CN112488947A - Model training and image processing method, device, equipment and computer readable medium - Google Patents
Model training and image processing method, device, equipment and computer readable medium Download PDFInfo
- Publication number
- CN112488947A CN112488947A CN202011403478.0A CN202011403478A CN112488947A CN 112488947 A CN112488947 A CN 112488947A CN 202011403478 A CN202011403478 A CN 202011403478A CN 112488947 A CN112488947 A CN 112488947A
- Authority
- CN
- China
- Prior art keywords
- image
- network
- training
- resolution
- batch
- 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
- 238000012549 training Methods 0.000 title claims abstract description 200
- 238000003672 processing method Methods 0.000 title claims description 10
- 238000000034 method Methods 0.000 claims abstract description 75
- 238000012545 processing Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 abstract description 16
- 238000010586 diagram Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 10
- 238000004891 communication Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 239000003795 chemical substances by application Substances 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000000926 separation method Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G06T5/73—
-
- 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
- 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
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4053—Super resolution, i.e. output image resolution higher than sensor resolution
- G06T3/4076—Super resolution, i.e. output image resolution higher than sensor resolution by iteratively correcting the provisional high resolution image using the original low-resolution image
-
- 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
- G06T2207/20081—Training; Learning
-
- 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]
Abstract
The embodiment of the disclosure discloses a model training method, a model training device, electronic equipment and a computer readable medium. One embodiment of the method comprises: obtaining a first batch of training samples, wherein each first training sample in the first batch of training samples comprises a first original blurred image and a first super-resolution image corresponding to the first original blurred image; inputting the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image; training an image super-resolution network using said first predictive sharp image and said first super-resolution sharp image in said first plurality of training samples. According to the embodiment, the output quality of the image super-resolution network in the prediction process is improved by performing cascade training on the image deblurring network and the image super-resolution network.
Description
Technical Field
Embodiments of the present disclosure relate to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a computer-readable medium for model training and image processing.
Background
In the field of image processing by using an artificial intelligence technology, a fuzzy image is usually input into an image deblurring network at first to obtain a predicted sharp image, and then the predicted sharp image is input into an image hyper-resolution network to obtain a predicted hyper-resolution image. The output quality of the image hyper-resolution network trained by the related technology in the prediction process is not good enough.
Disclosure of Invention
This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
Some embodiments of the present disclosure propose model training methods, apparatuses, devices and computer readable media.
In a first aspect, some embodiments of the present disclosure provide a model training method, the method comprising: obtaining a first batch of training samples, wherein each first training sample in the first batch of training samples comprises a first original blurred image and a first super-resolution image corresponding to the first original blurred image; inputting the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image; training an image super-resolution network using said first predictive sharp image and said first super-resolution sharp image in said first plurality of training samples.
In a second aspect, some embodiments of the present disclosure provide an image processing method, including: inputting a target image into an image deblurring network to obtain a clear image, wherein the image deblurring network is generated by an image deblurring network training method in any embodiment of the disclosure; and inputting the clear image into an image hyper-resolution network to obtain a hyper-resolution clear image, wherein the image hyper-resolution network is generated by an image hyper-resolution network training method in any embodiment of the disclosure.
In a third aspect, some embodiments of the present disclosure provide a model training apparatus, the apparatus comprising: a first obtaining unit, configured to obtain a first batch of training samples, where each first training sample in the first batch of training samples includes a first original blurred image and a first super-resolution image corresponding to the first original blurred image; a first input unit, configured to input the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image; a first training unit configured to train an image super-resolution network using the first predictive sharp image and the first super-resolution sharp image in the first plurality of training samples.
In a fourth aspect, some embodiments of the present disclosure provide an image processing apparatus comprising: a second input unit, configured to input a target image to an image deblurring network to obtain a sharp image, wherein the image deblurring network is generated by an image deblurring network training method in any embodiment of the present disclosure; and a third input unit configured to input the clear image into an image hyper-resolution network to obtain a hyper-resolution clear image, wherein the image hyper-resolution network is generated by an image hyper-resolution network training method in any embodiment of the disclosure.
In a fifth aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, which when executed by one or more processors, cause the one or more processors to implement the method as described in any of the implementations of the first aspect.
In a sixth aspect, some embodiments of the disclosure provide a computer readable medium having a computer program stored thereon, where the program when executed by a processor implements a method as described in any of the implementations of the first aspect.
One of the above-described various embodiments of the present disclosure has the following advantageous effects: the output quality of the image hyper-division network in the prediction process is improved. Specifically, the inventor finds that the reason why the output quality of the image hyper-resolution network trained by the related technology is not good in the prediction process is as follows: the related art is independent of training the image deblurring network and the image hyper-division network. That is, during the training process, the input of the image hyper-separation network does not include the output of the image deblurring network, but is a sample of other sources. The samples from these other sources have poor distribution with the output of the image deblurring network, which results in poor output quality of the image hyper-resolution network trained by the method in the prediction process. Based on the method, the scheme provides a training method for performing cascade training on the image deblurring network and the image hyper-division network. Specifically, the output of the image deblurring network in the training process is directly used for training the image hyper-segmentation network, and samples from other sources are not used for training the image hyper-segmentation network. In such cascade training, the input of the image hyper-separation network in the prediction process and the input of the image hyper-separation network in the training process are both the output of the image defogging network. That is, the input of the image hyper-molecular network in the prediction process and the input distribution in the training process are more consistent. Therefore, the output quality of the image hyper-division network in the prediction process is improved.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and features are not necessarily drawn to scale.
FIG. 1 is a schematic diagram of one application scenario of the model training method of some embodiments of the present disclosure;
FIG. 2 is a schematic diagram of one application scenario of the image processing method of some embodiments of the present disclosure;
FIG. 3 is a flow diagram of some embodiments of a model training method according to the present disclosure;
FIG. 4 is a flow diagram of some embodiments of an image processing method according to the present disclosure;
FIG. 5 is a flow diagram of further embodiments of a model training method according to the present disclosure;
FIG. 6 is a schematic structural diagram of some embodiments of a model training apparatus according to the present disclosure;
FIG. 7 is a schematic block diagram of some embodiments of an image processing apparatus according to the present disclosure;
FIG. 8 is a schematic structural diagram of an electronic device suitable for use in implementing some embodiments of the present disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.
It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings. The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.
It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.
The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.
The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Fig. 1 shows a schematic diagram of one application scenario in which the model training method of some embodiments of the present disclosure may be applied.
In the application scenario shown in fig. 1, first, the computing device 101 may obtain a first batch of training samples 102. Each of the first training samples 102 includes a first original blurred image and a first super-resolution image corresponding to the first original blurred image. Then, the first original blurred image in the first batch of training samples is input to a pre-trained image deblurring network 103, so as to obtain a predicted sharp image 104. Finally, the super-resolution network 105 is trained using the predictive-resolution image 104 and the super-resolution images in the training samples.
The computing device 101 may be hardware or software. When the computing device is hardware, it may be implemented as a distributed cluster composed of a plurality of servers or electronic devices, or may be implemented as a single server or a single electronic device. When the computing device is embodied as software, it may be implemented as multiple pieces of software or software modules, for example, to provide distributed services, or as a single piece of software or software module. And is not particularly limited herein.
It should be understood that the number of computing devices 101 in FIG. 1 is merely illustrative. There may be any number of computing devices 101, as desired for implementation.
With further reference to fig. 2, fig. 2 shows a schematic diagram of one application scenario in which the image processing method of some embodiments of the present disclosure may be applied.
In the application scenario shown in fig. 2, first, the computing device 201 may input a target image 202 to an image deblurring network 203, resulting in a sharp image 204. Then, the clear image 204 is input to the image super-resolution network 205, and a super-resolution clear image 206 is obtained.
The computing device 201 may be hardware or software. When the computing device is hardware, it may be implemented as a distributed cluster composed of a plurality of servers or electronic devices, or may be implemented as a single server or a single electronic device. When the computing device is embodied as software, it may be implemented as multiple pieces of software or software modules, for example, to provide distributed services, or as a single piece of software or software module. And is not particularly limited herein.
It should be understood that the number of computing devices 201 in FIG. 2 is merely illustrative. There may be any number of computing devices 201, as implementation needs dictate.
With continued reference to fig. 3, a flow 300 of some embodiments of a model training method according to the present disclosure is shown. The model training method comprises the following steps:
In some embodiments, each of the first training samples in the first plurality of training samples includes a first original blurred image and a first super-resolution image corresponding to the first original blurred image.
The first original blurred image may be an image in which the contrast is lower than a contrast preset threshold and the number of the pixel points is lower than a pixel point number preset threshold.
In some embodiments, the first super-resolution image may be an image having a contrast higher than or equal to the preset contrast threshold and a number of pixels higher than or equal to the preset number of pixels, which is the same as the content displayed by the first original blurred image.
In some embodiments, an executing agent of the model training method (e.g., the computing device shown in fig. 1) may obtain the first training samples via a wired connection or a wireless connection. It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, a uwb (ultra wideband) connection, and other wireless connection means now known or developed in the future.
In some embodiments, the image deblurring network may be any structure of network having an image feature extraction function. By way of example, the image deblurring network described above may include, but is not limited to, at least one of: GCANet (Gated Context Aggregation Network), DehazeNet (defogging Network), domain-adaptive image defogging Network, and Gated Fusion Network.
In some embodiments, the image super-resolution network may be any network with an image resolution improving function. By way of example, the image hyper-Resolution Network may include, but is not limited to, Super-Resolution Convolutional Neural Network (Super-Resolution Convolutional Neural Network), Pixel to Pixel GAN (generative countermeasure Network for pixels), and the like.
In some embodiments, the performing agent may train the hyperscoring network using the first predictive sharp image and the first super-resolution image in the first plurality of training samples by:
step one, inputting the first prediction clear image into the image hyper-resolution network to obtain a first prediction hyper-resolution clear image.
And step two, analyzing and comparing the first prediction super-resolution clear image with the first super-resolution clear image to obtain a first comparison result.
As an example, the executing entity may first determine a first difference value between each pixel value in the first predicted super-resolution image and a corresponding pixel value in the first super-resolution image, to obtain a plurality of first difference values. Then, the sum of the absolute values of the plurality of first differences is determined as the first comparison result.
And step three, determining a first loss value of the first prediction super-resolution clear image according to the first comparison result.
As an example, the execution body may determine a result of taking a logarithm of the first comparison result as the first loss value.
As still another example, the execution main body may directly determine the first comparison result as the first loss value.
And step four, responding to the situation that the image hyper-division network is not trained, and adjusting parameters in the image hyper-division network.
In some embodiments, the executing entity may determine whether the training of the image hyper-segmentation network is completed by determining whether a current training number is greater than a first preset number threshold.
In some embodiments, the executing entity may further determine whether the training of the image hyper-separation network is completed by determining whether the first loss value is smaller than a first preset loss value threshold.
According to the method provided by some embodiments of the disclosure, the output quality of the image super-resolution network in the prediction process is improved by performing cascade training on the image deblurring network and the image super-resolution network.
With continued reference to fig. 4, a flow 400 of some embodiments of an image processing method according to the present disclosure is shown. The image processing method comprises the following steps:
The image deblurring network is generated by an image deblurring network training method in any embodiment of the present disclosure.
With further reference to FIG. 5, a flow 500 of further embodiments of a model training method is illustrated. The process 500 of the model training method includes the following steps:
In some embodiments, the specific implementation of steps 501-503 and the technical effects thereof can refer to steps 301-303 in the embodiment corresponding to fig. 3, which are not described herein again.
In some embodiments, the second original blurred image may be a state where the contrast is lower than a preset contrast threshold. And the number of the pixel points is lower than the image with the preset threshold value of the number of the pixel points.
In some embodiments, the second super-resolution image may be the same as the second original blurred image. The contrast is higher than or equal to the preset contrast threshold. And the number of the pixel points is higher than or equal to the image with the preset threshold value of the number of the pixel points.
In some embodiments, the executing entity (e.g., the computing device shown in fig. 1) of the model training method may obtain the second training sample through a wired connection or a wireless connection. It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, a uwb (ultra wideband) connection, and other wireless connection means now known or developed in the future.
And 505, training the image deblurring network by using the second batch of training samples.
In some embodiments, the performing agent may train the image deblurring network using the second training sample by:
step one, inputting a second original blurred image in the second batch of training samples into the image deblurring network to obtain a second predicted sharp image.
And step two, inputting the second prediction clear image into the image hyper-resolution network to obtain a second prediction hyper-resolution clear image.
And step three, analyzing and comparing the second prediction super-resolution clear image with a second super-resolution clear image in the second batch of training samples to obtain a second comparison result.
As an example, the executing entity may first determine a second difference value between each pixel value in the second predicted super-resolution image and a corresponding pixel value in the second super-resolution image, to obtain a plurality of second difference values. Then, the sum of the absolute values of the plurality of second differences is determined as the second comparison result.
And step three, determining a second loss value of the second prediction super-resolution clear image according to the second comparison result.
As an example, the execution body may determine a result of taking a logarithm of the second comparison result as the second loss value.
As still another example, the execution main body may directly determine the second comparison result as the second loss value.
And step four, responding to the situation that the image deblurring network is not trained, and adjusting parameters in the image deblurring network.
In some embodiments, the performing subject may determine whether the training of the image deblurring network is completed by determining whether the training number is greater than a second preset number threshold.
In some embodiments, the executing entity may further determine whether the training of the image deblurring network is completed by determining whether the second loss value is smaller than a second loss value preset threshold.
In some embodiments, the third original blurred image may have a contrast ratio lower than a contrast ratio preset threshold. And the number of the pixel points is lower than the image with the preset threshold value of the number of the pixel points.
In some embodiments, the third super-resolution image may be the same as the third original blurred image. The contrast is higher than or equal to the preset contrast threshold. And the number of the pixel points is higher than or equal to the image with the preset threshold value of the number of the pixel points.
In some embodiments, the executing entity (e.g., the computing device shown in fig. 1) of the model training method may obtain the third training sample through a wired connection or a wireless connection. It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, a uwb (ultra wideband) connection, and other wireless connection means now known or developed in the future.
And step 507, training the image deblurring network and the image hyper-resolution network by using the third batch of training samples.
In some embodiments, the performing agent may train the image deblurring network and the image hyper-segmentation network using the third training sample by:
step one, inputting a third original blurred image in the third batch of training samples into the image deblurring network to obtain a third predicted sharp image.
And step two, inputting the third prediction clear image into the image hyper-resolution network to obtain a third prediction hyper-resolution clear image.
And step three, analyzing and comparing the third prediction super-resolution clear image with a third super-resolution clear image in the third batch of training samples to obtain a third comparison result.
As an example, the executing entity may first determine a third difference value between each pixel value in the third predicted super-resolution image and a corresponding pixel value in the third super-resolution image, and obtain a plurality of third difference values. Then, the sum of the absolute values of the plurality of third differences is determined as the third comparison result.
And step three, determining a third loss value of the third prediction super-resolution clear image according to the third comparison result.
As an example, the execution body may determine a result of taking a logarithm of the third comparison result as the third loss value.
As still another example, the execution main body may directly determine the third comparison result as the third loss value.
And step four, responding to the situation that the image deblurring network and the image hyper-resolution network are not trained, and adjusting parameters in the image deblurring network and the image hyper-resolution network.
In some embodiments, the performing subject may determine whether the image deblurring network and the image hyper-diversity network are trained by determining whether a current training number is greater than a third preset number threshold.
In some embodiments, the executing entity may further determine whether the training of the image deblurring network and the image hyper-segmentation network is completed by determining whether the third loss value is smaller than a preset third loss value threshold.
As can be seen from fig. 5, compared with the description of some embodiments corresponding to fig. 3, the scheme described in the flow 500 of the model training method in some embodiments corresponding to fig. 5 obtains a better quality image deblurring network by obtaining a second training sample to train the image deblurring network alone. On the basis, a third batch of training samples are obtained, and the image deblurring network and the image hyper-segmentation network are trained at the same time. The cascade prediction performance of the image deblurring network and the image hyper-division network is further improved.
With further reference to fig. 6, as an implementation of the methods illustrated in the above figures, the present disclosure provides some embodiments of a model training apparatus, which correspond to those illustrated in fig. 3, and which may be particularly applicable in various electronic devices.
As shown in FIG. 6, the model training apparatus 600 of some embodiments includes: a first acquisition unit 601, a first input unit 602, and a first training unit 603. The first obtaining unit 601 is configured to obtain a first plurality of training samples, where each of the first training samples in the first plurality of training samples includes a first original blurred image and a first super-resolution image corresponding to the first original blurred image; a first input unit 602, configured to input the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image; a first training unit 603 configured to train an image super-resolution network using the first predictive sharp image and the first super-resolution image in the first plurality of training samples.
In an optional implementation of some embodiments, the apparatus further comprises: a second obtaining unit, configured to obtain a second batch of training samples, where each second training sample in the second batch of training samples includes a second original blurred image and a second super-resolution image corresponding to the second original blurred image; a second training unit configured to train the image deblurring network using the second plurality of training samples.
In an optional implementation of some embodiments, the apparatus further comprises: a third obtaining unit configured to obtain a third plurality of training samples, each of the third plurality of training samples including a third original blurred image and a third super-resolution image corresponding to the third original blurred image; a third training unit configured to train the image deblurring network and the image hyper-resolution network using the third training samples.
It will be understood that the elements described in the apparatus 600 correspond to various steps in the method described with reference to fig. 3. Thus, the operations, features and resulting advantages described above with respect to the method are also applicable to the apparatus 600 and the units included therein, and are not described herein again.
With further reference to fig. 7, as an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of an image processing apparatus, which correspond to those shown in fig. 4, and which may be applied in particular in various electronic devices
As shown in FIG. 7, the model training apparatus 700 of some embodiments includes: a second input unit 701 and a third input unit 702. The second input unit 701 is configured to input a target image to an image deblurring network to obtain a sharp image, where the image deblurring network is generated by an image deblurring network training method in any embodiment of the present disclosure; a third input unit 702, configured to input the above-mentioned clear image into an image hyper-segmentation network, so as to obtain a hyper-segmented clear image, where the above-mentioned image hyper-segmentation network is generated by an image hyper-segmentation network training method in any embodiment of the present disclosure.
Referring now to fig. 8, a schematic diagram of an electronic device (e.g., a server or terminal device of fig. 1) 800 suitable for use in implementing some embodiments of the present disclosure is shown. The electronic device in some embodiments of the present disclosure may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle-mounted terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 8, an electronic device 800 may include a processing means (e.g., central processing unit, graphics processor, etc.) 801 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)802 or a program loaded from a storage means 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data necessary for the operation of the electronic apparatus 800 are also stored. The processing apparatus 801, the ROM 802, and the RAM 803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.
Generally, the following devices may be connected to the I/O interface 805: input devices 806 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 807 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; storage 808 including, for example, magnetic tape, hard disk, etc.; and a communication device 809. The communication means 809 may allow the electronic device 800 to communicate wirelessly or by wire with other devices to exchange data. While fig. 8 illustrates an electronic device 800 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 8 may represent one device or may represent multiple devices as desired.
In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In some such embodiments, the computer program may be downloaded and installed from a network through communications device 809, or installed from storage device 808, or installed from ROM 802. The computer program, when executed by the processing apparatus 801, performs the above-described functions defined in the methods of some embodiments of the present disclosure.
It should be noted that the computer readable medium described in some embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: obtaining a first batch of training samples, wherein each first training sample in the first batch of training samples comprises a first original blurred image and a first super-resolution image corresponding to the first original blurred image; inputting the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image; training an image super-resolution network using said first predictive sharp image and said first super-resolution sharp image in said first plurality of training samples.
Computer program code for carrying out operations for embodiments of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in some embodiments of the present disclosure may be implemented by software, and may also be implemented by hardware. The described units may also be provided in a processor, and may be described as: a processor includes a first acquisition unit, a first input unit, and a first training unit. Where the names of these units do not in some cases constitute a limitation on the unit itself, for example, the first acquisition unit may also be described as a "unit that acquires a first batch of training samples".
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), systems on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.
According to one or more embodiments of the present disclosure, there is provided a model training method including: obtaining a first batch of training samples, wherein each first training sample in the first batch of training samples comprises a first original blurred image and a first super-resolution image corresponding to the first original blurred image; inputting the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image; training an image super-resolution network using said first predictive sharp image and said first super-resolution sharp image in said first plurality of training samples.
In accordance with one or more embodiments of the present disclosure, a method further comprises: obtaining a second batch of training samples, wherein each second training sample in the second batch of training samples comprises a second original blurred image and a second super-resolution image corresponding to the second original blurred image; training the image deblurring network using the second set of training samples.
In accordance with one or more embodiments of the present disclosure, a method further comprises: obtaining a third batch of training samples, wherein each third training sample in the third batch of training samples comprises a third original blurred image and a third super-resolution sharp image corresponding to the third original blurred image; and training the image deblurring network and the image hyper-resolution network by using the third batch of training samples.
According to one or more embodiments of the present disclosure, there is provided an image processing method including: inputting a target image into an image deblurring network to obtain a clear image, wherein the image deblurring network is generated by an image deblurring network training method in any embodiment of the disclosure; and inputting the clear image into an image hyper-resolution network to obtain a hyper-resolution clear image, wherein the image hyper-resolution network is generated by an image hyper-resolution network training method in any embodiment of the disclosure.
According to one or more embodiments of the present disclosure, there is provided a model training apparatus including: a first obtaining unit, configured to obtain a first batch of training samples, where each first training sample in the first batch of training samples includes a first original blurred image and a first super-resolution image corresponding to the first original blurred image; a first input unit, configured to input the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image; a first training unit configured to train an image super-resolution network using the first predictive sharp image and the first super-resolution sharp image in the first plurality of training samples.
According to one or more embodiments of the present disclosure, an apparatus further comprises: a second obtaining unit, configured to obtain a second batch of training samples, where each second training sample in the second batch of training samples includes a second original blurred image and a second super-resolution image corresponding to the second original blurred image; a second training unit configured to train the image deblurring network using the second plurality of training samples.
According to one or more embodiments of the present disclosure, an apparatus further comprises: a third obtaining unit configured to obtain a third plurality of training samples, each of the third plurality of training samples including a third original blurred image and a third super-resolution image corresponding to the third original blurred image; a third training unit configured to train the image deblurring network and the image hyper-resolution network using the third training samples.
According to one or more embodiments of the present disclosure, there is provided an image processing apparatus including: a second input unit, configured to input a target image to an image deblurring network to obtain a sharp image, wherein the image deblurring network is generated by an image deblurring network training method in any embodiment of the present disclosure; and a third input unit configured to input the clear image into an image hyper-resolution network to obtain a hyper-resolution clear image, wherein the image hyper-resolution network is generated by an image hyper-resolution network training method in any embodiment of the disclosure.
According to one or more embodiments of the present disclosure, there is provided an electronic device including: one or more processors; a storage device having one or more programs stored thereon which, when executed by one or more processors, cause the one or more processors to implement a method as in any above.
According to one or more embodiments of the present disclosure, a computer-readable medium is provided, on which a computer program is stored, wherein the program, when executed by a processor, implements the method as any one of the above.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.
Claims (10)
1. A model training method, comprising:
obtaining a first batch of training samples, wherein each first training sample in the first batch of training samples comprises a first original blurred image and a first super-resolution image corresponding to the first original blurred image;
inputting the first original blurred image in the first batch of training samples into a pre-trained image deblurring network to obtain a first predicted sharp image;
training an image super-resolution network using the first predictive sharp image and the first super-resolution sharp image in the first batch of training samples.
2. The method of claim 1, wherein the method further comprises:
obtaining a second batch of training samples, wherein each second training sample in the second batch of training samples comprises a second original blurred image and a second super-resolution sharp image corresponding to the second original blurred image;
training the image deblurring network using the second batch of training samples.
3. The method of claim 2, wherein the method further comprises:
obtaining a third batch of training samples, wherein each third training sample in the third batch of training samples comprises a third original blurred image and a third super-resolution sharp image corresponding to the third original blurred image;
training the image deblurring network and the image hyper-segmentation network using the third batch of training samples.
4. An image processing method comprising:
inputting a target image into an image deblurring network to obtain a sharp image, wherein the image deblurring network is generated by the method of claim 2;
inputting the sharp image into an image hyper-resolution network to obtain a hyper-resolution sharp image, wherein the image hyper-resolution network is generated by the method of one of claim 1 or claim 3.
5. A model training apparatus comprising:
a first obtaining unit configured to obtain a first batch of training samples, each of the first batch of training samples including a first original blurred image and a first super-resolution image corresponding to the first original blurred image;
a first input unit configured to input the first original blurred image in the first batch of training samples to a pre-trained image deblurring network to obtain a first predicted sharp image;
a first training unit configured to train an image super-resolution network using the first predictive sharp image and the first super-resolution sharp image in the first batch of training samples.
6. The apparatus of claim 5, wherein the apparatus further comprises:
a second obtaining unit configured to obtain a second batch of training samples, each of the second batch of training samples including a second original blurred image and a second super-resolution image corresponding to the second original blurred image;
a second training unit configured to train the image deblurring network using the second batch of training samples.
7. The apparatus of claim 6, wherein the apparatus further comprises:
a third obtaining unit configured to obtain a third plurality of training samples, each of the third plurality of training samples including a third raw blurred image and a third super-resolution image corresponding to the third raw blurred image;
a third training unit configured to train the image deblurring network and the image hyper-diversity network using the third batch of training samples.
8. An image processing apparatus comprising:
a second input unit configured to input a target image to an image deblurring network resulting in a sharp image, wherein the image deblurring network is generated by the method of claim 2;
a third input unit configured to input the sharp image to an image hyper-resolution network resulting in a hyper-resolved sharp image, wherein the image hyper-resolution network is generated by the method of one of claim 1 or claim 3.
9. An electronic device, comprising:
one or more processors;
a storage device having one or more programs stored thereon,
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-3 or claim 4.
10. A computer readable medium, having a computer program stored thereon, wherein the program, when executed by a processor, implements the method of any of claims 1-3 or 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403478.0A CN112488947A (en) | 2020-12-04 | 2020-12-04 | Model training and image processing method, device, equipment and computer readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403478.0A CN112488947A (en) | 2020-12-04 | 2020-12-04 | Model training and image processing method, device, equipment and computer readable medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112488947A true CN112488947A (en) | 2021-03-12 |
Family
ID=74939360
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011403478.0A Pending CN112488947A (en) | 2020-12-04 | 2020-12-04 | Model training and image processing method, device, equipment and computer readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112488947A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112991212A (en) * | 2021-03-16 | 2021-06-18 | Oppo广东移动通信有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110136055A (en) * | 2018-02-02 | 2019-08-16 | 腾讯科技（深圳）有限公司 | Super-resolution method and device, storage medium, the electronic device of image |
| CN110569721A (en) * | 2019-08-01 | 2019-12-13 | 平安科技（深圳）有限公司 | Recognition model training method, image recognition method, device, equipment and medium |
| CN110647936A (en) * | 2019-09-20 | 2020-01-03 | 北京百度网讯科技有限公司 | Training method and device for video super-resolution reconstruction model and electronic equipment |
| CN110728626A (en) * | 2018-07-16 | 2020-01-24 | 宁波舜宇光电信息有限公司 | Image deblurring method and apparatus and training thereof |
| CN111105375A (en) * | 2019-12-17 | 2020-05-05 | 北京金山云网络技术有限公司 | Image generation method, model training method and device thereof, and electronic equipment |
| CN111402159A (en) * | 2020-03-13 | 2020-07-10 | 北京字节跳动网络技术有限公司 | Image processing method, image processing device, electronic equipment and computer readable medium |
| CN111553840A (en) * | 2020-04-10 | 2020-08-18 | 北京百度网讯科技有限公司 | Image super-resolution model training and processing method, device, equipment and medium |
| CN111598808A (en) * | 2020-05-18 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Image processing method, device and equipment and training method thereof |
| CN111695421A (en) * | 2020-04-30 | 2020-09-22 | 北京迈格威科技有限公司 | Image recognition method and device and electronic equipment |
| CN111784578A (en) * | 2020-06-28 | 2020-10-16 | Oppo广东移动通信有限公司 | Image processing method, image processing device, model training method, model training device, image processing equipment and storage medium |
| US20200372614A1 (en) * | 2019-05-22 | 2020-11-26 | Nec Laboratories America, Inc. | Image/video deblurring using convolutional neural networks with applications to sfm/slam with blurred images/videos |
-
2020
- 2020-12-04 CN CN202011403478.0A patent/CN112488947A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110136055A (en) * | 2018-02-02 | 2019-08-16 | 腾讯科技（深圳）有限公司 | Super-resolution method and device, storage medium, the electronic device of image |
| CN110728626A (en) * | 2018-07-16 | 2020-01-24 | 宁波舜宇光电信息有限公司 | Image deblurring method and apparatus and training thereof |
| US20200372614A1 (en) * | 2019-05-22 | 2020-11-26 | Nec Laboratories America, Inc. | Image/video deblurring using convolutional neural networks with applications to sfm/slam with blurred images/videos |
| CN110569721A (en) * | 2019-08-01 | 2019-12-13 | 平安科技（深圳）有限公司 | Recognition model training method, image recognition method, device, equipment and medium |
| CN110647936A (en) * | 2019-09-20 | 2020-01-03 | 北京百度网讯科技有限公司 | Training method and device for video super-resolution reconstruction model and electronic equipment |
| CN111105375A (en) * | 2019-12-17 | 2020-05-05 | 北京金山云网络技术有限公司 | Image generation method, model training method and device thereof, and electronic equipment |
| CN111402159A (en) * | 2020-03-13 | 2020-07-10 | 北京字节跳动网络技术有限公司 | Image processing method, image processing device, electronic equipment and computer readable medium |
| CN111553840A (en) * | 2020-04-10 | 2020-08-18 | 北京百度网讯科技有限公司 | Image super-resolution model training and processing method, device, equipment and medium |
| CN111695421A (en) * | 2020-04-30 | 2020-09-22 | 北京迈格威科技有限公司 | Image recognition method and device and electronic equipment |
| CN111598808A (en) * | 2020-05-18 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Image processing method, device and equipment and training method thereof |
| CN111784578A (en) * | 2020-06-28 | 2020-10-16 | Oppo广东移动通信有限公司 | Image processing method, image processing device, model training method, model training device, image processing equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| ZHEN CUI等: "Deep Network Cascade for Image Super-resolution", 《ECCV》, 31 December 2014 (2014-12-31), pages 49 - 64, XP055432639, DOI: 10.1007/978-3-319-10602-1_4 * |
| 郎晋伟: "光学投影断层成像超分辨率重建方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 2018, 15 April 2018 (2018-04-15), pages 138 - 2780 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112991212A (en) * | 2021-03-16 | 2021-06-18 | Oppo广东移动通信有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110298413B (en) | Image feature extraction method and device, storage medium and electronic equipment | |
| CN110413812B (en) | Neural network model training method and device, electronic equipment and storage medium | |
| CN111784712B (en) | Image processing method, device, equipment and computer readable medium | |
| CN111459364B (en) | Icon updating method and device and electronic equipment | |
| CN110991373A (en) | Image processing method, image processing apparatus, electronic device, and medium | |
| CN112257582A (en) | Foot posture determination method, device, equipment and computer readable medium | |
| CN112418249A (en) | Mask image generation method and device, electronic equipment and computer readable medium | |
| CN112257478A (en) | Code scanning method, device, terminal and storage medium | |
| CN111209432A (en) | Information acquisition method and device, electronic equipment and computer readable medium | |
| CN112418232A (en) | Image segmentation method and device, readable medium and electronic equipment | |
| CN112488947A (en) | Model training and image processing method, device, equipment and computer readable medium | |
| CN111756953A (en) | Video processing method, device, equipment and computer readable medium | |
| CN110852242A (en) | Watermark identification method, device, equipment and storage medium based on multi-scale network | |
| CN112764629B (en) | Augmented reality interface display method, device, equipment and computer readable medium | |
| CN111726476B (en) | Image processing method, device, equipment and computer readable medium | |
| CN110807114B (en) | Method, device, terminal and storage medium for picture display | |
| CN112418233A (en) | Image processing method, image processing device, readable medium and electronic equipment | |
| CN113177176A (en) | Feature construction method, content display method and related device | |
| CN112070034A (en) | Image recognition method and device, electronic equipment and computer readable medium | |
| CN111680754A (en) | Image classification method and device, electronic equipment and computer-readable storage medium | |
| CN112488943B (en) | Model training and image defogging method, device and equipment | |
| CN112215789B (en) | Image defogging method, device, equipment and computer readable medium | |
| CN112070163B (en) | Image segmentation model training and image segmentation method, device and equipment | |
| CN112346630B (en) | State determination method, device, equipment and computer readable medium | |
| CN113435528B (en) | Method, device, readable medium and electronic equipment for classifying objects |
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 |