CN111461248A - Photographic composition line matching method, device, equipment and storage medium - Google Patents
Photographic composition line matching method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111461248A CN111461248A CN202010274362.5A CN202010274362A CN111461248A CN 111461248 A CN111461248 A CN 111461248A CN 202010274362 A CN202010274362 A CN 202010274362A CN 111461248 A CN111461248 A CN 111461248A
- Authority
- CN
- China
- Prior art keywords
- target image
- sample set
- composition
- training
- unit
- 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 16
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 10
- 238000010276 construction Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 2
- 230000008676 import Effects 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000000875 corresponding effect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/55—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a device, equipment and a storage medium for matching a photographic composition line, and mainly relates to the technical field of computer information. The method comprises the steps of importing a classification sample set, wherein the classification sample set is a photo collection set which is classified by taking a manually calibrated construction line as a label; training the classified sample set through a convolutional neural network DenseNet, and constructing a composition model; acquiring a target image, wherein the target image is a shot picture; and identifying and classifying the target image by using the composition model to obtain a composition line of a corresponding category. The invention has the beneficial effects that: the intelligent shooting guidance system has the advantages that the network is improved and then is divided into multiple structural line types, and intelligent shooting guidance is performed.
Description
Technical Field
The invention relates to the technical field of computer information, in particular to a method, a device, equipment and a storage medium for matching a photographic composition line.
Background
Image recognition and object detection are important research problems in the field of computer vision, and have wide application prospects in various aspects such as face recognition, safety monitoring, dynamic tracking and the like. Image recognition refers to a technique of processing, analyzing, and understanding an image with a computer to recognize various different patterns of objects and objects. The object detection means that for any frame or continuous frame of images, a specific object is detected and identified, and the position and size information of the object is returned, for example, a bounding box surrounding the object is output.
With the continuous development and progress of smart phones in recent years, the functions of cameras carried by the smart phones are gradually increased. However, most of mobile phones are almost zero for guidance before photographing, and mainly after photographing is finished, great attention is paid to processing photos. There is no guidance as to how to make the user eject a more aesthetic photograph.
Current difficulty of guidance prior to taking photographs: 1) mainly has strong specialization, and people without related professional knowledge have difficulty in having some abstract understanding on the photos. 2) A huge correlated data set is required for training. 3) Before the neural network is not rolled, image processing is always processed by opencv and the like, and the image processing is difficult to realize due to large calculation amount. The guidance before photographing needs to be supported by a relatively professional knowledge theory. No such convolutional neural network can currently implement it.
Disclosure of Invention
The invention aims to provide a method, a device, equipment and a storage medium for matching photographic composition lines, which have the function of intelligently guiding shooting by improving a network and then dividing the network into a plurality of structural line types.
In order to achieve the purpose, the invention is realized by the following technical scheme:
a photographic composition line matching method, comprising:
importing a classification sample set, wherein the classification sample set is a photo collection which is classified by taking a manually calibrated construction line as a label;
training the classified sample set through a convolutional neural network DenseNet, and constructing a composition model;
acquiring a target image, wherein the target image is a shot picture;
and identifying and classifying the target image by using the composition model to obtain a composition line of a corresponding category.
The training of the classification sample set by the convolutional neural network DenseNet includes training with a high performance processing unit using the improved vgg under the PyTorch framework. After the model training is finished, the model is laid out on an Ali server by using a model service technology to perform running calculation.
As another aspect of the present invention, there is provided a photographic composition line matching apparatus including:
the system comprises an importing unit, a classifying unit and a processing unit, wherein the importing unit is configured to import a classified sample set, and the classified sample set is a photo collection set which is classified by taking a manually calibrated construction line as a label;
the training unit is configured and used for training the classification sample set through a convolutional neural network DenseNet and constructing a composition model;
an acquisition unit configured to acquire a target image, the target image being a photographed photograph;
and the identification unit is configured for identifying and classifying the target image by using the composition model to obtain the composition line of the corresponding category.
The acquisition unit and the recognition unit are configured in an APP of the voice communication device.
The acquiring of the target image includes:
the photo is shot through the voice communication device where the APP is located.
As another aspect of the present invention, there is provided an apparatus comprising:
one or more processors;
a memory for storing one or more programs,
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform a photographic composition line matching method as described above.
As another aspect of the present invention, there is provided a computer-readable storage medium storing a computer program which, when executed by a processor, implements a photographic composition line matching method as described above.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention applies a deep learning frame to identify and classify the structure lines for the first time, and the structure lines are classified into various structure line types after the network is improved, so that the accuracy is as high as 80 percent
2. The invention can be used in the photography industry to intelligently guide the photography.
Drawings
FIG. 1 is an exemplary diagram of step one in example 1 of the present invention.
FIG. 2 is a schematic diagram of step two of example 1 of the present invention.
Detailed Description
The invention will be further illustrated with reference to the following specific examples. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and these equivalents also fall within the scope of the present application.
Before the neural network is not rolled, image processing is always processed by opencv and the like, and the image processing is difficult to realize due to large calculation amount. The guidance before photographing needs to be supported by a relatively professional knowledge theory.
Example 1:
according to the reasons, picture recognition is slightly modified, a method for recognizing picture structure lines is adopted, pictures shot by a user are calculated, and then a closer composition classification is recommended.
The details of the three steps are as follows:
the method comprises the following steps: it is necessary for a professional to mark a large number of photos and divide the structure lines into a plurality of labels, and each type of structure line comprises 1 ten thousand photos with labels. The type is shown in figure 1:
step two: and modifying the network, and training by using the processed data set, as shown in the second figure.
After processing, the image data can directly enter a convolutional neural network for training. In terms of neural network selection, we used the DenseNet at the leading edge of the current comparison. DenseNet is a convolutional neural network with dense connections. In the network, any two layers have direct connection, that is, the input of each layer of the network is the union of the outputs of all the previous layers, and the feature map learned by the layer is directly transmitted to all the next layers as input. The network has fewer parameters than a conventional convolutional network because it does not need to relearn the redundant feature map. And improves the transfer of information and gradients in the network, which makes the network easier to train. The entire data was first partitioned into 70% data as the training set and 30% as the test data set. Training in a PyTorch framework with a high performance processing unit (GPU) using improved vgg. After the model training is finished, the model is distributed on an Ali server by using a model service technology to perform the running water operation, and the calculation time of each user hand drawing is about 0.7 second. And then, according to the result, matching the corresponding composition to recommend.
Step three: implementing interactive functionality on apps
After the trained model can identify the structure line of the user photo, a functional module is developed to identify the picture structure line and make a recommendation. Firstly, a user shoots on a mobile phone and clicks pictures to upload. Secondly, the uploaded pictures are processed in a plurality of steps and then input into an identification model, structure lines are drawn, and a structure line with the highest similarity is returned for recommendation.
Claims (7)
1. A photographic composition line matching method, comprising:
importing a classification sample set, wherein the classification sample set is a photo collection which is classified by taking a manually calibrated construction line as a label;
training the classified sample set through a convolutional neural network DenseNet, and constructing a composition model;
acquiring a target image, wherein the target image is a shot picture;
and identifying and classifying the target image by using the composition model to obtain a composition line of a corresponding category.
2. The method of claim 1, wherein said training the classified sample set through a convolutional neural network DenseNet comprises training with a high performance processing unit using improved vgg in a PyTorch framework. After the model training is finished, the model is laid out on an Ali server by using a model service technology to perform running calculation.
3. A photographic composition line matching apparatus, comprising:
the system comprises an importing unit, a classifying unit and a processing unit, wherein the importing unit is configured to import a classified sample set, and the classified sample set is a photo collection set which is classified by taking a manually calibrated construction line as a label;
the training unit is configured and used for training the classification sample set through a convolutional neural network DenseNet and constructing a composition model;
an acquisition unit configured to acquire a target image, the target image being a photographed photograph;
and the identification unit is configured for identifying and classifying the target image by using the composition model to obtain the composition line of the corresponding category.
4. The apparatus as claimed in claim 3, wherein the acquiring unit and the recognizing unit are disposed in APP of the voice communication apparatus.
5. The apparatus of claim 4, wherein said acquiring the target image comprises:
the photo is shot through the voice communication device where the APP is located.
6. An apparatus, characterized in that the apparatus comprises:
one or more processors;
a memory for storing one or more programs,
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method recited in claim 1 or 2.
7. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method of claim 1 or 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010274362.5A CN111461248A (en) | 2020-04-09 | 2020-04-09 | Photographic composition line matching method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010274362.5A CN111461248A (en) | 2020-04-09 | 2020-04-09 | Photographic composition line matching method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111461248A true CN111461248A (en) | 2020-07-28 |
Family
ID=71680606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010274362.5A Pending CN111461248A (en) | 2020-04-09 | 2020-04-09 | Photographic composition line matching method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111461248A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112528979A (en) * | 2021-02-10 | 2021-03-19 | 成都信息工程大学 | Transformer substation inspection robot obstacle distinguishing method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109040605A (en) * | 2018-11-05 | 2018-12-18 | 北京达佳互联信息技术有限公司 | Shoot bootstrap technique, device and mobile terminal and storage medium |
| CN109344715A (en) * | 2018-08-31 | 2019-02-15 | 北京达佳互联信息技术有限公司 | Intelligent composition control method, device, electronic equipment and storage medium |
| CN110889428A (en) * | 2019-10-21 | 2020-03-17 | 浙江大搜车软件技术有限公司 | Image recognition method and device, computer equipment and storage medium |
-
2020
- 2020-04-09 CN CN202010274362.5A patent/CN111461248A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344715A (en) * | 2018-08-31 | 2019-02-15 | 北京达佳互联信息技术有限公司 | Intelligent composition control method, device, electronic equipment and storage medium |
| CN109040605A (en) * | 2018-11-05 | 2018-12-18 | 北京达佳互联信息技术有限公司 | Shoot bootstrap technique, device and mobile terminal and storage medium |
| CN110889428A (en) * | 2019-10-21 | 2020-03-17 | 浙江大搜车软件技术有限公司 | Image recognition method and device, computer equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112528979A (en) * | 2021-02-10 | 2021-03-19 | 成都信息工程大学 | Transformer substation inspection robot obstacle distinguishing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111444908B (en) | Image recognition method, device, terminal and storage medium | |
| CN111062871B (en) | Image processing method and device, computer equipment and readable storage medium | |
| CN110659646A (en) | Automatic multitask certificate image processing method, device, equipment and readable storage medium | |
| CN110866471A (en) | Face image quality evaluation method and device, computer readable medium and communication terminal | |
| US20190205618A1 (en) | Method and apparatus for generating facial feature | |
| CN111445459B (en) | Image defect detection method and system based on depth twin network | |
| CN112287820A (en) | Face detection neural network, face detection neural network training method, face detection method and storage medium | |
| CN111340123A (en) | Image score label prediction method based on deep convolutional neural network | |
| CN103617432A (en) | Method and device for recognizing scenes | |
| CN107766864B (en) | Method and device for extracting features and method and device for object recognition | |
| CN112417947B (en) | Method and device for optimizing key point detection model and detecting face key points | |
| WO2022127814A1 (en) | Method and apparatus for detecting salient object in image, and device and storage medium | |
| CN113298018A (en) | False face video detection method and device based on optical flow field and facial muscle movement | |
| CN110363111B (en) | Face living body detection method, device and storage medium based on lens distortion principle | |
| CN113255501B (en) | Method, apparatus, medium and program product for generating form recognition model | |
| CN115410240A (en) | Intelligent face pockmark and color spot analysis method and device and storage medium | |
| CN111461248A (en) | Photographic composition line matching method, device, equipment and storage medium | |
| KR101334858B1 (en) | Automatic butterfly species identification system and method, and portable terminal having automatic butterfly species identification function using the same | |
| US10991085B2 (en) | Classifying panoramic images | |
| CN104166840A (en) | Focusing realization method based on video conference system | |
| CN111464743A (en) | Photographic composition matching method and system | |
| CN115410174B (en) | Two-stage vehicle insurance anti-fraud image acquisition quality inspection method, device and system | |
| CN114863224B (en) | Training method, image quality detection method, device and medium | |
| CN113065559B (en) | Image comparison method and device, electronic equipment and storage medium | |
| CN115909408A (en) | Pedestrian re-identification method and device based on Transformer network |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200728 |