CN116523792A - Image high-definition restoration method and system based on neural network - Google Patents
Image high-definition restoration method and system based on neural network Download PDFInfo
- Publication number
- CN116523792A CN116523792A CN202310516657.2A CN202310516657A CN116523792A CN 116523792 A CN116523792 A CN 116523792A CN 202310516657 A CN202310516657 A CN 202310516657A CN 116523792 A CN116523792 A CN 116523792A
- Authority
- CN
- China
- Prior art keywords
- image
- neural network
- image information
- definition
- algorithm
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000001914 filtration Methods 0.000 claims abstract description 13
- 230000009467 reduction Effects 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 4
- 230000005284 excitation Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 239000003086 colorant Substances 0.000 description 4
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration using non-spatial domain filtering
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/20024—Filtering details
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- 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]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, and particularly discloses an image high-definition reduction method based on a neural network, which comprises the following steps: acquiring image information of a target area through an image sensor, and acquiring original image information acquired by the image sensor; processing the original image information by adopting an image processing algorithm comprising a definition correlation algorithm (such as a Fourier transform algorithm) and a radius filtering method; the processed original image information is used as initial data and is input into a constructed neural network algorithm model, and high-definition restoration is carried out on the processed original image information; according to the invention, the original image information is subjected to refinement processing through the image processing algorithm, and then is input into the neural network algorithm model to perform high-definition restoration on the processed original image information, so that a high-resolution image is obtained, the resolution quality of the acquired image is improved, and the accurate identification processing of the video image information to be processed by workers is facilitated.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an image high-definition restoration method and system based on a neural network.
Background
In the current age, along with the surge development of artificial intelligence technology, not only the production and life of people are affected in various aspects, but also the development and progress of the world are promoted. In order to effectively use information collected from the respective terminals, real-time processing requirements for the information are also increasing. Particularly in the security field, the input amount of video image information is large, the calculated amount is large, and the power consumption is high.
In the prior art, the acquired video image information is affected by various factors, so that the video image information is blurred or noisy, the accuracy of the overall identification of the video image information by a worker is possibly affected, the worker can make wrong judgment, but partial noisy is still reserved in the processing of the image, the time is long, and the real image is difficult to restore.
Disclosure of Invention
The invention aims to provide an image high-definition restoration method and system based on a neural network, which are used for solving the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions:
an image high-definition reduction method based on a neural network comprises the following steps:
acquiring image information of a target area through an image sensor, and acquiring original image information acquired by the image sensor;
processing the original image information by adopting an image processing algorithm comprising a definition correlation algorithm (such as a Fourier transform algorithm) and a radius filtering method;
and taking the processed original image information as initial data, inputting the initial data into a constructed neural network algorithm model, and performing high-definition restoration on the processed original image information.
Preferably, the image information acquisition of the target area by the image sensor includes: a multi-layer image of image information of a target area is extracted by a multi-layer image feature extraction system, each layer image comprising at least two digital images based on different image sizes.
Preferably, the fourier transform algorithm transforms the image function f (x, y) into the frequency domain with a set of orthogonal function bases of the hilbert function space by treating f (x, y) asWhite light can be decomposed into seven colors of light through the lens, and the seven colors of light are f with different frequencies n (x, y) thus changing the image by filtering out light of certain frequencies.
Preferably, the radius filtering method removes noise points in the image by calculating the median of pixel values around the pixel points, the principle is to draw a circle by taking any point in the image as the center, calculate the number of points falling on the circle, and reserve the point when the number is larger than a preset value; and when the number is smaller than the preset value, eliminating the point.
Preferably, the neural network algorithm model is based on an artificial neural network mathematical model and learns in a certain learning mode, under the excitation of the environment where the neural network algorithm model is located, a plurality of sample modes are sequentially input into the network, the weight matrix of each layer of the network is adjusted according to the purpose of improving the image pixels, and the learning process is finished after the weights of each layer of the network are converged to a certain value.
An image high definition reduction system based on a neural network, comprising:
the target area image acquisition module is used for acquiring image information of a target area through the image sensor and acquiring original image information acquired by the image sensor;
the image preprocessing module is used for processing the original image information by adopting an image processing algorithm comprising a definition correlation algorithm (such as a Fourier transform algorithm) and a radius filtering method;
the image high-definition restoration module is used for taking the processed original image information as initial data, inputting the initial data into the constructed neural network algorithm model, and carrying out high-definition restoration on the processed original image information.
Preferably, the system further comprises: at least one processor; at least one memory for storing at least one program; the at least one program, when executed by the at least one processor, causes the at least one processor to implement a neural network-based image high definition reduction system as described above.
A computer readable storage medium having stored thereon a computer program which when executed by a processor implements a neural network based image high definition restoration method as defined in any of the preceding claims.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the original image information is subjected to refinement processing through the image processing algorithm, and then is input into the neural network algorithm model to perform high-definition restoration on the processed original image information, so that a high-resolution image is obtained, the resolution quality of the acquired image is improved, and the accurate identification processing of the video image information to be processed by workers is facilitated.
Drawings
FIG. 1 is a flow chart of an image high-definition restoration method based on a neural network;
fig. 2 is a diagram of a neural network-based image high-definition reduction system.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Examples:
referring to fig. 1-2, a neural network-based image high-definition reduction method includes:
acquiring image information of a target area through an image sensor, and acquiring original image information acquired by the image sensor;
processing the original image information by adopting an image processing algorithm comprising a definition correlation algorithm (such as a Fourier transform algorithm) and a radius filtering method;
and taking the processed original image information as initial data, inputting the initial data into a constructed neural network algorithm model, and performing high-definition restoration on the processed original image information.
Image information acquisition of the target area by the image sensor comprises: a multi-layer image of image information of a target area is extracted by a multi-layer image feature extraction system, each layer image comprising at least two digital images based on different image sizes.
The Fourier transform algorithm transforms the image function f (x, y) into the frequency domain by using a set of orthogonal function bases of the Hilbert function space, the principle being that f (x, y) is regarded as white light which can be decomposed into seven colors of light by a lens, the seven colors of light being f of different frequencies n (x, y) thus changing the image by filtering out light of certain frequencies.
The radius filtering method removes noise points in an image by calculating the median of pixel values around the pixel points, wherein the principle is to draw a circle by taking any point in the image as the center, calculate the number of points falling on the circle, and reserve the point when the number is larger than a preset value; and when the number is smaller than the preset value, eliminating the point.
The neural network algorithm model is based on an artificial neural network mathematical model and learns in a certain learning mode, under the excitation of the environment where the model is located, a plurality of sample modes are sequentially input into the network, the weight matrix of each layer of the network is adjusted according to the purpose of improving the image pixels, and the learning process is ended when the weights of each layer of the network are converged to a certain value.
A computer readable storage medium having stored thereon a computer program which when executed by a processor implements a neural network based image high definition restoration method as defined in any of the preceding claims.
An image high definition reduction system based on a neural network, comprising:
the target area image acquisition module is used for acquiring image information of a target area through the image sensor and acquiring original image information acquired by the image sensor;
the image preprocessing module is used for processing the original image information by adopting an image processing algorithm comprising a definition correlation algorithm (such as a Fourier transform algorithm) and a radius filtering method;
the image high-definition restoration module is used for taking the processed original image information as initial data, inputting the initial data into the constructed neural network algorithm model, and carrying out high-definition restoration on the processed original image information.
The system further comprises: at least one processor; at least one memory for storing at least one program; the at least one program, when executed by the at least one processor, causes the at least one processor to implement a neural network-based image high definition reduction system as described above.
According to the invention, the original image information is refined through the image processing algorithm, and then is input into the neural network algorithm model to perform high-definition restoration on the processed original image information, so that a high-resolution image is obtained, the resolution quality of the acquired image is improved, and the accurate identification processing of the video image information to be processed by workers is facilitated.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. The image high-definition restoration method based on the neural network is characterized by comprising the following steps of:
acquiring image information of a target area through an image sensor, and acquiring original image information acquired by the image sensor;
processing the original image information by adopting an image processing algorithm comprising a definition correlation algorithm (such as a Fourier transform algorithm) and a radius filtering method;
and taking the processed original image information as initial data, inputting the initial data into a constructed neural network algorithm model, and performing high-definition restoration on the processed original image information.
2. The neural network-based image high-definition restoration method according to claim 1, wherein the method comprises the following steps: the image information acquisition of the target area by the image sensor comprises the following steps: a multi-layer image of image information of a target area is extracted by a multi-layer image feature extraction system, each layer image comprising at least two digital images based on different image sizes.
3. The neural network-based image high-definition restoration method according to claim 1, wherein the method comprises the following steps: the Fourier transform algorithm transforms the image function f (x, y) into the frequency domain by using a set of orthogonal function bases of the Hilbert function space, the principle is that f (x, y) is regarded as white light, the white light can be decomposed into seven colored lights through a lens, and the seven colored lights are f of different frequencies n (x, y) thus changing the image by filtering out light of certain frequencies.
4. The neural network-based image high-definition restoration method according to claim 1, wherein the method comprises the following steps: the radius filtering method removes noise points in an image by calculating the median of pixel values around the pixel points, the principle is that a circle is drawn by taking any point in the image as the center, the number of points falling on the circle is calculated, and when the number is larger than a preset value, the point is reserved; and when the number is smaller than the preset value, eliminating the point.
5. The neural network-based image high-definition restoration method according to claim 1, wherein the method comprises the following steps: the neural network algorithm model is based on an artificial neural network mathematical model and learns in a certain learning mode (such as an image processing algorithm), under the excitation of the environment where the neural network algorithm model is positioned, a plurality of sample modes are sequentially input into the network, the weight matrix of each layer of the network is adjusted according to the purpose of improving the image pixels, and the learning process is finished after the weights of each layer of the network are converged to a certain value.
6. An image high definition reduction system based on a neural network, comprising:
the target area image acquisition module is used for acquiring image information of a target area through the image sensor and acquiring original image information acquired by the image sensor;
the image preprocessing module is used for processing the original image information by adopting an image processing algorithm comprising a definition correlation algorithm (such as a Fourier transform algorithm) and a radius filtering method;
the image high-definition restoration module is used for taking the processed original image information as initial data, inputting the initial data into the constructed neural network algorithm model, and carrying out high-definition restoration on the processed original image information.
7. The neural network-based image high-definition reduction system according to claim 6, wherein: the system further comprises: at least one processor; at least one memory for storing at least one program; the at least one program, when executed by the at least one processor, causes the at least one processor to implement a neural network-based image high definition reduction system as set forth in claim 6.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements a neural network based image high definition restoration method according to any of claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310516657.2A CN116523792A (en) | 2023-05-09 | 2023-05-09 | Image high-definition restoration method and system based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310516657.2A CN116523792A (en) | 2023-05-09 | 2023-05-09 | Image high-definition restoration method and system based on neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116523792A true CN116523792A (en) | 2023-08-01 |
Family
ID=87393822
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310516657.2A Pending CN116523792A (en) | 2023-05-09 | 2023-05-09 | Image high-definition restoration method and system based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116523792A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112381749A (en) * | 2020-11-24 | 2021-02-19 | 维沃移动通信有限公司 | Image processing method, image processing device and electronic equipment |
CN113284073A (en) * | 2021-07-08 | 2021-08-20 | 腾讯科技(深圳)有限公司 | Image restoration method, device and storage medium |
CN116055895A (en) * | 2023-03-29 | 2023-05-02 | 荣耀终端有限公司 | Image processing method and related device |
-
2023
- 2023-05-09 CN CN202310516657.2A patent/CN116523792A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112381749A (en) * | 2020-11-24 | 2021-02-19 | 维沃移动通信有限公司 | Image processing method, image processing device and electronic equipment |
CN113284073A (en) * | 2021-07-08 | 2021-08-20 | 腾讯科技(深圳)有限公司 | Image restoration method, device and storage medium |
CN116055895A (en) * | 2023-03-29 | 2023-05-02 | 荣耀终端有限公司 | Image processing method and related device |
Non-Patent Citations (2)
Title |
---|
中水珠江规划勘测设计有限公司等: "水陆一体化机动式船载三维时空信息获取技术", 30 June 2022, 黄河水利出版社, pages: 215 - 216 * |
陈家壁等: "光学信息技术原理及应用", 31 July 2002, 高等教育出版社, pages: 261 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111291696B (en) | Handwriting Dongba character recognition method based on convolutional neural network | |
CN112614136B (en) | Infrared small target real-time instance segmentation method and device | |
CN114359727B (en) | Tea disease identification method and system based on lightweight optimization Yolo v4 | |
CN113450288B (en) | Single image rain removing method and system based on deep convolutional neural network and storage medium | |
Wang et al. | MAGAN: Unsupervised low-light image enhancement guided by mixed-attention | |
CN108664839B (en) | Image processing method and device | |
CN110097522B (en) | Single outdoor image defogging method based on multi-scale convolution neural network | |
CN111861894A (en) | Image motion blur removing method based on generating type countermeasure network | |
CN111666813B (en) | Subcutaneous sweat gland extraction method of three-dimensional convolutional neural network based on non-local information | |
CN112200887A (en) | Multi-focus image fusion method based on gradient perception | |
Qi et al. | Research on the image segmentation of icing line based on NSCT and 2-D OSTU | |
CN115035567A (en) | Model training, incomplete face image recognition and reconstruction method, equipment and medium | |
CN108596840B (en) | Data set enhancement method for deep learning evaluation of vascular network development level | |
CN117876793A (en) | Hyperspectral image tree classification method and device | |
CN116523792A (en) | Image high-definition restoration method and system based on neural network | |
Van Noord et al. | Light-weight pixel context encoders for image inpainting | |
CN115937040A (en) | Image denoising method for assisting file digitization | |
CN115456908A (en) | Robust self-supervision image denoising method | |
CN114255203B (en) | Fry quantity estimation method and system | |
Wang et al. | Multi-patch and feature fusion network for single image Dehazing | |
CN113158997B (en) | Grain depot monitoring image denoising method, device and medium based on deep learning | |
CN111709889B (en) | Image dynamic real-time analysis system and method based on artificial intelligence | |
CN115115537B (en) | Image restoration method based on mask training | |
CN114581448B (en) | Image detection method, device, terminal equipment and storage medium | |
CN115908184B (en) | Automatic removal method and device for mole pattern |
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 |