CN112184732A - Intelligent image processing method - Google Patents
Intelligent image processing method Download PDFInfo
- Publication number
- CN112184732A CN112184732A CN202011038230.9A CN202011038230A CN112184732A CN 112184732 A CN112184732 A CN 112184732A CN 202011038230 A CN202011038230 A CN 202011038230A CN 112184732 A CN112184732 A CN 112184732A
- Authority
- CN
- China
- Prior art keywords
- image
- information
- block
- module
- blocks
- 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.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 40
- 230000006835 compression Effects 0.000 claims abstract description 29
- 238000007906 compression Methods 0.000 claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 27
- 238000013507 mapping Methods 0.000 claims abstract description 20
- 230000009467 reduction Effects 0.000 claims abstract description 12
- 230000008569 process Effects 0.000 claims description 5
- 230000006798 recombination Effects 0.000 claims description 3
- 238000005215 recombination Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 abstract description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 2
- 206010016759 Flat affect Diseases 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
-
- 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
- G06T2207/10004—Still image; Photographic image
Abstract
The invention discloses an intelligent image processing method, which comprises the following processing steps: A. the method comprises the steps of decomposing an acquired image to obtain contour information and main information, removing noise from the contour information through a threshold algorithm, and combining the contour information with the main information to achieve the purpose of removing the noise; B. the image is divided into two different blocks, the mapping block is compressed by a compression algorithm, and a compression block with the minimum error rate is searched for each domain block to replace the domain block, so that the compression effect is achieved. The original image is decomposed into the main information and the outline information, and only the noise reduction processing is needed to be carried out on the outline information, so that the method has the advantages of quickly removing the noise in the original image and completely reserving the effective information of the original image; by adopting the method for encoding the defined block and the mapping block, the beneficial effects of high compression ratio, low distortion ratio and quick and accurate calculation can be realized.
Description
Technical Field
The invention relates to the field of computers and intelligent computing, in particular to an intelligent image processing method.
Background
With the progress of science and technology, the digital image processing technology has wide application field and rapid development, and the high efficiency and precision of the image are particularly important for the image processing in the process of acquisition and transmission, and a good image processing method not only can rapidly code the image, but also can reserve the useful information of the image to the maximum extent. In the prior art, the image information is seriously lost by a quick coding algorithm, and the coding algorithm is relatively complex on the premise of keeping the image information.
Disclosure of Invention
The invention discloses an intelligent image processing method, which aims to solve the technical problems that image information is seriously lost by a quick coding algorithm and the coding algorithm is relatively complex when reserved image information is complete during the conventional image processing, and realize the technical effects of high compression ratio and good original image information storage during the image processing.
The technical scheme of the invention is as follows:
an intelligent image processing method comprises the following steps:
A. decomposing original image information into main information and contour information, and removing noise through a threshold algorithm;
decomposing original image information into main information and contour information, wherein the following method is adopted:
wherein f (m, n) is image discrete acquisition data, p (i, j) is image main information data, q (i, j) is image contour information, and h, j is a pair of quadrature mirror image filter banks (QMF);
the decomposed profile information q (i, j) needs to be denoised by a threshold algorithm:
by setting the magnitude of the threshold μ, the noise information in the contour information Q (i, j) is removed to obtain new image information Q (i, j), and then the image is reorganized:
F(i,j)＝∑∑p(i,j)hi-2m,j-2n+∑∑Q(i,j)gi-2m,j-2n
wherein F (i, j) is image information obtained by the recombination;
B. dividing the image into a defined block and a mapping block, and replacing the most matched defined block with the compressed mapping block to realize the compression coding of the image; the specific process is as follows:
(1) averagely dividing the image into a blocks F of size M × NM×NCalled definition block, and divides the image into beta blocks F of 2 Mx 2N size2M×2NCalled mapping block, mapping block F is obtained by the following method2M×2NAnd (3) performing compression transformation:
(2) each submodule that defines a block finds the best matching block from the compressed blocks.
Further, for each sub-module defining a block to find the best matching block from the compressed blocks, the specific method is as follows:
e (F, ω) is the image distortion rate, the defined blocks obtain the best matching blocks by finding the compressed blocks with the lowest distortion rate, and when each defined block finds the best matching block, the blocks are recombined, i.e. the encoding is completed.
Further, the intelligent image processing system based on which the image intelligent processing method is based comprises:
the image acquisition receiving module is used for acquiring related image information and carrying out digital processing on the acquired image;
the image noise reduction processing module is used for carrying out noise reduction processing on the received image and is connected with the image acquisition receiving module through a data bus;
the image compression coding module is used for recoding the image and is connected with the image noise reduction processing module through a data bus;
the image storage module stores the processed image, and the image output module is connected with the image compression coding module through a data bus.
Further, when the intelligent image processing system is applied, the image acquisition receiving module receives the acquired image and then sends the image to the image denoising processing module, the image denoising processing module removes the drying process of the image and then sends the processed image to the image compression coding module, the image compression coding module compresses and codes the image to obtain a final processed image, and the final processed image is sent to the image storage module for storing and sending the image through a network.
The invention has at least the following beneficial effects:
(1) the method and the device decompose the original image into the main information and the outline information, and only need to perform noise reduction processing on the outline information, so that the method and the device have the advantages of quickly removing noise in the original image and completely reserving effective information of the original image.
(2) The invention adopts the method of defining the block and mapping the block to encode, has the advantages of quick and accurate calculation, and has the advantages of high compression ratio and low distortion ratio.
Drawings
FIG. 1 is a block diagram of an intelligent image processing method according to the present invention;
FIG. 2 is a flowchart of an image denoising method according to the present invention;
FIG. 3 is a flow chart of a method for compressing an image according to the present invention.
Detailed Description
For a better understanding of the present invention, reference will now be made in detail to the embodiments illustrated in the accompanying drawings.
The invention discloses an intelligent image processing method, which comprises the following processing steps:
A. and decomposing the original image information into main information and contour information, and removing noise through a threshold algorithm.
The obtained image information is decomposed into two parts, namely main information and outline information, and the image is decomposed by adopting the following method:
wherein m and n are pixel coordinate values of the discrete image, and i and j are pixel coordinate values of the original image; f (m, n) is image discrete acquisition data, p (i, j) is image main information data, q (i, j) is image contour information, and h, g is a pair of quadrature mirror filter banks (QMF).
The decomposed profile information q (i, j) needs to be denoised by a threshold algorithm:
the threshold algorithm can remove noise information in the profile information q (i, j) by setting the size of the threshold mu, and random distribution factors are used as denominators, so that the influence of threshold selection rationality on the denoising effect can be reduced; new image information Q (i, j) is obtained, after which the images are recombined:
F(i,j)＝∑∑p(i,j)hi-2m,j-2n+∑∑Q(i,j)gi-2m,j-2n
here, F (i, j) is image information obtained by the recombination.
The beneficial effects of the treatment are as follows: the noise information in the original image can be effectively and quickly removed, and the information of the original image is well reserved.
B. And dividing the image into a defined block and a mapping block, and replacing the best matched defined block by the compressed mapping block to realize the compression coding of the image.
Averagely dividing the image into a blocks F of size M × NM×NCalled definition block, and divides the image into beta blocks F of 2 Mx 2N size2M×2NCalled mapping block, mapping block F is obtained by the following method2M×2NTo carry outCompression transformation:
by the method, each mapping block can be effectively and quickly compressed, each mapping block simultaneously refers to the information of the adjacent blocks, and the obtained compressed block information is more complete.
For each sub-module defining a block to find the best matching block from the compressed blocks, the method is as follows:
e (F, ω) is the image distortion rate, and the block is defined to obtain the best matching block by searching the compressed block with the lowest distortion rate. When the best matching block is found in each defined block, the blocks are recombined, and the coding is completed. The images processed by the steps A and B can be used for storage or real-time transmission, and the storage space or the network resource occupation in transmission can be saved to the maximum extent on the premise of saving effective information to the maximum extent.
The invention also discloses an image intelligent processing system, as shown in fig. 1, comprising an image acquisition receiving module 10, an image denoising processing module 20, an image compression coding module 30 and an image storage module 40.
The image collecting and receiving module 10 is configured to collect relevant image information, perform digital processing on the collected image, and transmit the collected digital information to the image denoising processing module 20. The image acquisition receiving module 10 is connected with the image noise reduction processing module 20 through a data bus.
The image denoising processing module 20 is configured to perform denoising processing on a received image, where in an acquisition process of the image, a noise point that does not exist in an actual image is inevitably generated, which may cause information distortion to affect loss of useful information of the image for the image processing performed later, referring to the figure, the present invention performs information decomposition on the image to obtain main information and contour information of the image, then processes the contour information to remove the existing noise information, and then combines the processed contour information with the main information obtained by decomposition to achieve an effect of removing noise; the image noise reduction processing module 20 is connected with the image compression coding module 30 through a data bus.
The image compression encoding module 30 is configured to re-encode the image, and after the noise reduction processing is performed by the module 20, the size of the image is not changed, and the image needs to be re-encoded, so that the re-encoded image can be stored in a smaller space than the original image. As shown in fig. 3, the image compression/encoding module 30 firstly divides the image into a plurality of M × N blocks FM×NCalled definition block, and divides the image into beta blocks F of 2 Mx 2N size2M×2NReferred to as mapping blocks, the mapping blocks herein may overlap each other. And each sub-module of the defined block searches the best matched block from the compressed blocks for replacement, thereby achieving the compression effect.
The image compression encoding module 30 is connected with the image output module 40 through a data bus.
The image storage module 40 stores the image processed by the present invention, and because the original image is denoised and compressed, the obtained image can be stored in a smaller space than the original image, so as to save the storage space.
The operation flow of the intelligent image processing system applying the intelligent image processing method is as follows:
the image collecting and receiving module 10 sends the collected image to the image denoising processing module 20 after receiving the image, and the image denoising processing module 20 sends the processed image to the image compression coding module 30 after the image is subjected to the image denoising processing by adopting the method A of the invention; the image compression and encoding module 30 compresses and encodes the image according to the method B of the present invention to obtain a final processed image, and sends the image to the image storage module 40 for storage and sending the image through a network.
In summary, the original image is decomposed into the main information and the contour information, and only the noise reduction processing is performed on the contour information, so that the method has the beneficial effects of quickly removing the noise in the original image and completely retaining the effective information of the original image. Meanwhile, the method for coding the defined block and the mapping block is adopted, and the method has the advantages of being fast and accurate in calculation, high in compression ratio and low in distortion rate.
It should be understood that the above are only preferred embodiments of the present invention, and any modification made based on the spirit of the present invention should be within the scope of the present invention.
Claims (4)
1. An intelligent image processing method is characterized by comprising the following steps:
A. decomposing original image information into main information and contour information, and removing noise through a threshold algorithm; decomposing original image information into main information and contour information, wherein the following method is adopted:
wherein f (m, n) is image discrete acquisition data, p (i, j) is image main information data, q (i, j) is image contour information, and h, j is a pair of quadrature mirror image filter banks (QMF);
the decomposed profile information q (i, j) needs to be denoised by a threshold algorithm:
by setting the magnitude of the threshold μ, the noise information in the contour information Q (i, j) is removed to obtain new image information Q (i, j), and then the image is reorganized:
F(i,j)＝∑∑p(i,j)hi-2m,j-2n+∑∑Q(i,j)gi-2m,j-2n
wherein F (i, j) is image information obtained by the recombination;
B. dividing the image into a defined block and a mapping block, and replacing the most matched defined block with the compressed mapping block to realize the compression coding of the image; the specific process is as follows:
(1) averagely dividing the image into a blocks F of size M × NM′NCalled definition block, and divides the image into beta blocks F of 2 Mx 2N size2M′2NCalled mapping block, mapping block F is obtained by the following method2M′2NAnd (3) performing compression transformation:
(2) each submodule that defines a block finds the best matching block from the compressed blocks.
2. The intelligent image processing method as claimed in claim 1, wherein for each sub-module defining a block, the best matching block is found from the compressed blocks by:
e (F, ω) is the image distortion rate, the defined blocks obtain the best matching blocks by finding the compressed blocks with the lowest distortion rate, and when each defined block finds the best matching block, the blocks are recombined, i.e. the encoding is completed.
3. The intelligent image processing method according to claim 2, wherein the intelligent image processing system on which the intelligent image processing method is based comprises:
the image acquisition receiving module is used for acquiring related image information and carrying out digital processing on the acquired image;
the image noise reduction processing module is used for carrying out noise reduction processing on the received image and is connected with the image acquisition receiving module through a data bus;
the image compression coding module is used for recoding the image and is connected with the image noise reduction processing module through a data bus;
the image storage module stores the processed image, and the image output module is connected with the image compression coding module through a data bus.
4. The intelligent image processing method according to claim 3, wherein when the intelligent image processing system is applied, the image acquisition and reception module receives the acquired image and then sends the image to the image denoising processing module, the image denoising processing module performs denoising processing on the image and then sends the processed image to the image compression and encoding module, the image compression and encoding module performs compression and encoding on the image to obtain a final processed image, and the final processed image is sent to the image storage module for storing and sending the image through the network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011038230.9A CN112184732B (en) | 2020-09-27 | 2020-09-27 | Intelligent image processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011038230.9A CN112184732B (en) | 2020-09-27 | 2020-09-27 | Intelligent image processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112184732A true CN112184732A (en) | 2021-01-05 |
| CN112184732B CN112184732B (en) | 2022-05-24 |
Family
ID=73944427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011038230.9A Active CN112184732B (en) | 2020-09-27 | 2020-09-27 | Intelligent image processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112184732B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03102579A (en) * | 1989-09-18 | 1991-04-26 | Fujitsu Ltd | Noise removal system for image data |
| CN102523453A (en) * | 2011-12-29 | 2012-06-27 | 西安空间无线电技术研究所 | Super large compression method and transmission system for images |
| CN104766289A (en) * | 2015-03-20 | 2015-07-08 | 华南理工大学 | Image denoising and compressing method |
| CN106612439A (en) * | 2016-02-04 | 2017-05-03 | 四川用联信息技术有限公司 | Adaptive fast fractal image compression method |
| CN109391818A (en) * | 2018-11-30 | 2019-02-26 | 昆明理工大学 | A kind of fast search Fractal Image Compression Approach based on dct transform |
| US20200267386A1 (en) * | 2019-02-18 | 2020-08-20 | Shanghai Zhaoxin Semiconductor Co., Ltd. | Image compression method and image processing system |
-
2020
- 2020-09-27 CN CN202011038230.9A patent/CN112184732B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03102579A (en) * | 1989-09-18 | 1991-04-26 | Fujitsu Ltd | Noise removal system for image data |
| CN102523453A (en) * | 2011-12-29 | 2012-06-27 | 西安空间无线电技术研究所 | Super large compression method and transmission system for images |
| CN104766289A (en) * | 2015-03-20 | 2015-07-08 | 华南理工大学 | Image denoising and compressing method |
| CN106612439A (en) * | 2016-02-04 | 2017-05-03 | 四川用联信息技术有限公司 | Adaptive fast fractal image compression method |
| CN109391818A (en) * | 2018-11-30 | 2019-02-26 | 昆明理工大学 | A kind of fast search Fractal Image Compression Approach based on dct transform |
| US20200267386A1 (en) * | 2019-02-18 | 2020-08-20 | Shanghai Zhaoxin Semiconductor Co., Ltd. | Image compression method and image processing system |
Non-Patent Citations (1)
| Title |
|---|
| 李建国: "正交镜像滤波器组的原理及实现", 《中国优秀硕士学位论文全文数据库（电子期刊）信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112184732B (en) | 2022-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111641832B (en) | Encoding method, decoding method, device, electronic device and storage medium | |
| Nashat et al. | Image compression based upon wavelet transform and a statistical threshold | |
| CN114742911A (en) | Image compressed sensing reconstruction method, system, equipment and medium | |
| CN112150356A (en) | Single compressed image super-resolution reconstruction method based on cascade framework | |
| CN112184732B (en) | Intelligent image processing method | |
| Singh et al. | Review of image compression techniques | |
| CN112085667A (en) | Deblocking effect removing method and device based on pseudo-analog video transmission | |
| CN112004092A (en) | BP neural network image compression method based on improved image block classification algorithm | |
| CN1281065C (en) | Tree-structure-based grade tree aggregation-divided video image compression method | |
| CN115695793A (en) | JPEG image compression system | |
| CN112509107B (en) | Point cloud attribute re-coloring method, device and encoder | |
| CN112381206A (en) | Deep neural network compression method, system, storage medium and computer equipment | |
| CN108848385B (en) | Block data compression method suitable for micro-nano satellite | |
| CN114882133B (en) | Image coding and decoding method, system, device and medium | |
| CN101267557B (en) | A method for image compression based on compound vector quantification | |
| Li et al. | A wideband spectrum data segment compression algorithm in cognitive radio networks | |
| CN104346819A (en) | Maximum error image compression method based on limited wavelet synopses | |
| CN108668131B (en) | Image compression coding method based on sparse representation | |
| CN115065817B (en) | Hologram compression method, encoder and hologram reproduction module | |
| Aggarwal et al. | ECG Image Compression: Essentially Non-Oscillatory Interpolation Technique and Lifting Schemes | |
| Mahmood | Suggested hybrid Transform Technique for image compression | |
| CN113052258A (en) | Convolution method, model and computer equipment based on middle layer characteristic diagram compression | |
| Mohanty et al. | A novel speck algorithm for faster image compression | |
| Rao et al. | Analysis And Synthesis of Image Using Wavelet Transform and Compression Methods | |
| CN117974504A (en) | Low-rank prior-driven non-local matching hyperspectral image recovery system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220505 Address after: 528000 workshop A-2, No. 1, Park 3, Lutang Wucun, Chengbei Industrial Park, Luocun, No. 17, Beihu 1st Road, Luocun, Shishan town, Nanhai District, Foshan City, Guangdong Province Applicant after: Foshan Sanli Intelligent Equipment Technology Co.,Ltd. Address before: Room 213, Yijing building, No.1 Hengshan Road, Yantai Economic and Technological Development Zone, Shandong Province Applicant before: Shandong Yanhuang Industrial Design Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |