CN116228543A - Image interpolation method based on information recovery mechanism - Google Patents
Image interpolation method based on information recovery mechanism Download PDFInfo
- Publication number
- CN116228543A CN116228543A CN202310270523.7A CN202310270523A CN116228543A CN 116228543 A CN116228543 A CN 116228543A CN 202310270523 A CN202310270523 A CN 202310270523A CN 116228543 A CN116228543 A CN 116228543A
- Authority
- CN
- China
- Prior art keywords
- image
- interpolation
- inverse
- preliminary
- resolution
- 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 49
- 230000007246 mechanism Effects 0.000 title claims abstract description 16
- 238000011084 recovery Methods 0.000 title claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims abstract description 26
- 230000008569 process Effects 0.000 claims abstract description 15
- 238000005070 sampling Methods 0.000 claims description 4
- 230000006872 improvement Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
-
- 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
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Television Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image interpolation method based on an information recovery mechanism, and belongs to the technical field of image interpolation. The method comprises the steps of firstly carrying out interpolation calculation on an original image to obtain a preliminary interpolation image. Secondly, an information recovery mechanism is constructed for the preliminary interpolation image, namely loss information in the preliminary difference image is quantized, and specifically, an error image is obtained through two steps of inverse interpolation calculation and error calculation. Finally, the preliminary interpolation image and the error image are superimposed to obtain a final interpolation image. The image interpolation method disclosed by the invention has the advantages of simple principle, can reduce the edge blurring and saw tooth phenomena brought in the interpolation process, and has higher popularization value in the application fields of satellite remote sensing, smart cities, ultra-high definition televisions and the like.
Description
Technical Field
The invention belongs to the technical field of image interpolation, and particularly relates to an image interpolation method based on an information recovery mechanism.
Background
The digital image interpolation technology is mainly used for estimating the pixel value of an unknown sampling point by using the pixel value of the known sampling point of an image, is a generation process of image data, and has wide application in the fields of satellite remote sensing, smart cities, ultra-high definition televisions and the like. In the image processing process, the resolution of the image is adjusted according to the actual application requirement, and the image interpolation is involved in the adjustment process. From the mathematical perspective, the image interpolation is to apply a certain kernel function and known neighborhood pixel points to fit the gray value of the unknown pixel points, and the quality of the interpolation method directly influences the subsequent practical engineering application.
Currently, the existing image interpolation techniques are classified into a linear interpolation method and a nonlinear interpolation method. Common linear interpolation methods include a neighbor interpolation method, a bilinear interpolation method, and the like. In the neighbor interpolation method, the pixel value to be interpolated is set as the pixel value of the nearest point in the original image, and the interpolated image has the jaggies and the blockiness. Similar problems exist with linear interpolation methods. In recent years, some nonlinear interpolation methods, such as interpolation methods based on machine learning, can improve the detail information of the interpolation image to a certain extent, but the methods have some uncontrollable factors, a large number of images are required to be subjected to model training, and the trained models usually have no generalization capability, so that local image distortion and blurring problems can occur in the image interpolation process.
In a word, the main problem of the existing image interpolation method is that detail information is lost in the interpolation process, and edge blurring and jaggies are easy to occur. Therefore, how to realize an image interpolation method with clear texture details is very important, and has very important research significance and practical value for the fields of satellite remote sensing, smart cities, ultra-high definition televisions and the like.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an image interpolation method based on an information recovery mechanism, and the technical scheme adopted by the invention is as follows:
an image interpolation method based on an information recovery mechanism comprises the following steps:
step S1: image interpolation is carried out on an original image F (x, y) to obtain a preliminary interpolation image I (x ', y'), wherein the value range of an original image variable x is x=1, 2,..M, the value range of a variable y is y=1, 2,..N, M and N are the number of rows and columns of the image respectively, the value range of the preliminary interpolation image variable x 'is x' =1, 2,..;
the next three steps are used to implement an information retrieval mechanism for image interpolation to obtain a high quality interpolated image.
Step S2: performing inverse interpolation calculation on the preliminary interpolation image I (x ', y') obtained in step S1, that is, if the image interpolation in step S1 is resolution improvement, the inverse interpolation is to reduce the resolution of I (x ', y') and obtain an image H (x, y), and this process is called inverse downsampling, and if the interpolation in step S1 is to reduce the resolution of the image, the inverse interpolation is to increase the resolution of I (x ', y') and obtain an image H (x, y), and this process is called inverse upsampling, H (x, y) is taken as a result image of the inverse interpolation, and the resolution thereof is the same as that of the original image F (x, y);
step S3: calculating errors, namely taking a difference between an original image F (x, y) and a reverse interpolation result image H (x, y) obtained in the step S2, recording the difference as A (x, y), and then carrying out interpolation calculation on the A (x, y) by using an interpolation method in the step S1 to obtain an error image E (x ', y');
step S4: and (3) superposing and outputting, namely superposing the preliminary interpolation image I (x ', y') of the step S1 and the error image E (x ', y') of the step S3 to obtain a final interpolation output image S (x ', y').
Preferably, the calculation formula of the image interpolation in the step S1 is:
Preferably, the calculation formula of the image downsampling in the step S2 is as follows:
Preferably, the calculation formula of the image up-sampling in the step S2 is:
Preferably, the calculation formula of the interpolation output image S (x ', y') in the step S4 is:
s (x ', y') =i (x ', y') +λ·e (x ', y'), where the parameter λ=1.5.
Compared with the prior art, the invention has the following beneficial effects: the invention has simple principle, acquires the error image by constructing the information recovery mechanism in the interpolation process, and superimposes the error image on the final interpolation image and outputs the final interpolation image, thereby reducing the edge blurring and the sawtooth phenomenon brought in the interpolation process. The interpolation method is suitable for the application fields of satellite remote sensing, smart cities and ultra-high definition televisions, and has higher popularization value.
Drawings
FIG. 1 is a block diagram of the steps of the present invention.
Detailed Description
In order to facilitate the understanding of the technical solution of the present invention, the technical solution of the present invention will now be described in further detail with reference to the drawings and examples of the specification.
The invention provides an image interpolation method based on an information recovery mechanism, which adopts the following technical scheme:
referring to fig. 1, the specific implementation steps of an image interpolation method based on an information recycling mechanism are as follows:
step S1: image interpolation is carried out on an original image F (x, y) to obtain a preliminary interpolation image I (x ', y'), wherein the value range of an original image variable x is x=1, 2, M, the value range of a variable y is y=1, 2, N, M and N are the number of rows and the number of columns of the image respectively, the value range of the preliminary interpolation image variable x 'is x' =1, 2, and the value range of round (M.A), y 'is y' =1, 2, and the value range of round (N.A), round () "is a rounding function, and A is the scale factor of image scaling. The scale factor A being greater than 1 indicates that the original image needs to be subjected to enlargement processing, and 0 < A < 1 indicates that the original image needs to be subjected to reduction processing.
In a specific interpolation process, a corresponding relation between coordinates of a preliminary interpolation image I (x ', y') and coordinates of an original image F (x, y) needs to be constructed according to a scale coefficient A of image scaling, and an inverse coordinate mapping calculation formula is established, namely, coordinate values of a (x ', y') coordinate system are divided by the scale coefficient A, so that pixel points needing interpolation after scaling are determined. The calculation formula of the image interpolation is as follows:
Step S2: performing inverse interpolation calculation, namely, if the image interpolation in the step S1 is to perform resolution improvement, the resolution of the image I (x ', y') is reduced and an image H (x, y) is obtained, and the process is called inverse downsampling; if the interpolation in step S1 is to reduce the resolution of the image, then the inverse interpolation is to increase the resolution of I (x ', y') and obtain the image H (x, y), and the process is referred to as inverse upsampling. H (x, y) is the same resolution as the original image F (x, y) as the result image of the inverse interpolation.
The calculation formula of the image inverse downsampling is as follows:
The calculation formula of the image inverse upsampling is:
Step S3: error calculation, namely, taking a difference between an original image F (x, y) and a reverse interpolation result image H (x, y) obtained in the step S2, and recording the difference as A (x, y), wherein a specific calculation formula is as follows:
A(x，y)＝|F(x，y)-H(x，y)|。
then, the interpolation method in step S1 is used to perform interpolation calculation on a (x, y) to obtain an error image E (x ', y').
Step S4: and (3) superposing and outputting, namely superposing the preliminary interpolation image I (x ', y') of the step S1 and the error image E (x ', y') of the step S3 to obtain a final interpolation output image S (x ', y'). The calculation formula of S (x ', y') is:
s (x ', y') =i (x ', y') +λ·e (x ', y'), where the parameter λ=1.5.
In the superposition output link, the interpolated image is output according to the type of the original image F (x, y), if F (x, y) is a gray level image, S (x ', y') is directly output, if F (x, y) is a color image, the color image is respectively processed in the steps in a sub-channel mode, and the sub-channel interpolation results are combined into one color image to be output.
It should be noted that the above embodiments can be freely combined as needed. The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (5)
1. An image interpolation method based on an information recovery mechanism is characterized in that: the method comprises the following specific steps:
step S1: image interpolation is carried out on an original image F (x, y) to obtain a preliminary interpolation image I (x ', y'), wherein the value range of an original image variable x is x=1, 2,..M, the value range of a variable y is y=1, 2,..N, M and N are the number of rows and columns of the image respectively, the value range of the preliminary interpolation image variable x 'is x' =1, 2,..;
step S2: performing inverse interpolation calculation on the preliminary interpolation image I (x ', y') obtained in step S1, that is, if the image interpolation in step S1 is resolution improvement, the inverse interpolation is to reduce the resolution of I (x ', y') and obtain an image H (x, y), and this process is called inverse downsampling, and if the interpolation in step S1 is to reduce the resolution of the image, the inverse interpolation is to increase the resolution of I (x ', y') and obtain an image H (x, y), and this process is called inverse upsampling, H (x, y) is taken as a result image of the inverse interpolation, and the resolution thereof is the same as that of the original image F (x, y);
step S3: calculating errors, namely taking a difference between an original image F (x, y) and a reverse interpolation result image H (x, y) obtained in the step S2, recording the difference as A (x, y), and then carrying out interpolation calculation on the A (x, y) by using an interpolation method in the step S1 to obtain an error image E (x ', y');
step S4: step S4: and (3) superposing and outputting, namely superposing the preliminary interpolation image I (x ', y') of the step S1 and the error image E (x ', y') of the step S3 to obtain a final interpolation output image S (x ', y').
2. The method for image interpolation based on an information reclamation mechanism as recited in claim 1, wherein: the calculation formula of the image interpolation in the step S1 is as follows:
3. The method for image interpolation based on an information reclamation mechanism as recited in claim 1, wherein: the calculation formula of the image inverse downsampling in the step S2 is as follows:
4. The method for image interpolation based on an information reclamation mechanism as recited in claim 1, wherein: the calculation formula of the image up-sampling in the step S2 is as follows:
5. The method for image interpolation based on an information reclamation mechanism as recited in claim 1, wherein: the calculation formula of the interpolation output image S (x ', y') in the step S4 is as follows: s (x ', y') =i (x ', y') +λ·e (x ', y'), where the parameter λ=1.5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310270523.7A CN116228543A (en) | 2023-03-04 | 2023-03-04 | Image interpolation method based on information recovery mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310270523.7A CN116228543A (en) | 2023-03-04 | 2023-03-04 | Image interpolation method based on information recovery mechanism |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116228543A true CN116228543A (en) | 2023-06-06 |
Family
ID=86571247
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310270523.7A Pending CN116228543A (en) | 2023-03-04 | 2023-03-04 | Image interpolation method based on information recovery mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116228543A (en) |
-
2023
- 2023-03-04 CN CN202310270523.7A patent/CN116228543A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921786B (en) | Image super-resolution reconstruction method based on residual convolutional neural network | |
| CN106910161B (en) | Single image super-resolution reconstruction method based on deep convolutional neural network | |
| CN111915484B (en) | Reference image guiding super-resolution method based on dense matching and self-adaptive fusion | |
| CN103034973B (en) | Based on the adaptive image scaling method of bicubic interpolation | |
| CN111861884B (en) | Satellite cloud image super-resolution reconstruction method based on deep learning | |
| KR20210037620A (en) | Correction Method And Device for Distorted Document Image | |
| US20030179935A1 (en) | Image processing apparatus, image processing method, image processing program, and computer-readable record medium storing image processing program | |
| CN110349087B (en) | RGB-D image high-quality grid generation method based on adaptive convolution | |
| CN107680043B (en) | Single image super-resolution output method based on graph model | |
| Chen et al. | Single image super-resolution using deep CNN with dense skip connections and inception-resnet | |
| CN111861886B (en) | Image super-resolution reconstruction method based on multi-scale feedback network | |
| CN112419150B (en) | Image super-resolution reconstruction method of arbitrary multiple based on bilateral upsampling network | |
| CN103473797B (en) | Spatial domain based on compressed sensing sampling data correction can downscaled images reconstructing method | |
| CN105427243A (en) | Video super-resolution reconstruction method based on adaptive interpolation kernel learning | |
| CN111242999B (en) | Parallax estimation optimization method based on up-sampling and accurate re-matching | |
| Shen et al. | Deeper super-resolution generative adversarial network with gradient penalty for sonar image enhancement | |
| CN109064394B (en) | Image super-resolution reconstruction method based on convolutional neural network | |
| CN115511705A (en) | Image super-resolution reconstruction method based on deformable residual convolution neural network | |
| CN113240584A (en) | Multitask gesture picture super-resolution method based on picture edge information | |
| CN113487482A (en) | Self-adaptive super-resolution method based on meta-migration learning | |
| CN116029943B (en) | Infrared image super-resolution enhancement method based on deep learning | |
| CN113240581A (en) | Real world image super-resolution method for unknown fuzzy kernel | |
| CN110211037B (en) | Image super-resolution method based on multi-stage sparse dictionary learning | |
| CN116228543A (en) | Image interpolation method based on information recovery mechanism | |
| CN111179171A (en) | Image super-resolution reconstruction method based on residual module and attention mechanism |
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 |