CN111127366A - Portrait picture restoration method, device and equipment - Google Patents
Portrait picture restoration method, device and equipment Download PDFInfo
- Publication number
- CN111127366A CN111127366A CN201911368204.XA CN201911368204A CN111127366A CN 111127366 A CN111127366 A CN 111127366A CN 201911368204 A CN201911368204 A CN 201911368204A CN 111127366 A CN111127366 A CN 111127366A
- Authority
- CN
- China
- Prior art keywords
- portrait
- picture
- portrait picture
- pictures
- processed
- 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 54
- 230000009467 reduction Effects 0.000 claims abstract description 27
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 23
- 238000012360 testing method Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 13
- 238000011176 pooling Methods 0.000 claims description 13
- 238000012795 verification Methods 0.000 claims description 10
- 238000012935 Averaging Methods 0.000 claims description 4
- 238000002790 cross-validation Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 7
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 230000006872 improvement Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The application discloses a portrait picture restoration method, which comprises the following steps: acquiring a pre-trained portrait picture restoration model, wherein the portrait picture restoration model comprises an encoder and a decoder based on a convolutional neural network structure; and if the obtained portrait picture is determined to be the processed portrait picture, reducing the processed portrait picture into a portrait original picture through the portrait picture reduction model. In the embodiment of the description, the pre-trained portrait picture restoration model is obtained, so that the processed portrait picture can be restored to the portrait original picture, and the restoration effect of the processed portrait picture is improved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, and a device for restoring a portrait picture.
Background
With the progress of the image processing technology at present, the processed image may be greatly different from the original image of the user, and there are many environments in which the original image of the user needs to be used, such as a door path using face recognition, input of image information of an examinee, and the like, but the acquired image cannot be guaranteed to be the unprocessed original image when information is acquired, and if the processed image is used during input into a system, a recognition error may occur during later recognition, so that the processed image needs to be restored.
However, in the prior art, the effect of reducing the processed portrait picture is not significant, and the requirements of users cannot be met.
Disclosure of Invention
In view of this, embodiments of the present application provide a method, an apparatus, and a device for reducing a portrait picture, which are used to solve the problem in the prior art that an effect of reducing a processed portrait picture is not significant.
The embodiment of the application adopts the following technical scheme:
the embodiment of the application provides a portrait picture restoration method, which comprises the following steps:
acquiring a pre-trained portrait picture restoration model, wherein the portrait picture restoration model comprises an encoder and a decoder based on a convolutional neural network structure;
and if the obtained portrait picture is determined to be the processed portrait picture, reducing the processed portrait picture into a portrait original picture through the portrait picture reduction model.
Further, before obtaining the pre-trained portrait picture restoration model, the method further includes:
acquiring a plurality of groups of portrait original pictures and processed portrait pictures;
establishing an initial portrait picture reduction model;
and training the portrait picture restoration model according to the multiple groups of portrait original pictures and the processed portrait pictures to obtain the portrait picture restoration model meeting preset conditions.
Further, after acquiring a plurality of groups of portrait original pictures and processed portrait pictures, the method further includes:
and carrying out noise reduction, binarization, character segmentation and normalization processing on the multiple groups of portrait original pictures and the processed portrait pictures, and inputting the processed portrait original pictures and the processed portrait pictures into an image processor so as to obtain portrait original pictures and processed portrait pictures with specific sizes.
Further, after acquiring a plurality of groups of portrait original pictures and processed portrait pictures, the method further includes:
and marking the original portrait picture and the processed portrait picture of the same person so as to divide the original portrait picture and the processed portrait picture of the same person into the same group.
Further, the encoder includes a plurality of convolutional layers and pooling layers, and/or,
the decoder comprises a plurality of convolutional layers and deconvolution layers;
the encoder compresses the processed portrait picture, and sends the compressed picture to the decoder, and the decoder receives the compressed picture and decodes the compressed picture to obtain a restored portrait picture.
Further, the encoder comprises a multi-layer structure, wherein at least one layer of the structure comprises a convolutional layer and a pooling layer;
the structure of the convolutional layer comprises at least one of the following: 5x5x96, 3x3x128, 3x3x256, 3x3x384, and 3x3x 256;
the structure of the pooling layer comprises at least one of: 3x3x96/2, 3x3x128/2, 3x3x256/2, and 3x3x 384/2.
Further, the decoder comprises a plurality of layers of structures, wherein at least one layer of the structures comprises a convolutional layer and a deconvolution layer;
the structure of the convolutional layer comprises at least one of the following: 3x3x384, 3x3x256, 3x3x128, 3x3x96, 5x5x 3;
the structure of the deconvolution layer includes at least one of: 3x3x256/2, 3x3x128/2, 3x3x96/2, 3x3x 3/2.
Further, the training of the portrait image restoration model according to the multiple groups of portrait original images and the processed portrait images obtains a portrait image restoration model meeting preset conditions, and specifically includes:
constructing a data set by a plurality of groups of portrait original pictures and processed portrait pictures, and dividing the data set into a training test set and a verification set according to a preset proportion;
screening out a training test set according to a ten-fold cross-validation method;
inputting the portrait pictures subjected to centralized training test processing into the initial portrait picture restoration model to obtain a plurality of preselected portrait picture restoration models;
respectively inputting the portrait pictures subjected to centralized verification processing into a plurality of preselected portrait picture reduction models, and calculating the reduced portrait pictures obtained by each preselected portrait picture reduction model according to an averaging method;
comparing the restored portrait picture obtained by each preselected portrait picture restoration model with the corresponding portrait original picture;
and selecting a portrait picture restoration model meeting preset conditions from the plurality of preselected portrait picture restoration models according to the comparison result.
The embodiment of the present application further provides a portrait picture restoration device, which includes:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a pre-trained portrait picture reduction model, and the portrait picture reduction model comprises an encoder and a decoder based on a convolutional neural network structure;
and the restoring unit is used for restoring the processed portrait picture into a portrait original picture through the portrait picture restoring model if the acquired portrait picture is determined to be the processed portrait picture.
An embodiment of the present application further provides a portrait picture restoration device, which is characterized in that the device includes a memory for storing computer program instructions and a processor for executing the program instructions, where the computer program instructions, when executed by the processor, trigger the device to perform the following means:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a pre-trained portrait picture reduction model, and the portrait picture reduction model comprises an encoder and a decoder based on a convolutional neural network structure;
and the restoring unit is used for restoring the processed portrait picture into a portrait original picture through the portrait picture restoring model if the acquired portrait picture is determined to be the processed portrait picture.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects: in the embodiment of the description, the pre-trained portrait picture restoration model is obtained, so that the processed portrait picture can be restored to the portrait original picture, and the restoration effect of the processed portrait picture is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic flowchart of a method for restoring a portrait picture according to a first embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a portrait picture restoration method provided in the second embodiment of the present specification;
fig. 3 is a schematic structural diagram of a portrait picture restoration apparatus provided in the third embodiment of the present specification.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, 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 application.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for restoring a portrait picture according to an embodiment of the present disclosure.
The embodiment of the present specification may be implemented by a portrait picture restoration system, and specifically includes:
step S101, obtaining a pre-trained human image restoration model, wherein the human image restoration model comprises an encoder and a decoder based on a convolutional neural network structure.
And S102, if the obtained portrait picture is determined to be the processed portrait picture, restoring the processed portrait picture to be the portrait original picture through the portrait picture restoring model.
In the embodiment of the description, the pre-trained portrait picture restoration model is obtained, so that the processed portrait picture can be restored to the portrait original picture, and the restoration effect of the processed portrait picture is improved.
Corresponding to the above embodiments, fig. 2 is a schematic flow chart of a portrait picture restoration method provided in the second embodiment of this specification.
The embodiment of the present specification may be implemented by a portrait picture restoration system, and specifically includes:
step S201, a plurality of groups of original portrait pictures and processed portrait pictures are obtained.
In step S201 of the embodiment of the present specification, a plurality of groups of original portrait pictures and processed portrait pictures may be constructed as a data set. After acquiring multiple groups of original portrait pictures and processed portrait pictures, the embodiments of the present specification may further perform the following steps:
and performing noise reduction, binarization, character segmentation and normalization processing on a plurality of groups of portrait original pictures and processed portrait pictures in the data set, and inputting the processed portrait original pictures and the processed portrait pictures into an image processor so as to obtain portrait original pictures and processed portrait pictures with specific sizes, for example, the specific size may be 227x227x 3. The image processor can apply OpenCV, which is a cross-platform computer vision library based on BSD license and can run on Linux, Windows, Android and Mac OS operating systems. The method is light and efficient, is composed of a series of C functions and a small number of C + + classes, provides interfaces of languages such as Python, Ruby, MATLAB and the like, and realizes a plurality of general algorithms in the aspects of image processing and computer vision.
After acquiring multiple groups of original portrait pictures and processed portrait pictures, the embodiments of the present specification may further perform the following steps:
and marking the original portrait picture and the processed portrait picture of the same person so as to divide the original portrait picture and the processed portrait picture of the same person into the same group.
Step S202, an initial portrait picture reduction model is established.
In step S202 in the embodiment of this specification, the encoder includes a plurality of convolution layers and a pooling layer, and/or the decoder includes a plurality of convolution layers and a deconvolution layer, where the encoder compresses the processed portrait picture and sends the compressed picture to the decoder, and the decoder receives the compressed picture and then decodes the compressed picture to obtain a restored portrait picture.
The encoder comprises a multi-layer structure, at least one of the layers comprising a convolutional layer and a pooling layer; the structure of the convolutional layer comprises at least one of the following: 5x5x96, 3x3x128, 3x3x256, 3x3x384, and 3x3x 256; the structure of the pooling layer comprises at least one of: 3x3x96/2, 3x3x128/2, 3x3x256/2, and 3x3x 384/2.
The decoder comprises a plurality of layers of structures, at least one of which comprises a convolutional layer and a deconvolution layer; the structure of the convolutional layer comprises at least one of the following: 3x3x384, 3x3x256, 3x3x128, 3x3x96, 5x5x 3; the structure of the deconvolution layer includes at least one of: 3x3x256/2, 3x3x128/2, 3x3x96/2, 3x3x 3/2.
Further, in step S202 in the embodiment of the present specification, the structure of the encoder is preferably:
the first layer is a 5x5x96 convolutional layer and a 3x3x96/2 pooling layer, the second layer is a 3x3x128 pooling layer and a 3x3x128/2 convolutional layer, the third layer is a 3x3x256 convolutional layer and a 3x3x256/2 pooling layer, the fourth layer is a 3x3x384 convolutional layer and a 3x3x384/2 pooling layer, and the fifth layer is a 3x3x256 convolutional layer.
Further, in step S203 of the embodiment of the present specification, the structure of the decoder is preferably:
the first layer is a 3x3x384 convolutional layer, the second layer is a 3x3x256/2 deconvolution layer and a 3x3x256 convolutional layer, the third layer is a 3x3x128/2 deconvolution layer and a 3x3x128 convolutional layer, the fourth layer is a 3x3x96/2 deconvolution layer and a 3x3x96 convolutional layer, and the fifth layer is a 3x3x3/2 deconvolution layer and a 5x5x3 convolutional layer.
Step S203, training the portrait picture restoration model according to the multiple groups of portrait original pictures and the processed portrait pictures to obtain a portrait picture restoration model meeting preset conditions.
In step S203 in this embodiment of the present specification, the training of the portrait image restoration model according to the multiple groups of portrait original pictures and the processed portrait images to obtain a portrait image restoration model meeting preset conditions includes:
constructing a data set by a plurality of groups of portrait original pictures and processed portrait pictures, and dividing the data set into a training test set and a verification set according to a preset proportion, for example, the training set can be divided into 4: 1, dividing the image into a training test set and a verification set, namely taking 20% of pictures as the verification set and 80% of pictures as the training test set;
screening out a training test set according to a ten-fold cross-validation method, wherein the training test set can be divided and validated by using the ten-fold cross-validation method, for example, one tenth of the training test set is used as the test set, and the rest nine tenths are used as the training set. That is, 10% of pictures in the training set are randomly selected as a test set, the rest 90% are used as the training set, and the process is repeated for ten times to obtain 10 complete training test sets;
inputting the portrait pictures processed in the training test set into the portrait picture restoration model to obtain a plurality of preselected portrait picture restoration models, for example, training the portrait picture restoration models respectively by using 10 training test sets to obtain 10 preselected portrait picture restoration models;
respectively inputting the portrait pictures subjected to centralized verification processing into a plurality of preselected portrait picture reduction models, calculating the reduced portrait pictures obtained by each preselected portrait picture reduction model according to an averaging method, for example, respectively inputting the pictures in a verification set into 10 preselected portrait picture reduction models to obtain the reduced portrait pictures, and calculating the reduced portrait pictures obtained by each preselected portrait picture reduction model according to the averaging method, wherein parameters of the pictures are corrected by using a training test set, overfitting is prevented, underfitting is avoided, the model hyperparameters are corrected by using the verification set, and the preselected portrait picture reduction models are preliminarily evaluated;
comparing the restored portrait picture obtained by each preselected portrait picture restoration model with the corresponding portrait original picture, and selecting an optimal portrait picture restoration model according to a cost function, namely calculating the square of the difference value between the pixel of the restored portrait picture and the pixel of the portrait original picture corresponding to the restored portrait picture to obtain a cost value, and selecting the preselected portrait picture restoration model with the minimum difference value according to the cost value;
and selecting a portrait picture restoration model meeting preset conditions from the plurality of preselected portrait picture restoration models according to the comparison result, wherein the portrait picture restoration model meeting the preset conditions is the portrait picture restoration model with the least square of the pixel difference value.
And step S204, restoring the processed portrait picture into a portrait original picture through the acquired portrait picture restoration model meeting the conditions.
In the embodiment of the description, the pre-trained portrait picture restoration model is obtained, so that the processed portrait picture can be restored to the portrait original picture, and the restoration effect of the processed portrait picture is improved.
Fig. 3 is a schematic structural diagram of a portrait photo restoration device provided in the third embodiment of the present disclosure.
The embodiment of the present specification may be implemented by a portrait image restoration system, and specifically includes: an acquisition unit 1 and a reduction unit 2.
The acquiring unit 1 is used for acquiring a pre-trained human image restoration model, wherein the human image restoration model comprises an encoder and a decoder based on a convolutional neural network structure;
and the restoring unit 2 is used for restoring the processed portrait picture into a portrait original picture through the portrait picture restoring model if the acquired portrait picture is determined to be the processed portrait picture.
An embodiment of the present specification further provides a portrait picture restoration device, which is characterized in that the device includes a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute the following means:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a pre-trained portrait picture reduction model, and the portrait picture reduction model comprises an encoder and a decoder based on a convolutional neural network structure;
and the restoring unit is used for restoring the processed portrait picture into a portrait original picture through the portrait picture restoring model if the acquired portrait picture is determined to be the processed portrait picture.
In the embodiment of the description, the pre-trained portrait picture restoration model is obtained, so that the processed portrait picture can be restored to the portrait original picture, and the restoration effect of the processed portrait picture is improved.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Language Description Language), traffic, pl (core unified Programming Language), HDCal, JHDL (Java Hardware Description Language), langue, Lola, HDL, laspam, hardsradware (Hardware Description Language), vhjhd (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A portrait picture restoration method is characterized by comprising the following steps:
acquiring a pre-trained portrait picture restoration model, wherein the portrait picture restoration model comprises an encoder and a decoder based on a convolutional neural network structure;
and if the obtained portrait picture is determined to be the processed portrait picture, reducing the processed portrait picture into a portrait original picture through the portrait picture reduction model.
2. The method for restoring a human image picture as claimed in claim 1, wherein before the obtaining of the pre-trained human image picture restoration model, the method further comprises:
acquiring a plurality of groups of portrait original pictures and processed portrait pictures;
establishing an initial portrait picture reduction model;
and training the portrait picture restoration model according to the multiple groups of portrait original pictures and the processed portrait pictures to obtain the portrait picture restoration model meeting preset conditions.
3. The portrait picture restoration method according to claim 2, wherein after the multiple groups of portrait original pictures and the processed portrait pictures are obtained, the method further comprises:
and carrying out noise reduction, binarization, character segmentation and normalization processing on the multiple groups of portrait original pictures and the processed portrait pictures, and inputting the processed portrait original pictures and the processed portrait pictures into an image processor so as to obtain portrait original pictures and processed portrait pictures with specific sizes.
4. The portrait picture restoration method according to claim 2, wherein after the multiple groups of portrait original pictures and the processed portrait pictures are obtained, the method further comprises:
and marking the original portrait picture and the processed portrait picture of the same person so as to divide the original portrait picture and the processed portrait picture of the same person into the same group.
5. The method for restoring a human image picture as set forth in claim 1,
the encoder includes a plurality of convolutional layers and pooling layers, and/or,
the decoder comprises a plurality of convolutional layers and deconvolution layers;
the encoder compresses the processed portrait picture, and sends the compressed picture to the decoder, and the decoder receives the compressed picture and decodes the compressed picture to obtain a restored portrait picture.
6. The portrait picture restoration method according to claim 5, wherein the encoder comprises a multi-layer structure, wherein at least one layer of the structure comprises a convolutional layer and a pooling layer;
the structure of the convolutional layer comprises at least one of the following: 5x5x96, 3x3x128, 3x3x256, 3x3x384, and 3x3x 256;
the structure of the pooling layer comprises at least one of: 3x3x96/2, 3x3x128/2, 3x3x256/2, and 3x3x 384/2.
7. The portrait picture restoration method according to claim 5, wherein the decoder comprises a plurality of layers, at least one of the layers comprising a convolutional layer and a deconvolution layer;
the structure of the convolutional layer comprises at least one of the following: 3x3x384, 3x3x256, 3x3x128, 3x3x96, 5x5x 3;
the structure of the deconvolution layer includes at least one of: 3x3x256/2, 3x3x128/2, 3x3x96/2, 3x3x 3/2.
8. The portrait picture restoration method according to any one of claims 2 to 4, wherein the training of the portrait picture restoration model according to the multiple groups of portrait original pictures and the processed portrait pictures to obtain a portrait picture restoration model meeting preset conditions specifically comprises:
constructing a data set by a plurality of groups of portrait original pictures and processed portrait pictures, and dividing the data set into a training test set and a verification set according to a preset proportion;
screening out a training test set according to a ten-fold cross-validation method;
inputting the portrait pictures subjected to centralized training test processing into the initial portrait picture restoration model to obtain a plurality of preselected portrait picture restoration models;
respectively inputting the portrait pictures subjected to centralized verification processing into a plurality of preselected portrait picture reduction models, and calculating the reduced portrait pictures obtained by each preselected portrait picture reduction model according to an averaging method;
comparing the restored portrait picture obtained by each preselected portrait picture restoration model with the corresponding portrait original picture;
and selecting a portrait picture restoration model meeting preset conditions from the plurality of preselected portrait picture restoration models according to the comparison result.
9. A portrait picture restoration device, characterized in that the device comprises:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a pre-trained portrait picture reduction model, and the portrait picture reduction model comprises an encoder and a decoder based on a convolutional neural network structure;
and the restoring unit is used for restoring the processed portrait picture into a portrait original picture through the portrait picture restoring model if the acquired portrait picture is determined to be the processed portrait picture.
10. A portrait picture restoration device, comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the device to perform the apparatus of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911368204.XA CN111127366A (en) | 2019-12-26 | 2019-12-26 | Portrait picture restoration method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911368204.XA CN111127366A (en) | 2019-12-26 | 2019-12-26 | Portrait picture restoration method, device and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111127366A true CN111127366A (en) | 2020-05-08 |
Family
ID=70503208
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911368204.XA Pending CN111127366A (en) | 2019-12-26 | 2019-12-26 | Portrait picture restoration method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111127366A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108537747A (en) * | 2018-03-22 | 2018-09-14 | 南京大学 | A kind of image repair method based on the convolutional neural networks with symmetrical parallel link |
| CN108961174A (en) * | 2018-05-24 | 2018-12-07 | 北京飞搜科技有限公司 | A kind of image repair method, device and electronic equipment |
| CN109259784A (en) * | 2018-08-27 | 2019-01-25 | 上海铱硙医疗科技有限公司 | AI prediction technique, device, equipment and the storage medium of cerebral infarction |
| CN110276731A (en) * | 2019-06-17 | 2019-09-24 | 艾瑞迈迪科技石家庄有限公司 | Endoscopic image color restoring method and device |
| CN110458763A (en) * | 2019-07-08 | 2019-11-15 | 深圳中兴网信科技有限公司 | Restoring method, system, the medium of night color image based on deep learning |
-
2019
- 2019-12-26 CN CN201911368204.XA patent/CN111127366A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108537747A (en) * | 2018-03-22 | 2018-09-14 | 南京大学 | A kind of image repair method based on the convolutional neural networks with symmetrical parallel link |
| CN108961174A (en) * | 2018-05-24 | 2018-12-07 | 北京飞搜科技有限公司 | A kind of image repair method, device and electronic equipment |
| CN109259784A (en) * | 2018-08-27 | 2019-01-25 | 上海铱硙医疗科技有限公司 | AI prediction technique, device, equipment and the storage medium of cerebral infarction |
| CN110276731A (en) * | 2019-06-17 | 2019-09-24 | 艾瑞迈迪科技石家庄有限公司 | Endoscopic image color restoring method and device |
| CN110458763A (en) * | 2019-07-08 | 2019-11-15 | 深圳中兴网信科技有限公司 | Restoring method, system, the medium of night color image based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108170667B (en) | Word vector processing method, device and equipment | |
| CN107957989B (en) | Cluster-based word vector processing method, device and equipment | |
| CN112036236B (en) | Image detection method, device and medium based on GhostNet | |
| CN109034183B (en) | Target detection method, device and equipment | |
| CN113223101B (en) | Image processing method, device and equipment based on privacy protection | |
| CN111652351A (en) | Deployment method, device and medium of neural network model | |
| CN115981870B (en) | Data processing method and device, storage medium and electronic equipment | |
| CN112308113A (en) | Target identification method, device and medium based on semi-supervision | |
| CN111340014A (en) | Living body detection method, living body detection device, living body detection apparatus, and storage medium | |
| CN112347512A (en) | Image processing method, device, equipment and storage medium | |
| CN107292002B (en) | Method and device for reconstructing digital core | |
| CN113888415B (en) | Model training and image restoration method and device | |
| CN116341642B (en) | Data processing method and device, storage medium and electronic equipment | |
| CN115511754B (en) | Low-illumination image enhancement method based on improved Zero-DCE network | |
| CN111652053A (en) | Employee attendance checking method, device and medium | |
| CN115545572B (en) | Method, device, equipment and storage medium for business wind control | |
| CN111127366A (en) | Portrait picture restoration method, device and equipment | |
| CN111753662A (en) | Traffic accident handling method, equipment and medium | |
| CN115034367A (en) | Model deployment method and device | |
| CN112949642B (en) | Character generation method and device, storage medium and electronic equipment | |
| CN115984154A (en) | Image fusion method and device, storage medium and electronic equipment | |
| CN109325127B (en) | Risk identification method and device | |
| CN113538463A (en) | Aneurysm segmentation method, device and equipment | |
| CN111539961A (en) | Target segmentation method, device and equipment | |
| CN111652074A (en) | Face recognition method, device, equipment and 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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210714 Address after: 250101 building S02, 1036 Chaochao Road, high tech Zone, Jinan City, Shandong Province Applicant after: Shandong Inspur Scientific Research Institute Co.,Ltd. Address before: Floor 6, Chaochao Road, Shandong Province Applicant before: SHANDONG INSPUR ARTIFICIAL INTELLIGENCE RESEARCH INSTITUTE Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210823 Address after: 250101 North 6th floor, S05 building, Inspur Science Park, No. 1036, Inspur Road, high tech Zone, Jinan City, Shandong Province Applicant after: Jinan Tengming Information Technology Co.,Ltd. Address before: 250101 building S02, 1036 Chaochao Road, high tech Zone, Jinan City, Shandong Province Applicant before: Shandong Inspur Scientific Research Institute Co.,Ltd. |
|
| TA01 | Transfer of patent application right |