image blocks are contained in the sample library.
As a preferred embodiment of the present invention, the process of establishing the deep learning network and training the deep neural network in step 5 is a process of solving an activation function, a weight and an offset value of each hidden layer node, wherein the activation function is a function that is enhanced after summing each input, and typical activation functions include Sigmoid, tanh and ReLU, and each activation function is as follows:
tanh=tanh(x)
ReLU=max(0,x)。
as a preferred embodiment of the present invention, the intelligent marker adopts OpenCV-based image detection, which can detect most images, prepare for labeling, and perform histogram equalization processing to enhance the facial image characteristics, where the formula is as follows:
in a preferred embodiment of the present invention, H, W in the formula (1) represents the height and width of an image, histo (k) represents the number of pixels with a pixel gray value of k, histo in the formula (2) represents the histogram of an output image, hist represents the histogram of an input image, a balanced mapping relationship is established, the formula (3) represents a mathematical model of accumulating histograms, and a q value is obtained by conversion, the q value in the formula (4) represents a pixel mapping value after equalization, and image matrixing processing is performed, O in the formula 5 represents a pixel value, and I represents an input pixel value.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, the analysis and calculation capabilities of complex data and hidden characteristics of a deep learning network are utilized from the aspects of automatic big data marking and image target detection, the big data sample image library is analyzed according to image characteristics, images are distinguished, and the accuracy and stability of image identification are guaranteed; when a shooting terminal enters a shooting mode, image information in a current view-finding frame is obtained, the composition type of a current shot object is determined according to the image information, the shooting composition mode is selected according to the composition type, intelligent marking is carried out in the composition mode of a target image, accurate scene selection of the image is ensured, meanwhile, the individual characteristics and the group characteristics of a shot object are determined through an image characteristic identification module, the use habit of a user is confirmed through a deep learning network, and after the characteristic identification result of the shot object is confirmed, the image result is output, so that the composition is completed, the picture processing efficiency is improved, and the purpose of corresponding user experience is improved.
Drawings
FIG. 1 is a schematic diagram of a patterning process of the present invention;
FIG. 2 is a schematic diagram of an image feature recognition process according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-2, the present invention provides a technical solution: an aesthetic composition method for coupling between individual aesthetics and photographic aesthetics, comprising: the aesthetic composition method of the shooting terminal comprises the following steps:
step 1: when the shooting terminal enters a shooting mode, acquiring image information in a current view-finding frame;
step 2: determining the composition type of the current shot object according to the image information, and selecting a shooting composition mode according to the composition type;
and step 3: intelligently marking in a composition mode of a target image to ensure accurate scene selection of the image;
and 4, step 4: determining individual characteristics and group characteristics of the shot object through an image characteristic identification module;
and 5: and confirming the use habit of the user through the deep learning network, and outputting an image result after confirming the characteristic recognition result of the shot object so as to complete composition.
Further improved, as shown in fig. 2: the image feature recognition module of step 4 includes an individual feature including a physical feature unique to the photographic subject and a group feature including a physical feature common to the photographic subject.
In a further improvement, the deep learning network of step 5 includes a sample library, and at least 10 ten thousand frames of image blocks are included in the sample library.
Further improved, the process of establishing the deep learning network and training the deep neural network in step 5 is a process of solving an activation function, a weight and an offset value of each hidden layer node, wherein the activation function is a function that is enhanced after summing each input, and typical activation functions include Sigmoid, tanh and ReLU, and each activation function is as follows:
tanh=tanh(x)
ReLU=max(0,x)
the essence of the deep neural network is that the weights are multiplied by the input values plus the offset values. For solving the weight value and the deviant, the solution can be realized by combining back propagation and gradient descent, the weight value of each node is initialized by random numbers at first, and then the output value calculated by the deep neural network is compared with the real output value. If the comparison value is larger than the difference, modifying the weight of the current layer node. When the comparison value is not much different, the weight of the lower layer is modified. And pushing forward according to the rule, and gradually recommending the weight to the first layer.
Further improved, the intelligent mark adopts image detection based on OpenCV, which can detect most images, prepare for labeling, and perform histogram equalization processing to enhance the face image characteristics, wherein the formula is as follows:
as a preferred embodiment of the present invention, H, W in the formula (1) represents the height and width of an image, histo (k) represents the number of pixels of which the gray value of a pixel is k, histo in the formula (2) represents the histogram of an output image, hisi represents the histogram of an input image, a balanced mapping relationship is established, the formula (3) represents a mathematical model for accumulating histograms, a q value is obtained by conversion, q value in the formula (4) represents a pixel mapping value after equalization, image matrixing processing is performed, O in the formula 5 represents a pixel value, and I represents an input pixel value, a deep neural network frame is established and trained to form an identification mechanism, so as to achieve the purpose of accurately identifying a human face; meanwhile, automatic marking is established, so that automatic marking is achieved after data acquisition, and the efficiency and quality of data engineering are guaranteed.
According to the method, the analysis and calculation capabilities of complex data and hidden characteristics of a deep learning network are utilized from the aspects of automatic big data marking and image target detection, the big data sample image library is analyzed according to image characteristics, images are distinguished, and the accuracy and stability of image identification are guaranteed; when a shooting terminal enters a shooting mode, image information in a current view-finding frame is obtained, the composition type of a current shot object is determined according to the image information, the shooting composition mode is selected according to the composition type, intelligent marking is carried out in the composition mode of a target image, accurate scene selection of the image is ensured, meanwhile, the individual characteristics and the group characteristics of a shot object are determined through an image characteristic identification module, the use habit of a user is confirmed through a deep learning network, and after the characteristic identification result of the shot object is confirmed, the image result is output, so that the composition is completed, the picture processing efficiency is improved, and the purpose of corresponding user experience is improved.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. An aesthetic composition method for coupling between personal aesthetics and photographic aesthetics, comprising: the method comprises the following steps: the aesthetic composition method of the shooting terminal comprises the following steps:
step 1: when the shooting terminal enters a shooting mode, acquiring image information in a current view-finding frame;
step 2: determining the composition type of the current shot object according to the image information, and selecting a shooting composition mode according to the composition type;
and step 3: intelligently marking in a composition mode of a target image to ensure accurate scene selection of the image;
and 4, step 4: determining individual characteristics and group characteristics of the shot object through an image characteristic identification module;
and 5: and confirming the use habit of the user through the deep learning network, and outputting an image result after confirming the characteristic recognition result of the shot object so as to complete composition.
2. An aesthetic composition method for coupling aesthetic personal appearance with photographic aesthetic appearance according to claim 1, characterized in that: the image feature recognition module of step 4 includes an individual feature including a physical feature unique to the photographic subject and a group feature including a physical feature common to the photographic subject.
3. An aesthetic composition method for coupling aesthetic personal appearance with photographic aesthetic appearance according to claim 1, characterized in that: the deep learning network of the step 5 comprises a sample base, and at least 10 ten thousand frames of image blocks are contained in the sample base.
4. An aesthetic composition method for coupling aesthetic personal appearance with photographic aesthetic appearance according to claim 1, characterized in that: the process of establishing the deep learning network and training the deep neural network in the step 5 is a process of solving an activation function, a weight and an offset value of each hidden layer node, wherein the activation function is a function which is enhanced after summing all the inputs, typical activation functions include Sigmoid, tanh and ReLU, and all the activation functions are as follows:
tanh=tanh(x)
ReLU=max(0,x)。
5. an aesthetic composition method for coupling aesthetic personal appearance with photographic aesthetic appearance according to claim 1, characterized in that: the intelligent mark adopts image detection based on OpenCV, most images can be detected in the mode, preparation is made for marking, histogram equalization processing is carried out for enhancing the characteristics of the face image, and the formula is as follows:
6. an aesthetic composition method for coupling aesthetic personal appearance with photographic aesthetic appearance according to claim 5, characterized in that: h, W in the formula (1) represents the height and width of an image, histo (k) represents the number of pixels with the gray value of the pixel being k, histo in the formula (2) represents the histogram of an output image, hisi represents the histogram of an input image, an equalization mapping relation is established, the formula (3) represents a mathematical model for accumulating the histograms, conversion is carried out to obtain a q value, the q value in the formula (4) represents the pixel mapping value after equalization, image matrixing processing is carried out, O in the formula 5 represents the pixel value, and I represents the input pixel value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111484987.5A CN114140694A (en) | 2021-12-07 | 2021-12-07 | Aesthetic composition method for coupling individual aesthetics with photographic aesthetics |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111484987.5A CN114140694A (en) | 2021-12-07 | 2021-12-07 | Aesthetic composition method for coupling individual aesthetics with photographic aesthetics |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114140694A true CN114140694A (en) | 2022-03-04 |
Family
ID=80384724
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111484987.5A Pending CN114140694A (en) | 2021-12-07 | 2021-12-07 | Aesthetic composition method for coupling individual aesthetics with photographic aesthetics |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114140694A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101990068A (en) * | 2009-07-31 | 2011-03-23 | 卡西欧计算机株式会社 | Image processing device and method |
CN108513066A (en) * | 2018-03-28 | 2018-09-07 | 努比亚技术有限公司 | It takes pictures composition guidance method, mobile terminal and storage medium |
CN109767397A (en) * | 2019-01-09 | 2019-05-17 | 三星电子(中国)研发中心 | A kind of image optimization method and system based on artificial intelligence |
CN111757012A (en) * | 2020-07-16 | 2020-10-09 | 盐城工学院 | Image processing method based on combination of individual and photographic aesthetics |
CN113411511A (en) * | 2021-06-29 | 2021-09-17 | 中国科学院长春光学精密机械与物理研究所 | High frame frequency imaging system image preprocessing method based on histogram analysis |
-
2021
- 2021-12-07 CN CN202111484987.5A patent/CN114140694A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101990068A (en) * | 2009-07-31 | 2011-03-23 | 卡西欧计算机株式会社 | Image processing device and method |
CN108513066A (en) * | 2018-03-28 | 2018-09-07 | 努比亚技术有限公司 | It takes pictures composition guidance method, mobile terminal and storage medium |
CN109767397A (en) * | 2019-01-09 | 2019-05-17 | 三星电子(中国)研发中心 | A kind of image optimization method and system based on artificial intelligence |
CN111757012A (en) * | 2020-07-16 | 2020-10-09 | 盐城工学院 | Image processing method based on combination of individual and photographic aesthetics |
CN113411511A (en) * | 2021-06-29 | 2021-09-17 | 中国科学院长春光学精密机械与物理研究所 | High frame frequency imaging system image preprocessing method based on histogram analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Li et al. | Fast multi-exposure image fusion with median filter and recursive filter | |
KR101731771B1 (en) | Automated selection of keeper images from a burst photo captured set | |
CN110807759B (en) | Method and device for evaluating photo quality, electronic equipment and readable storage medium | |
US8212911B2 (en) | Imaging apparatus, imaging system, and imaging method displaying recommendation information | |
CN105243371B (en) | A kind of detection method, system and the camera terminal of face U.S. face degree | |
WO2019210555A1 (en) | People counting method and device based on deep neural network and storage medium | |
CN108234870A (en) | Image processing method, device, terminal and storage medium | |
CN108810406B (en) | Portrait light effect processing method, device, terminal and computer readable storage medium | |
CN108198177A (en) | Image acquiring method, device, terminal and storage medium | |
TWI532361B (en) | Automatic photographing method and system thereof | |
WO2022227393A1 (en) | Image photographing method and apparatus, electronic device, and computer readable storage medium | |
CN111182212B (en) | Image processing method, image processing apparatus, storage medium, and electronic device | |
US9881086B2 (en) | Image shooting device, image shooting method, and recording medium | |
CN107424117B (en) | Image beautifying method and device, computer readable storage medium and computer equipment | |
WO2016206344A1 (en) | White balance correction method, device and computer storage medium | |
WO2023020201A1 (en) | Image enhancement method and electronic device | |
WO2019090502A1 (en) | Intelligent terminal-based image capturing method and image capturing system | |
WO2015196681A1 (en) | Picture processing method and electronic device | |
CN114466133B (en) | Photographing method and device | |
CN109872340B (en) | Composition method, electronic device and computer readable storage medium | |
CN110110742B (en) | Multi-feature fusion method and device, electronic equipment and storage medium | |
WO2023071933A1 (en) | Camera photographing parameter adjustment method and apparatus and electronic device | |
CN114140694A (en) | Aesthetic composition method for coupling individual aesthetics with photographic aesthetics | |
CN111757012A (en) | Image processing method based on combination of individual and photographic aesthetics | |
CN110519509A (en) | Composition evaluation method, method for imaging, device, electronic equipment, storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220304 |
|
RJ01 | Rejection of invention patent application after publication |