CN111833355A - Method for scratching picture - Google Patents

Method for scratching picture Download PDF

Info

Publication number
CN111833355A
CN111833355A CN202010507559.9A CN202010507559A CN111833355A CN 111833355 A CN111833355 A CN 111833355A CN 202010507559 A CN202010507559 A CN 202010507559A CN 111833355 A CN111833355 A CN 111833355A
Authority
CN
China
Prior art keywords
picture
data
matting
outputting
matrix data
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
Application number
CN202010507559.9A
Other languages
Chinese (zh)
Inventor
甘凌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Yiqi Network Technology Co ltd
Original Assignee
Hangzhou Yiqi Network Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Yiqi Network Technology Co ltd filed Critical Hangzhou Yiqi Network Technology Co ltd
Priority to CN202010507559.9A priority Critical patent/CN111833355A/en
Publication of CN111833355A publication Critical patent/CN111833355A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T9/00—Image coding
    • G06T9/002—Image coding using neural networks
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention discloses a method for matting pictures, which comprises the following steps: the method comprises the steps of decoding and converting picture coded data into picture matrix data, enabling the picture matrix data to be in an opencv format, identifying the picture matrix data by using a convolutional neural network on a GPU, outputting a feature diagram, and finally matting the feature diagram on a CPU by using a Gaussian function and storing the feature diagram into the picture format. The method is used for identifying and scratching the picture, and has the advantages of quicker identification and more accurate scratching.

Description

Method for scratching picture
Technical Field
The invention relates to the field of picture matting, in particular to a picture matting method.
Background
Deep learning is a branch of machine learning, is an algorithm which takes an artificial neural network as a framework and performs characterization learning on data, and has the advantage that a strong deep network is used for autonomously learning image features to replace manually acquired features.
In the prior art, the time for identifying and scratching the picture is long and the accuracy is low by adopting an artificial mode, so that the picture is correspondingly operated by adopting an artificial intelligence method, and the picture is quicker and more accurate in identification and scratching.
Disclosure of Invention
The invention provides a method for matting pictures, which aims to solve the problems of long time and low accuracy rate in picture identification matting in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention discloses a method for matting pictures, which comprises the following steps:
acquiring picture coded data, decoding the picture coded data, and outputting picture matrix data;
processing the picture matrix data by using a convolutional neural network, and outputting a characteristic diagram;
and calculating the characteristic graph according to Gaussian filtering, and matting to obtain a picture.
Acquiring picture coding data.
Preferably, the obtaining of the web text, the calculating of the feature map according to gaussian filtering, and the matting to obtain the picture comprise:
according to a gaussian function:
Figure BDA0002527088010000021
the feature map is calculated where the parameters in the gaussian template are determined by the coordinates of (x, y) and σ represents the standard deviation.
Preferably, acquiring picture encoded data, decoding the picture encoded data, and outputting picture matrix data includes:
analyzing the picture coding data by utilizing a fromstering function, and outputting matrix data;
and decoding the matrix data by using an imdecode function, and outputting picture matrix data.
Preferably, the processing the picture matrix data by using a convolutional neural network to output a feature map includes:
and processing the picture matrix by using a prediction model, a convolutional layer and a pooling layer of the convolutional neural network, and outputting a characteristic diagram.
A device for matting pictures, comprising:
the decoding module is used for acquiring picture coded data, decoding the picture coded data and outputting picture matrix data;
the processing module is used for processing the picture matrix data by utilizing a convolutional neural network and outputting a characteristic diagram;
and the matting module calculates the characteristic graph according to Gaussian filtering and obtains a picture by matting.
Preferably, the matting module comprises:
a Gaussian function unit that, according to a Gaussian function:
Figure BDA0002527088010000022
the feature map is calculated where the parameters in the gaussian template are determined by the coordinates of (x, y) and σ represents the standard deviation.
Preferably, the decoding module includes:
a fromming function unit which analyzes the picture coding data by using a fromming function and outputs matrix data;
and the imdecode function unit is used for decoding the matrix data by utilizing the imdecode function and outputting the picture matrix data.
Preferably, the processing module includes:
and the characteristic diagram unit is used for processing the picture matrix by utilizing the prediction model, the convolutional layer and the pooling layer of the convolutional neural network and outputting a characteristic diagram. .
An electronic device comprising a memory and a processor, the memory for storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement a method of matting a picture as claimed in any one of the above.
A computer-readable storage medium having stored thereon a computer program for causing a computer to carry out a method of matting a picture as claimed in any one of the above.
The invention has the following beneficial effects:
the method comprises the steps of decoding and converting picture coded data into picture matrix data, enabling the picture matrix data to be in an opencv format, identifying the picture matrix data by using a convolutional neural network on a GPU, outputting a feature diagram, and finally matting the feature diagram on a CPU by using a Gaussian function and storing the feature diagram into the picture format. The method is used for identifying and scratching the picture, and has the advantages of quicker identification and more accurate scratching.
Drawings
FIG. 1 is a first flowchart of a method for matting pictures according to an embodiment of the present invention;
FIG. 2 is a second flowchart of a method for matting pictures according to an embodiment of the present invention;
FIG. 3 is a third flowchart of a method for matting a picture according to an embodiment of the present invention;
FIG. 4 is a fourth flowchart of a method for matting pictures according to an embodiment of the present invention;
fig. 5 is a flowchart of a specific implementation of a method for matting pictures according to an embodiment of the present invention.
FIG. 6 is a schematic diagram of an apparatus for matting pictures according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of an acquisition module of an apparatus for matting pictures according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a matching module of an apparatus for matting pictures according to an embodiment of the present invention;
FIG. 9 is a schematic diagram of an output module of an apparatus for matting pictures according to an embodiment of the present invention;
FIG. 10 is a flowchart of an embodiment of an apparatus for matting pictures according to an embodiment of the present invention;
fig. 11 is a schematic diagram of an electronic device implementing a method for matting pictures according to an embodiment of 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 accompanying drawings, 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.
Before the technical solution of the present invention is introduced, a scenario to which the technical solution of the present invention may be applicable is exemplarily described.
Example 1
As shown in fig. 1, a method for matting pictures includes the following steps:
s110, acquiring picture coded data, decoding the picture coded data, and outputting picture matrix data;
s120, processing the image matrix data by using a convolutional neural network, and outputting a characteristic diagram;
and S130, calculating the characteristic diagram according to Gaussian filtering, and matting to obtain a picture.
The method comprises the steps of decoding and converting picture coded data into picture matrix data, enabling the picture matrix data to be in an opencv format, identifying the picture matrix data by using a convolutional neural network on a GPU, outputting a feature diagram, and finally matting the feature diagram on a CPU by using a Gaussian function and storing the feature diagram into the picture format. The GPU is called Graphics processing Unit (vision processor) to be specially responsible for executing drawing work and parallel computing work, the CPU is called Central Processing Unit (CPU) to be used for explaining computer instructions and processing data in computer software, and the method is suitable for logic structure judgment and small-scale numerical calculation.
Example 2
As shown in fig. 2, a method for matting pictures includes:
s210, acquiring picture coded data, decoding the picture coded data, and outputting picture matrix data;
s220, processing the image matrix data by using a convolutional neural network, and outputting a characteristic diagram;
s230, according to a Gaussian function:
Figure BDA0002527088010000051
calculating the feature map, wherein parameters in the Gaussian template are determined by coordinates of (x, y), and sigma represents standard deviation;
and (2) calculating the characteristic graph according to a Gaussian function, wherein (X, y) in the method represents a Gaussian kernel, the parameter value is 3X 3, sigma represents the standard deviation of the Gaussian kernel function in the X direction, the parameter value is 0, and the picture is extracted and stored in formats of png, jpg, tif and the like. The image matting method based on the Gaussian function has the advantage of being accurate in matting.
Example 3
As shown in fig. 3, a method for matting pictures includes:
s310, acquiring picture coded data, decoding the picture coded data, and outputting picture matrix data;
s320, analyzing the picture coding data by utilizing a fromstering function, and outputting matrix data;
s330, decoding the matrix data by using an imdecode function, and outputting picture matrix data;
in embodiment 3, since data is decoded from a character string by using a fromstering function, picture coded data can be parsed into numpy matrix data, and since data is decoded into an image format by using an immecode function for restoring an image from network transmission data, matrix data can be decoded into picture matrix data. By the method, the picture can be acquired more quickly.
Example 4
As shown in fig. 4, a method for matting pictures includes:
s410, acquiring picture coded data, decoding the picture coded data, and outputting picture matrix data;
s420, processing the picture matrix by using a prediction model, a convolutional layer and a pooling layer of the convolutional neural network, and outputting a characteristic diagram;
the prediction model of the convolutional neural network in embodiment 4 is generated by training the labeled data through the convolutional neural network on the GPU according to eighty percent of training data and twenty percent of validation data, and storing the model for the result generated by each Epoch (one generation of training). And processing the picture matrix by using a convolutional neural network, and calculating to obtain a characteristic diagram.
Example 5
As shown in fig. 5, one specific embodiment may be:
s510, decoding the picture coded data and outputting picture matrix data;
firstly, decoding pictures uploaded by a user by using base64, then coding base64 into numpy matrix data by using a fromming function, and finally converting the numpy matrix data into opencv picture matrix data by using an imdecode function to complete the identification of the uploaded pictures.
S520, establishing a convolutional neural network prediction model;
the prediction model of the convolutional neural network is generated by training the labeled data through the convolutional neural network on the GPU according to eighty percent of training data and twenty percent of verification data, and storing a model for a result generated by each Epoch (generation training). The convolutional neural network comprises a prediction model, a convolutional layer and a pooling layer, and is deployed on a GPU server.
S530, the convolutional neural network processes the picture matrix through the prediction model, the convolutional layer and the pooling layer and outputs a characteristic diagram;
and the server transmits the picture to neural network reasoning on the GPU, and generates a characteristic diagram in a matrix format through convolutional neural network reasoning.
S540, calculating the characteristic graph by using a Gaussian function, and matting and storing the characteristic graph;
the characteristic graph is calculated according to a Gaussian function, in the method, (X, y) represents a Gaussian kernel, the parameter value is 3X 3, sigma represents the standard deviation of the Gaussian kernel function in the X direction, the parameter value is 0, a picture is obtained by scratching in a CPU and is stored in formats of png, jpg, tif and the like, then the picture in a memory is encrypted through base64 to generate JSON format data, and the picture scratching by utilizing the Gaussian function has the advantage of accurate scratching.
Example 6
As shown in fig. 6, a device for matting pictures comprises:
the decoding module 10 acquires picture encoded data, decodes the picture encoded data, and outputs picture matrix data;
the processing module 20 is used for processing the image matrix data by using a convolutional neural network and outputting a characteristic diagram;
and the matting module 30 calculates the characteristic graph according to Gaussian filtering and obtains a picture by matting.
One embodiment of the above apparatus may be: the decoding module 10 obtains the picture coded data, decodes the picture coded data, and outputs the picture matrix data, the processing module 20 processes the picture matrix data by using a convolutional neural network, and outputs a characteristic diagram, and finally, the matting module 30 calculates the characteristic diagram according to Gaussian filtering, and scratches to obtain a picture.
Example 7
As shown in fig. 7, a matting module 30 of an apparatus for matting a picture includes:
gaussian function section 32, according to the gaussian function:
Figure BDA0002527088010000081
the feature map is calculated where the parameters in the gaussian template are determined by the coordinates of (x, y) and σ represents the standard deviation.
One embodiment of the above described device's matting module 30 can be: gaussian function section 32, according to the gaussian function:
Figure BDA0002527088010000082
the feature map is calculated where the parameters in the gaussian template are determined by the coordinates of (x, y) and σ represents the standard deviation.
Example 8
As shown in fig. 8, a decoding module 10 of an apparatus for matting pictures includes:
a hashing function unit 12, which analyzes the picture coding data by using a hashing function and outputs matrix data;
and an imdecode function unit 14 for decoding the matrix data by using an imdecode function and outputting picture matrix data.
One embodiment of the decoding module 10 of the above apparatus may be: the fromming function unit 12 analyzes the picture coding data by using a fromming function, outputs matrix data, and then the codec function unit 14 decodes the matrix data by using the codec function, and outputs picture matrix data.
Example 9
As shown in fig. 9, a processing module 20 of a device for matting pictures comprises:
and a feature map unit 22 for processing the image matrix by using the prediction model, the convolutional layer and the pooling layer of the convolutional neural network, and outputting a feature map.
One embodiment of the processing module 20 of the above apparatus may be: and a feature map unit 22 for processing the image matrix by using the prediction model, the convolutional layer and the pooling layer of the convolutional neural network, and outputting a feature map.
Example 10
As shown in fig. 10, one specific embodiment may be:
s1010, decoding the picture coded data and outputting picture matrix data;
firstly, decoding pictures uploaded by a user by using base64, then coding base64 into numpy matrix data by using a fromming function, and finally converting the numpy matrix data into opencv picture matrix data by using an imdecode function to complete the identification of the uploaded pictures.
S1020, establishing a convolutional neural network prediction model;
the prediction model of the convolutional neural network is generated by training the labeled data through the convolutional neural network on the GPU according to eighty percent of training data and twenty percent of verification data, and storing a model for a result generated by each Epoch (generation training). The convolutional neural network comprises a prediction model, a convolutional layer and a pooling layer, and is deployed on a GPU server.
S1030, the convolutional neural network processes the picture matrix through the prediction model, the convolutional layer and the pooling layer, and outputs a characteristic diagram;
and the server transmits the picture to neural network reasoning on the GPU, and generates a characteristic diagram in a matrix format through convolutional neural network reasoning.
S1040, calculating the feature graph by using a Gaussian function, and matting to obtain and store the picture;
the characteristic graph is calculated according to a Gaussian function, in the method, (X, y) represents a Gaussian kernel, the parameter value is 3X 3, sigma represents the standard deviation of the Gaussian kernel function in the X direction, the parameter value is 0, a picture is obtained by scratching in a CPU and is stored in formats of png, jpg, tif and the like, then the picture in a memory is encrypted through base64 to generate JSON format data, and the picture scratching by utilizing the Gaussian function has the advantage of accurate scratching.
Example 11
As shown in fig. 11, an electronic device comprises a memory 1101 and a processor 1102, the memory 1101 is used for storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor 1102 to implement a method of matting pictures as described above.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the electronic device described above may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.
A computer-readable storage medium storing a computer program which, when executed by a computer, causes the computer to implement a method of matting a picture as described above.
Illustratively, the computer program may be divided into one or more modules/units, which are stored in the memory 1101 and executed by the processor 1102 to implement the present invention. One or more modules/units may be a series of computer program instruction segments capable of performing certain functions, the instruction segments being used to describe the execution of a computer program in a computer device.
The computer device may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The computer device may include, but is not limited to, a memory 1101, a processor 1102. Those skilled in the art will appreciate that the present embodiments are merely exemplary of a computing device and are not intended to limit the computing device, and may include more or fewer components, or some of the components may be combined, or different components, e.g., the computing device may also include input output devices, network access devices, buses, etc.
The processor 1102 may be a Central Processing Unit (CPU), other general purpose processor 1102, a digital signal processor 1102 (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general purpose processor 1102 may be a microprocessor 1102 or the processor 1102 may be any conventional processor 1102 or the like.
The storage 1101 may be an internal storage unit of the computer device, such as a hard disk or a memory of the computer device. The memory 1101 may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash card (FlashCard), etc. provided on the computer device. Further, the memory 1101 may also include both an internal storage unit and an external storage device of the computer device. The memory 1101 is used to store computer programs and other programs and data required by the computer device. The memory 1101 may also be used to temporarily store data that has been output or is to be output.
The above description is only an embodiment of the present invention, but the technical features of the present invention are not limited thereto, and any changes or modifications within the technical field of the present invention by those skilled in the art are covered by the claims of the present invention.

Claims (10)

1. A method of matting pictures, comprising:
acquiring picture coded data, decoding the picture coded data, and outputting picture matrix data;
processing the picture matrix data by using a convolutional neural network, and outputting a characteristic diagram;
and calculating the characteristic graph according to Gaussian filtering, and matting to obtain a picture.
2. The method of claim 1, wherein the calculating the feature map according to gaussian filtering to obtain the image comprises:
according to a gaussian function:
Figure FDA0002527086000000011
the feature map is calculated where the parameters in the gaussian template are determined by the coordinates of (x, y) and σ represents the standard deviation.
3. The method of claim 1, wherein obtaining encoded picture data, decoding the encoded picture data, and outputting picture matrix data comprises:
analyzing the picture coding data by utilizing a fromstering function, and outputting matrix data;
and decoding the matrix data by using an imdecode function, and outputting picture matrix data.
4. The method of claim 3, wherein the processing the picture matrix data by using a convolutional neural network to output a feature map comprises:
and processing the picture matrix by using a prediction model, a convolutional layer and a pooling layer of the convolutional neural network, and outputting a characteristic diagram.
5. A device for matting pictures, comprising:
the decoding module is used for acquiring picture coded data, decoding the picture coded data and outputting picture matrix data;
the processing module is used for processing the picture matrix data by utilizing a convolutional neural network and outputting a characteristic diagram;
and the matting module calculates the characteristic graph according to Gaussian filtering and obtains a picture by matting.
6. The device of claim 5, wherein the matting module comprises:
a Gaussian function unit that, according to a Gaussian function:
Figure FDA0002527086000000021
the feature map is calculated where the parameters in the gaussian template are determined by the coordinates of (x, y) and σ represents the standard deviation.
7. The apparatus for matting picture according to claim 5, wherein said decoding module comprises:
a fromming function unit which analyzes the picture coding data by using a fromming function and outputs matrix data;
and the imdecode function unit is used for decoding the matrix data by utilizing the imdecode function and outputting the picture matrix data.
8. The device for matting pictures according to claim 7, wherein the processing module includes:
and the characteristic diagram unit is used for processing the picture matrix by utilizing the prediction model, the convolutional layer and the pooling layer of the convolutional neural network and outputting a characteristic diagram.
9. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer instructions, and wherein the one or more computer instructions are executed by the processor to implement a fine-grained sentiment dictionary based network sentiment analysis method as claimed in any one of claims 1 to 4.
10. A computer-readable storage medium storing a computer program, wherein the computer program is configured to enable a computer to execute the method for network emotion analysis based on a fine-grained emotion dictionary according to any one of claims 1 to 4.
CN202010507559.9A 2020-06-05 2020-06-05 Method for scratching picture Pending CN111833355A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010507559.9A CN111833355A (en) 2020-06-05 2020-06-05 Method for scratching picture

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010507559.9A CN111833355A (en) 2020-06-05 2020-06-05 Method for scratching picture

Publications (1)

Publication Number Publication Date
CN111833355A true CN111833355A (en) 2020-10-27

Family

ID=72897962

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010507559.9A Pending CN111833355A (en) 2020-06-05 2020-06-05 Method for scratching picture

Country Status (1)

Country Link
CN (1) CN111833355A (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104732534A (en) * 2015-03-18 2015-06-24 中国人民公安大学 Method and system for matting conspicuous object in image
CN106023220A (en) * 2016-05-26 2016-10-12 史方 Vehicle exterior part image segmentation method based on deep learning
CN108460770A (en) * 2016-12-13 2018-08-28 华为技术有限公司 Scratch drawing method and device
US20190030371A1 (en) * 2017-07-28 2019-01-31 Elekta, Inc. Automated image segmentation using dcnn such as for radiation therapy
CN109410211A (en) * 2017-08-18 2019-03-01 北京猎户星空科技有限公司 The dividing method and device of target object in a kind of image
CN109886971A (en) * 2019-01-24 2019-06-14 西安交通大学 A kind of image partition method and system based on convolutional neural networks
CN110827312A (en) * 2019-11-12 2020-02-21 北京深境智能科技有限公司 Learning method based on cooperative visual attention neural network

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104732534A (en) * 2015-03-18 2015-06-24 中国人民公安大学 Method and system for matting conspicuous object in image
CN106023220A (en) * 2016-05-26 2016-10-12 史方 Vehicle exterior part image segmentation method based on deep learning
CN108460770A (en) * 2016-12-13 2018-08-28 华为技术有限公司 Scratch drawing method and device
US20190030371A1 (en) * 2017-07-28 2019-01-31 Elekta, Inc. Automated image segmentation using dcnn such as for radiation therapy
CN109410211A (en) * 2017-08-18 2019-03-01 北京猎户星空科技有限公司 The dividing method and device of target object in a kind of image
CN109886971A (en) * 2019-01-24 2019-06-14 西安交通大学 A kind of image partition method and system based on convolutional neural networks
CN110827312A (en) * 2019-11-12 2020-02-21 北京深境智能科技有限公司 Learning method based on cooperative visual attention neural network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
GMHAPPY: "python cv2展示网络图片、图片编解码、及与base64转换_cv2读取网络图片", Retrieved from the Internet <URL:https://blog.csdn.net/ctwy291314/article/details/91493156?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522170954502816800185878624%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fall.%2522%257D&request_id=170954502816800185878624&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~first_rank_ecpm_v1~rank_v31_ecpm-1-91493156-null-null.142^v99^pc_search_result_base3&utm_term=pathon%20cv2%E5%B1%95%E7%A4%BA%E7%BD%91%E7%BB%9C%E5%9B%BE%E7%89%87%20%E5%9B%BE%E7%89%87%E7%BC%96%E8%A7%A3%E7%A0%81&spm=1018.2226.3001.4187> *

Similar Documents

Publication Publication Date Title
CN111782839A (en) Image question-answering method, image question-answering device, computer equipment and medium
CN111709406B (en) Text line identification method and device, readable storage medium and electronic equipment
WO2020082610A1 (en) Identity card information verification method and apparatus, device, and computer readable storage medium
CN112016543A (en) Text recognition network, neural network training method and related equipment
US11494614B2 (en) Subsampling training data during artificial neural network training
CN113704531A (en) Image processing method, image processing device, electronic equipment and computer readable storage medium
CN107967461B (en) SVM (support vector machine) differential model training and face verification method, device, terminal and storage medium
CN110765843B (en) Face verification method, device, computer equipment and storage medium
CN108491812B (en) Method and device for generating face recognition model
CN112767320A (en) Image detection method, image detection device, electronic equipment and storage medium
TWI803243B (en) Method for expanding images, computer device and storage medium
CN113887408B (en) Method, device, equipment and storage medium for detecting activated face video
CN113177449A (en) Face recognition method and device, computer equipment and storage medium
JP2023526899A (en) Methods, devices, media and program products for generating image inpainting models
CN109271526A (en) Method for text detection, device, electronic equipment and computer readable storage medium
CN109829857B (en) Method and device for correcting inclined image based on generation countermeasure network
CN115100659A (en) Text recognition method and device, electronic equipment and storage medium
CN114299590A (en) Training method of face completion model, face completion method and system
CN110414522A (en) A kind of character identifying method and device
CN114529785A (en) Model training method, video generation method and device, equipment and medium
CN110889290B (en) Text encoding method and apparatus, text encoding validity checking method and apparatus
CN115147547B (en) Human body reconstruction method and device
CN111833355A (en) Method for scratching picture
CN115565186A (en) Method and device for training character recognition model, electronic equipment and storage medium
CN113379637A (en) Image restoration method, system, medium, and device based on progressive learning strategy

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