CN106548169A - Fuzzy literal Enhancement Method and device based on deep neural network - Google Patents
Fuzzy literal Enhancement Method and device based on deep neural network Download PDFInfo
- Publication number
- CN106548169A CN106548169A CN201610945012.0A CN201610945012A CN106548169A CN 106548169 A CN106548169 A CN 106548169A CN 201610945012 A CN201610945012 A CN 201610945012A CN 106548169 A CN106548169 A CN 106548169A
- Authority
- CN
- China
- Prior art keywords
- image block
- image
- neural network
- deep neural
- block
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of fuzzy literal Enhancement Method based on deep neural network, including：Set up reference database；Test image of the collection comprising word；Test image is divided into into multiple test image blocks by image block division rule；Index by target search of each test image block in reference database, filter out the multiple pre-set image blocks most like with test image block；It is restored image block by multiple most like pre-set image block Weighted Fusions according to fusion coefficients, the image adjacent restored image block Weighted Fusion of correspondence is obtained into restored image.The present invention also provides a kind of fuzzy literal intensifier based on deep neural network.Introduce deep neural network feature, improve the robustness of image block when reference data being set up with image block retrieval；Also clearly image can will be recovered to comprising fuzzy character image by the data base for training even at off-line state, is easy to show or recognize word in image, improves the resolution and definition of word in image.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of fuzzy literal increasing based on deep neural network
Strong method and device.
Background technology
With the development and the progress of science and technology of society, requirement more and more higher of the people to image processing techniquess.Based on image
In process, Text region is used as an important basic technology, with huge using value and wide application prospect, especially
It is the Text region of natural scene image.For example, Text region in image, OCR (Optical are carried out by OCR technique
Character Recognition, optical character identification) refer to that electronic equipment (such as scanner or digital camera) is checked on paper
The word of printing, determines its shape by the pattern for detecting dark, bright, shape is translated into computer with character recognition method then
The process of word.
Word is a kind of important information carrier, and according to incompletely statistics, the theme for still having 90% information resources at present is
There is provided by document information.With developing rapidly for scientific and technological information, these information automations are identified into that one becomes a kind of
Trend and focus.Word automatic identification rate in high-quality text image can reach more than 99%.
However, in the prior art, as the decline of picture quality, particularly image pixel be not high or image itself is unclear
The image blur phenomena brought by Chu, causes the discrimination of word also to decline therewith.
The content of the invention
The shortcoming of prior art in view of the above, it is an object of the invention to provide a kind of based on deep neural network
Fuzzy literal Enhancement Method and device, during for solving image blurring in prior art, it is impossible to accurately identify word in image
Problem.
For achieving the above object and other related purposes, the present invention provides a kind of fuzzy literal based on deep neural network
Enhanced method, including：
Set up reference database；
Test image of the collection comprising word；
The test image is divided into into multiple test image blocks by image block division rule；
Index by target search of test image block each described in the reference database, filter out and the test
The most like multiple pre-set image blocks of image block；
According to fusion coefficients by multiple most like pre-set image block Weighted Fusions be restored image block, by described image pair
Adjacent restored image block Weighted Fusion is answered to obtain restored image.
Another object of the present invention is to provide a kind of fuzzy literal based on deep neural network enhanced device, wrap
Include：
Reference database, for setting up reference database；
Acquisition module, for test image of the collection comprising word；
Processing module, for the test image is divided into multiple test image blocks by image block division rule；
Retrieval module, for indexing by target search of test image block each described in the reference database, sieve
Select the multiple pre-set image blocks most like with the test image block；
Fusion Module, for according to fusion coefficients by multiple most like pre-set image block Weighted Fusions be restored image
Described image correspondence adjacent restored image block Weighted Fusion is obtained restored image by block.
As described above, the fuzzy literal Enhancement Method based on deep neural network and device of the present invention, have with following
Beneficial effect：
By building reference database and identification model being trained in the data base, collection includes the test of word to the present invention
Test image is divided into multiple test image blocks by image, based on the deep neural network characteristic matching image in data base
The most like pre-set image block of block, the multiple most like pre-set image blocks of Weighted Fusion obtain restored image block, will by picture position
Adjacent restored image block is recovered to picture rich in detail.To introduce deep neural network special when reference data being set up with image block retrieval
Levy, improve the robustness of image block；The data base of training can also be passed through by comprising fuzzy word even at off-line state
Image restoration is easy to show or recognize word in image into clearly image, improves the resolution of word in image and clear
Degree.
Description of the drawings
Fig. 1 is shown as the present invention and provides a kind of fuzzy literal Enhancement Method flow chart based on deep neural network；
Fig. 2 is shown as the present invention and provides a kind of the detailed of step S1 in the fuzzy literal Enhancement Method based on deep neural network
Thin flow chart；
Fig. 3 is shown as image segmentation during the present invention provides a kind of fuzzy literal Enhancement Method based on deep neural network and shows
It is intended to；
Fig. 4 is shown as the present invention and provides a kind of the detailed of step S4 in the fuzzy literal Enhancement Method based on deep neural network
Thin flow chart；
Fig. 5 is shown as the present invention and provides a kind of the detailed of step S5 in the fuzzy literal Enhancement Method based on deep neural network
Thin flow chart；
Fig. 6 is shown as Cellular structure during the present invention provides a kind of fuzzy literal Enhancement Method based on deep neural network and shows
It is intended to；
Fig. 7 is shown as the first enforcement stream that the present invention provides a kind of fuzzy literal Enhancement Method based on deep neural network
Cheng Tu；
Fig. 8 is shown as the present invention and provides a kind of fuzzy literal intensifier structured flowchart based on deep neural network；
Fig. 9 is shown as the present invention and provides database structure in a kind of fuzzy literal intensifier based on deep neural network
Block diagram；
Figure 10 is shown as retrieving module during the present invention provides a kind of fuzzy literal intensifier based on deep neural network
Structured flowchart；
Figure 11 is shown as the present invention and provides Fusion Module in a kind of fuzzy literal intensifier based on deep neural network
Structured flowchart.
Specific embodiment
Embodiments of the present invention are illustrated below by way of specific instantiation, those skilled in the art can be by this specification
Disclosed content understands other advantages and effect of the present invention easily.The present invention can also pass through concrete realities different in addition
The mode of applying is carried out or applies, the every details in this specification can also based on different viewpoints with application, without departing from
Various modifications and changes are carried out under the spirit of the present invention.It should be noted that, in the case where not conflicting, following examples and enforcement
Feature in example can be mutually combined.
It should be noted that the diagram provided in following examples only illustrates the basic structure of the present invention in a schematic way
Think, the component relevant with the present invention is only shown then in schema rather than according to component count during actual enforcement, shape and size
Draw, which is actual when the implementing kenel of each component, quantity and ratio can be a kind of random change, and its assembly layout kenel
It is likely more complexity.
Embodiment 1
Fig. 1 is referred to, and a kind of flow chart of the fuzzy literal Enhancement Method based on deep neural network is provided for the present invention,
Including：
Step S1, sets up reference database；
Specifically, the purpose of reference database is set up in order to build a priori storehouse for word deblurring, specially
Door is used to aid in fuzzy image enhancement.
Step S2, test image of the collection comprising word；
Specifically, collection is fuzzy literal image (picture) to be restored comprising character image, and the word includes word
Row, line of text, character etc..
The test image is divided into multiple test image blocks by image block division rule by step S3；
Specifically, the test image normalized is obtained into standardized format first, is carried out according still further to pixel spot size
Uniform piecemeal is processed, and is divided into multiple test image blocks, wherein, image block division rule be each described image block press word and
Piecemeal position is identified.
Step S4, in the reference database with test image block each described as target search index, filter out with
The most like multiple pre-set image blocks of the test image block；
Specifically, to split the test image block of gained as target retrieval, enter according to above-mentioned target in reference database
Line retrieval, estimates for search mark according to the distance between pre-set image block in the test image block and reference database of target retrieval
Standard, the less representative of distance measure value are more similar.
Step S5, according to fusion coefficients by multiple most like pre-set image block Weighted Fusions be restored image block, by institute
State the adjacent restored image block Weighted Fusion of image correspondence and obtain restored image.
Specifically, most like multiple pre-set image blocks are weighted into fusion first, it is right according to institute in image to be restored
The equal Weighted Fusion of image block answered obtains its correspondence restored image block；All restored image blocks are pressed into pixel Weighted Fusion one by one
Obtain restored image.
In the present embodiment, introduce deep neural network feature when reference data being set up with image block retrieval, improve
The robustness of image block；Also will can be recovered to comprising fuzzy character image by the data base for training even at off-line state
Clearly image, is easy to show or recognize word in image, improves the resolution and definition of word in image.
Embodiment 2
Fig. 2 is shown as the present invention and provides a kind of the detailed of step S1 in the fuzzy literal Enhancement Method based on deep neural network
Thin flow chart；
Step S101, gathers word clearly image, and wherein each word image comprising multiple multi-forms；
Specifically, clearly image is high-quality picture to the word of collection, and in order to consider the coverage rate of word with
And picture number, the picture for being gathered should at least include Chinese characters in common use storehouse, secondary Chinese characters in common use storehouse and other common characters.Separately
Outward, it is considered to which it is different that same word may write (express) mode, so, each word image comprising multiple multi-forms,
At least ensure the word integrity as far as possible in priori storehouse, for supporting feature space when follow-up blurred picture strengthens.
Step S102, by described image normalized and is divided into multiple pre-set image blocks, wherein, described in each
Pre-set image block is identified with piecemeal position by word；
Specifically, by the word of collection, clearly image is normalized, normalization main to the size of image,
Gray processing, contrast enhancing etc., convert images into corresponding sole criterion form；Secondly, normalized image is carried out point
Block process, is divided into multiple pre-set image blocks, and the size (Block Size) of piecemeal process can be arranged on 10 to 40 pictures
Plain left and right, the half being preferably fixed as image block size herein, i.e., between adjacent image block, registration is 50%, and image
Displacement increment between block is wide, high 50% of the image block, as shown in Fig. 3 is aobvious, provides a kind of based on depth for the present invention
Image segmentation schematic diagram in the fuzzy literal Enhancement Method of neutral net；In figure, each pre-set image block size is 16*
16pix, pre-set image block have coincidence between 8pix and adjacent another pre-set image block, according to word and piecemeal position in image
Put and be identified jointly, i.e., the pre-set image block that same word and same position are partitioned into just is identified as same label, sentences
It is set to a class pre-set image block.
Step S103, adopts pre-set image described in convolutional neural networks Algorithm for Training based on softmaxwithloss structures
Block obtains deep learning module；
Specifically, it is clear using word and identify identical image block, optimized based on class object
Using convolutional neural networks Algorithm for Training, each pre-set image block obtains deep learning module to softmaxwithloss structures, its
In, the softmaxwithloss structures are as follows：
In formula (1), vectors of the z for full articulamentum output in convolutional neural networks, z=(z1,z2,…zn)；F (z) is
The output of softmax.
Step S104, is output as index with the deep learning module, sets up the identification model of reference database.
Specifically, for the deep learning module (deep neural network) for training, last full connection preferably by which
Output of the layer as characteristics of image modeling, models for image block and indexes；Before restored image, need to each word institute
Corresponding image block carries out deep neural network feature extraction, carries out feature modeling index, generates the knowledge with regard to reference database
Other model；When to needing the broad image for recovering to process, the feature of offline index is can be used directly, so as to convenient, fast
The maximum top n image block of correspondence similarity is found promptly.
In the present embodiment, by building reference database, the flow process of fuzzy literal image recovery is not only shortened, is improved
The efficiency that fuzzy literal image increases；Meanwhile, the deep neural network of introducing can increase substantially the robust of image block search
Property, improve the restorability of fuzzy literal image.
Embodiment 3
Fig. 4 is shown as the present invention and provides a kind of the detailed of step S4 in the fuzzy literal Enhancement Method based on deep neural network
Thin flow chart；
Step S401, extracts the deep neural network feature of each test image block；
Specifically, 10 most like with which pre-set image is retrieved in reference database for each test image block
Block, in search, is measured with L1 distances using degree neural network characteristics.
Step S402, calculates the corresponding deep neural network feature of each test image block and reference data as follows
The distance between deep neural network feature of pre-set image block in storehouse is estimated；
D (p, q)=| | p-q | | (2)
In formula (2), distance measures of the d (p, q) for deep neural network feature between test image block and pre-set image block,
P, q are respectively the CNN characteristic vectors of image block；
Formula (2) is specifically launched, equation below can be obtained：
D (p, q)=| p1-q1|+|p2-q2|+…|pn-qn|
Wherein, p, q are respectively the CNN characteristic vectors of image block；P=(p1,p2,…,pn) and q=(q1,q2,…,qn), often
Individual feature is n-dimensional vector.
Step S403, it is most like pre-set image block to filter out the minimum image block of multiple distance measure values.
Specifically, the 10 default figures minimum apart from measure value in reference database between target search image block
As block is most like pre-set image block.
In the present embodiment, by using deep neural network feature, which has higher sign ability, in search high definition
Image block when have more robust, for the blurred picture under complicated true environment has good restorability, improve fuzzy
Image restoration ability.
Embodiment 4
Fig. 5 is shown as the present invention and provides a kind of the detailed of step S5 in the fuzzy literal Enhancement Method based on deep neural network
Thin flow chart；
Step S501, by reflecting between each position in image corresponding image block to be restored and most like pre-set image block
The relation of penetrating is fixed as one to ten, and by ten most like pre-set image blocks, Weighted Fusion is calculated as follows, is restored
Image block；
In formula (3), f (x, y) be fusion after image block function, gkDuring (x, y) is the reference database that correspondence is searched
Most like pre-set image block, ω (xk) for fusion coefficients；
Specifically, the fusion coefficients can be expressed as following form：
Wherein, xiRepresent the characteristic parameter p of the fuzz testing image block and pre-set image block q of front ten for retrievingiFeature
Front ten similar pre-set image block is permeated restored image block by the inverse of parameter distance by image block function, therefore,
All test image blocks corresponding in the image of parked can be merged in a manner described, be obtained the image weighting fusion
Restored image block.
Each described restored image block is divided into four cell elements, as follows with cell element as substantially single by step S502
Position individual element Weighted Fusion obtains restored image；
In formula (4), g (x, y) is the cell element after final fusion, fk(x, y) is four overlapping cell elements of correspondence, and (x, y) is
The station location marker of cell element pixel, ωk(x, y) is weight coefficient.
Specifically, when being divided into image block due to image, the overlapping ratio of setting is 50%, therefore, will be every in image
Individual image block is divided into four cell elements in a manner described, as shown in fig. 6, providing a kind of based on deep neural network for the present invention
Cellular structure schematic diagram in fuzzy literal Enhancement Method；In figure, the image block of 16*16pix is divided into the born of the same parents of four 8*8pix
Unit, is weighted fusion according to formula (4), wherein, ωk(x, y) for the concrete calculation of weight coefficient is：
Wherein, | | Pk(x,y)||2For Euclidean distance formula, be on correspondence cell element point (x, y) in correspondence image block
The distance of the heart (x', y'), ω 'k(x, y) is the value after weight coefficient normalization, adopts which finally to merge weight coefficient.
In the present embodiment, by being weighted fusion respectively to image block and cell element, and in first time Weighted Fusion
On the basis of carry out second Weighted Fusion again, improve the definition of fuzzy literal, be easy to the identification of later stage character image.
Embodiment 5
Fig. 7 is shown as the first enforcement stream that the present invention provides a kind of fuzzy literal Enhancement Method based on deep neural network
Cheng Tu, including：Word unclear " heresy " word image is restored, first, successively its normalized is processed with piecemeal,
Obtain 16 test image blocks；Location difference according to image is identified to 16 test image blocks respectively, such as：Figure
Picture block 1 is retrieved in reference database corresponding to image block 16, respectively the test image block with mark as target retrieval
This 10 most like pre-set image blocks are weighted fusion by 10 most like pre-set image blocks, respectively obtain correspondence survey
The restored image block (image block of denoising) of examination image block (image block 1 is to image block 16)；By " heresy " word image correspondence restored map
As block obtains " heresy " word image denoising after for ultimate unit individual element Weighted Fusion by cell element, as shown in Figure 7, it will be apparent that
The definition of fuzzy literal is increased, is easy to visual understanding；Meanwhile, for the image block comprising literal line, also can be according to we
Method carries out the recovery of fuzzy literal.
Embodiment
Fig. 8 is shown as the present invention and provides a kind of fuzzy literal intensifier structured flowchart based on deep neural network；Bag
Include：
Reference database 1, for setting up reference database；
Specifically, the purpose of reference database is set up in order to build a priori storehouse for word deblurring, specially
Door is used to aid in fuzzy image enhancement.
Acquisition module 2, for test image of the collection comprising word；
Specifically, collection is fuzzy literal test image (picture) to be restored comprising character image, and the word is included
Literal line, line of text, character etc..
Processing module 3, for the test image is divided into multiple test image blocks by image block division rule；
Specifically, the test image normalized is obtained into standardized format first, is carried out according still further to pixel spot size
Uniform piecemeal is processed, and is divided into multiple test image blocks, wherein, image block division rule be each described image block press word and
Piecemeal position is identified.
Retrieval module 4, for indexing by target search of test image block each described in the reference database, sieve
Select the multiple pre-set image blocks most like with described image block；
Specifically, to split the test image block of gained as target retrieval, enter according to above-mentioned target in reference database
Line retrieval, estimates for search mark according to the distance between pre-set image block in the test image block and reference database of target retrieval
Standard, the less representative of distance measure value are more similar.
Fusion Module 5, for according to fusion coefficients by multiple most like pre-set image block Weighted Fusions be restored image
Described image correspondence adjacent restored image block Weighted Fusion is obtained restored image by block.
Specifically, most like multiple pre-set image blocks are weighted into fusion first, it is right according to institute in image to be restored
The equal Weighted Fusion of image block answered obtains its correspondence restored image block；All restored image blocks are weighted one by one according to pixel and is melted
Conjunction obtains restored image.
In the present embodiment, introduce deep neural network feature when reference data being set up with image block retrieval, improve
The robustness of image block；Also will can be recovered to comprising fuzzy character image by the data base for training even at off-line state
Clearly image, is easy to show or recognize word in image, improves the resolution and definition of word in image.
Fig. 9 is shown as the present invention and provides database structure in a kind of fuzzy literal intensifier based on deep neural network
Block diagram, including：
Collecting unit 11, for gathering word clearly image, and wherein each word includes multiple multi-forms
Image；
Specifically, clearly image is high-quality picture to the word of collection, and in order to consider word coverage rate and
Picture number, the picture for being gathered should at least include Chinese characters in common use storehouse, secondary Chinese characters in common use storehouse and other common characters, in addition,
Consider that same word may write (express) mode difference, so, each word image comprising multiple multi-forms, at least
Ensure the word integrity as far as possible in priori storehouse, for supporting feature space when follow-up blurred picture strengthens.
Processing unit 12, by described image normalized and is divided into multiple pre-set image blocks, wherein, each institute
State pre-set image block to be identified with piecemeal position by word；
Specifically, by the word of collection, clearly image is normalized, normalization main to the size of image,
Gray processing, contrast enhancing etc., convert images into corresponding sole criterion form；Secondly, normalized image is carried out point
Block process, is divided into multiple pre-set image blocks, and the size (Block Size) of piecemeal process can be arranged on 10 to 40 pictures
Plain left and right, the half being preferably fixed as image block size herein, i.e., between adjacent pre-set image block, registration is 50%, and
Displacement increment between pre-set image block is wide, high 50% of the image block, as shown in Fig. 3 is aobvious, provides one kind for the present invention
Based on image segmentation schematic diagram in the fuzzy literal Enhancement Method of deep neural network；In figure, each pre-set image block size is
16*16pix, image block have coincidence between 8pix and adjacent another pre-set image block, according to word and piecemeal position in image
Put and be identified jointly, i.e., the image block that same word and same position are partitioned into just is identified as same label, is judged to
One class image block.
Training unit 13, for adopting pre- described in convolutional neural networks Algorithm for Training based on softmaxwithloss structures
If image block obtains deep learning module；
Specifically, it is clear using word and identify identical image block, optimized based on class object
Using convolutional neural networks Algorithm for Training, each pre-set image block obtains deep learning module to softmaxwithloss structures, its
In, the softmaxwithloss structures are as follows：
In formula (1), vectors of the z for full articulamentum output in convolutional neural networks, z=(z1,z2,…zn)；F (z) is
The output of softmax.
Model Identification unit 14, is output as index with the deep learning module, sets up the identification mould of reference database
Type.
Specifically, for the deep learning module (deep neural network) for training, last full connection preferably by which
Output of the layer as characteristics of image modeling, models for image block and indexes；Before restored image, need to each word institute
Corresponding image block carries out deep neural network feature extraction, carries out feature modeling index, generates the knowledge with regard to reference database
Other model；When to needing the broad image for recovering to process, the feature of offline index is can be used directly, so as to convenient, fast
The maximum top n image block of correspondence similarity is found promptly.
In the present embodiment, by building reference database, the flow process of fuzzy literal image recovery is not only shortened, is improved
The efficiency that fuzzy literal image increases；Meanwhile, the deep neural network of introducing can increase substantially the robust of image block search
Property, improve the restorability of fuzzy literal image.
Figure 10 is shown as retrieving module during the present invention provides a kind of fuzzy literal intensifier based on deep neural network
Structured flowchart, including：
Extraction unit 41, for extracting the deep neural network feature of each test image block；
Specifically, 10 most like with which pre-set image is retrieved in reference database for each test image block
Block, in search, is measured with L1 distances using degree neural network characteristics.
Computing unit 42, for calculating the corresponding deep neural network feature of each test image block as follows with ginseng
Examine the distance between deep neural network feature of pre-set image block in data base to estimate,
D (p, q)=| | p-q | | (2)
In formula (2), distance measures of the d (p, q) for deep neural network feature between test image block and pre-set image block,
P, q are respectively the CNN characteristic vectors of image block；
Formula (2) is specifically launched, equation below can be obtained：
D (p, q)=| p1-q1|+|p2-q2|+…|pn-qn|
Wherein, p, q are respectively the CNN characteristic vectors of image block；P=(p1,p2,…,pn) and q=(q1,q2,…,qn), often
Individual feature is n-dimensional vector
Screening unit 43, is most like pre-set image block for filtering out the minimum image block of multiple distance measure values.
Specifically, the 10 default figures minimum apart from measure value in reference database between target search image block
As block is most like pre-set image block.
In the present embodiment, by using deep neural network feature, which has higher sign ability, in search high definition
Image block when have more robust, for the blurred picture under complicated true environment has good restorability, improve fuzzy
Image restoration ability.
Figure 11 is shown as the present invention and provides Fusion Module in a kind of fuzzy literal intensifier based on deep neural network
Structured flowchart, including：
First integrated unit 51, for by each position in image corresponding image block to be restored and most like pre-set image
Mapping relations between block are fixed as one to ten, and by ten most like image blocks, Weighted Fusion is calculated as follows, is obtained
To restored image block；
In formula (3), f (x, y) be fusion after image block function, gkDuring (x, y) is the reference database that correspondence is searched
Most like image block, ω (xk) for fusion coefficients；
Specifically, the fusion coefficients can be expressed as following form：
Wherein, xiRepresent the characteristic parameter p of the fuzz testing image block and pre-set image block q of front ten for retrievingiFeature
Front ten similar image block is permeated restored image block by the inverse of parameter distance by image block function, therefore, can be by
In the image of parked, corresponding all test image blocks are merged in a manner described, obtain the figure of the image weighting fusion
As block.
Second integrated unit 52, for each described restored image block is divided into four cell elements, as follows with born of the same parents
Unit obtains restored image for ultimate unit individual element Weighted Fusion；
In formula (4), g (x, y) is the cell element after final fusion, fk(x, y) is four overlapping cell elements of correspondence, and (x, y) is
The station location marker of cell element pixel, ωk(x, y) is weight coefficient.
Specifically, when being divided into image block due to image, the overlapping ratio of setting is 50%, therefore, will be every in image
Individual restored image block is divided into four cell elements in the manner described above, as shown in fig. 7, providing a kind of based on depth nerve for the present invention
Cellular structure schematic diagram in the fuzzy literal Enhancement Method of network；In figure, the image block of 16*16pix is divided into four 8*8pix
Cell element, be weighted fusion according to formula (4), wherein, ωk(x, y) for the concrete calculation of weight coefficient is：
Wherein, | | Pk(x,y)||2For Euclidean distance formula, be on correspondence cell element point (x, y) in correspondence image block
The distance of the heart (x', y'), ω 'k(x, y) is the value after weight coefficient normalization, adopts which finally to merge weight coefficient.
In the present embodiment, by being weighted fusion respectively to image block and cell element, and in first time Weighted Fusion
On the basis of carry out second Weighted Fusion again, improve the definition of fuzzy literal, be easy to the identification of later stage character image.
In sum, the present invention is by building reference database and in the data base, training identification model, collection to include
Test image is divided into multiple image blocks to be tested by the test image of word, special based on deep neural network in data base
Levy and match the most like pre-set image block of the image block, the multiple most like pre-set image blocks of Weighted Fusion obtain restored image block,
Adjacent restored image block is recovered to into picture rich in detail by picture position.When reference data being set up with image block retrieval introduce depth
Neural network characteristics, improve the robustness of image block；The data base that training can also be passed through even at off-line state will include
Fuzzy character image is recovered to clearly image, is easy to show or recognize word in image, improves the knowledge of word in image
Not Du and definition.So, the present invention effectively overcomes various shortcoming of the prior art and has high industrial utilization.
The principle and its effect of above-described embodiment only illustrative present invention, it is of the invention not for limiting.It is any ripe
The personage for knowing this technology all can carry out modifications and changes to above-described embodiment under the spirit and the scope without prejudice to the present invention.Cause
This, those of ordinary skill in the art is complete with institute under technological thought without departing from disclosed spirit such as
Into all equivalent modifications or change, should by the present invention claim be covered.
Claims (10)
1. a kind of enhanced method of fuzzy literal based on deep neural network, it is characterised in that include：
Set up reference database；
Test image of the collection comprising word；
The test image is divided into into multiple test image blocks by image block division rule；
Index by target search of test image block each described in the reference database, filter out and the test image
The most like multiple pre-set image blocks of block；
According to fusion coefficients by multiple most like pre-set image block Weighted Fusions be restored image block, by described image correspondence phase
Adjacent restored image block Weighted Fusion obtains restored image.
2. the enhanced method of the fuzzy literal based on deep neural network according to claim 1, it is characterised in that described
The step of setting up reference database, including：
Gather word clearly image, and wherein each word image comprising multiple multi-forms；
By described image normalized and multiple pre-set image blocks are divided into, wherein, each described pre-set image block is pressed
Word is identified with piecemeal position；
Deep learning is obtained using pre-set image block described in convolutional neural networks Algorithm for Training based on softmaxwithloss structures
Module；
Index is output as with the deep learning module, the identification model of reference database is set up.
3. the enhanced method of the fuzzy literal based on deep neural network according to claim 2, it is characterised in that described
Deep learning module is obtained using pre-set image block described in convolutional neural networks Algorithm for Training based on softmaxwithloss structures
The step of, including：
It is clear using word and identify identical pre-set image block, based on the softmaxwithloss structures that class object optimizes
Using convolutional neural networks Algorithm for Training, each image block obtains deep learning module, wherein, the softmaxwithloss knots
Structure is as follows：
In formula (1), vectors of the z for full articulamentum output in convolutional neural networks, z=(z1,z2,…zn)；F (z) is softmax
Output.
4. the enhanced method of the fuzzy literal based on deep neural network according to claim 1, it is characterised in that described
Index by target search of test image block each described in the reference database, filter out most like with described image block
Multiple pre-set image blocks the step of, including：
The deep neural network feature of each test image block is extracted, each corresponding depth of test image block is calculated as follows
The distance between deep neural network feature of pre-set image block in degree neural network characteristics and reference database is estimated, screening
It is most like pre-set image block to go out the minimum image block of multiple distance measure values；
D (p, q)=| | p-q | | (2)
In formula (2), distance measures of the d (p, q) for deep neural network feature between test image block and pre-set image block, p, q
Respectively CNN characteristic vectors of image block.
5. the enhanced method of the fuzzy literal based on deep neural network according to claim 1, it is characterised in that described
According to fusion coefficients by multiple most like pre-set image block Weighted Fusions be restored image block, will described image correspondence it is adjacent
The step of restored image block Weighted Fusion obtains restored image, including：
Mapping relations between each position in image corresponding image block to be restored and most like pre-set image block are fixed as
One to ten, by ten most like image blocks, Weighted Fusion is calculated as follows, obtains restored image block；
In formula (3), f (x, y) be fusion after image block function, gk(x, y) is most like in the reference database that searches of correspondence
Pre-set image block, ω (xk) for fusion coefficients；
Each described restored image block is divided into into four cell elements, is weighted by ultimate unit individual element of cell element as follows
Fusion obtains restored image；
In formula (4), g (x, y) is the cell element after final fusion, fk(x, y) is four overlapping cell elements of correspondence, and (x, y) is cell element picture
The station location marker of vegetarian refreshments, ωk(x, y) is weight coefficient.
6. the enhanced device of a kind of fuzzy literal based on deep neural network, it is characterised in that include：
Reference database, for setting up reference database；
Acquisition module, for test image of the collection comprising word；
Processing module, for the test image is divided into multiple test image blocks by image block division rule；
Retrieval module, for indexing by target search of test image block each described in the reference database, filters out
The multiple pre-set image blocks most like with the test image block；
Fusion Module, for according to fusion coefficients by multiple most like pre-set image block Weighted Fusions be restored image block, will
The adjacent restored image block Weighted Fusion of described image correspondence obtains restored image.
7. the enhanced device of the fuzzy literal based on deep neural network according to claim 6, it is characterised in that described
Reference database includes：
Collecting unit, for gathering word clearly image, and wherein each word image comprising multiple multi-forms；
Processing unit, by described image normalized and is divided into multiple pre-set image blocks, wherein, each is described default
Image block is identified with piecemeal position by word；
Training unit, for adopting pre-set image described in convolutional neural networks Algorithm for Training based on softmaxwithloss structures
Block obtains deep learning module；
Model Identification unit, is output as index with the deep learning module, sets up the identification model of reference database.
8. the enhanced device of the fuzzy literal based on deep neural network according to claim 7, it is characterised in that described
Training unit is specifically included：
It is clear using word and identify identical pre-set image block, based on the softmaxwithloss structures that class object optimizes
Using convolutional neural networks Algorithm for Training, each pre-set image block obtains deep learning module, wherein, it is described
Softmaxwithloss structures are as follows：
In formula (1), vectors of the z for full articulamentum output in convolutional neural networks, z=(z1,z2,…zn)；F (z) is softmax
Output.
9. the enhanced device of the fuzzy literal based on deep neural network according to claim 6, it is characterised in that described
Retrieval module is specifically included：
Extraction unit, for extracting the deep neural network feature of each test image block；
Computing unit, for calculating the corresponding deep neural network feature of each test image block and reference data as follows
The distance between deep neural network feature of pre-set image block in storehouse is estimated,
D (p, q)=| | p-q | | (2)
In formula (2), distance measures of the d (p, q) for deep neural network feature between test image block and pre-set image block, p, q
Respectively CNN characteristic vectors of image block；
Screening unit, is most like image block for filtering out the minimum image block of multiple distance measure values.
10. the enhanced device of the fuzzy literal based on deep neural network according to claim 6, it is characterised in that institute
State Fusion Module to specifically include：
First integrated unit, for by between each position in image corresponding image block to be restored and most like pre-set image block
Mapping relations be fixed as one to ten, by ten most like image blocks, Weighted Fusion is calculated as follows, is restored
Image block；
In formula (3), f (x, y) be fusion after image block function, gk(x, y) is most like in the reference database that searches of correspondence
Pre-set image block, ω (xk) for fusion coefficients；
Second integrated unit, for each described restored image block is divided into four cell elements, as follows with cell element as base
Our unit's individual element Weighted Fusion obtains restored image；
In formula (4), g (x, y) is the cell element after final fusion, fk(x, y) is four overlapping cell elements of correspondence, and (x, y) is cell element picture
The station location marker of vegetarian refreshments, ωk(x, y) is weight coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610945012.0A CN106548169B (en) | 2016-11-02 | 2016-11-02 | Fuzzy literal Enhancement Method and device based on deep neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610945012.0A CN106548169B (en) | 2016-11-02 | 2016-11-02 | Fuzzy literal Enhancement Method and device based on deep neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106548169A true CN106548169A (en) | 2017-03-29 |
| CN106548169B CN106548169B (en) | 2019-04-23 |
Family
ID=58393693
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610945012.0A Active CN106548169B (en) | 2016-11-02 | 2016-11-02 | Fuzzy literal Enhancement Method and device based on deep neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106548169B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609560A (en) * | 2017-09-27 | 2018-01-19 | 北京小米移动软件有限公司 | Character recognition method and device |
| CN107609195A (en) * | 2017-10-18 | 2018-01-19 | 广东小天才科技有限公司 | One kind searches topic method and device |
| CN108230253A (en) * | 2017-05-08 | 2018-06-29 | 北京市商汤科技开发有限公司 | Image recovery method, device, electronic equipment and computer storage media |
| CN108921825A (en) * | 2018-06-12 | 2018-11-30 | 北京羽医甘蓝信息技术有限公司 | The method and device of the facial skin points shape defect of detection based on deep learning |
| CN109107161A (en) * | 2018-08-17 | 2019-01-01 | 深圳市腾讯网络信息技术有限公司 | A kind of control method of game object, device, medium and equipment |
| CN110557572A (en) * | 2018-05-31 | 2019-12-10 | 杭州海康威视数字技术股份有限公司 | image processing method and device and convolutional neural network system |
| RU2709661C1 (en) * | 2018-09-19 | 2019-12-19 | Общество с ограниченной ответственностью "Аби Продакшн" | Training neural networks for image processing using synthetic photorealistic containing image signs |
| CN111223058A (en) * | 2019-12-27 | 2020-06-02 | 杭州雄迈集成电路技术股份有限公司 | Image enhancement method |
| CN111598796A (en) * | 2020-04-27 | 2020-08-28 | Oppo广东移动通信有限公司 | Image processing method and device, electronic device and storage medium |
| WO2022100490A1 (en) * | 2020-11-16 | 2022-05-19 | Huawei Technologies Co.,Ltd. | Methods and systems for deblurring blurry images |
| CN114581686A (en) * | 2022-01-26 | 2022-06-03 | 南京富岛油气智控科技有限公司 | Oil pumping working condition fusion reasoning identification method based on indicator diagram Hash search |
| CN117152022A (en) * | 2023-10-25 | 2023-12-01 | 荣耀终端有限公司 | Image processing method and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427249A (en) * | 2015-10-30 | 2016-03-23 | 南京邮电大学 | Wind power image quality enhancing method based on robustness nuclear norm regular regression |
| CN105574820A (en) * | 2015-12-04 | 2016-05-11 | 南京云石医疗科技有限公司 | Deep learning-based adaptive ultrasound image enhancement method |
| CN105718912A (en) * | 2016-01-26 | 2016-06-29 | 浙江捷尚视觉科技股份有限公司 | Vehicle characteristic object detection method based on deep learning |
| CN105931180A (en) * | 2016-04-12 | 2016-09-07 | 西安电子科技大学 | Salient information guided image irregular mosaic splicing method |
-
2016
- 2016-11-02 CN CN201610945012.0A patent/CN106548169B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427249A (en) * | 2015-10-30 | 2016-03-23 | 南京邮电大学 | Wind power image quality enhancing method based on robustness nuclear norm regular regression |
| CN105574820A (en) * | 2015-12-04 | 2016-05-11 | 南京云石医疗科技有限公司 | Deep learning-based adaptive ultrasound image enhancement method |
| CN105718912A (en) * | 2016-01-26 | 2016-06-29 | 浙江捷尚视觉科技股份有限公司 | Vehicle characteristic object detection method based on deep learning |
| CN105931180A (en) * | 2016-04-12 | 2016-09-07 | 西安电子科技大学 | Salient information guided image irregular mosaic splicing method |
Non-Patent Citations (1)
| Title |
|---|
| 吴炜: "《基于学习的图像增强技术》", 28 February 2013, 西安电子科技大学出版社 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230253A (en) * | 2017-05-08 | 2018-06-29 | 北京市商汤科技开发有限公司 | Image recovery method, device, electronic equipment and computer storage media |
| CN108230253B (en) * | 2017-05-08 | 2020-11-27 | 北京市商汤科技开发有限公司 | Image restoration method and device, electronic equipment and computer storage medium |
| CN107609560A (en) * | 2017-09-27 | 2018-01-19 | 北京小米移动软件有限公司 | Character recognition method and device |
| CN107609195A (en) * | 2017-10-18 | 2018-01-19 | 广东小天才科技有限公司 | One kind searches topic method and device |
| CN110557572A (en) * | 2018-05-31 | 2019-12-10 | 杭州海康威视数字技术股份有限公司 | image processing method and device and convolutional neural network system |
| CN110557572B (en) * | 2018-05-31 | 2021-04-27 | 杭州海康威视数字技术股份有限公司 | Image processing method and device and convolutional neural network system |
| CN108921825A (en) * | 2018-06-12 | 2018-11-30 | 北京羽医甘蓝信息技术有限公司 | The method and device of the facial skin points shape defect of detection based on deep learning |
| CN109107161A (en) * | 2018-08-17 | 2019-01-01 | 深圳市腾讯网络信息技术有限公司 | A kind of control method of game object, device, medium and equipment |
| RU2709661C1 (en) * | 2018-09-19 | 2019-12-19 | Общество с ограниченной ответственностью "Аби Продакшн" | Training neural networks for image processing using synthetic photorealistic containing image signs |
| CN111223058B (en) * | 2019-12-27 | 2023-07-18 | 杭州雄迈集成电路技术股份有限公司 | Image enhancement method |
| CN111223058A (en) * | 2019-12-27 | 2020-06-02 | 杭州雄迈集成电路技术股份有限公司 | Image enhancement method |
| CN111598796A (en) * | 2020-04-27 | 2020-08-28 | Oppo广东移动通信有限公司 | Image processing method and device, electronic device and storage medium |
| CN111598796B (en) * | 2020-04-27 | 2023-09-05 | Oppo广东移动通信有限公司 | Image processing method and device, electronic equipment and storage medium |
| WO2022100490A1 (en) * | 2020-11-16 | 2022-05-19 | Huawei Technologies Co.,Ltd. | Methods and systems for deblurring blurry images |
| US11741579B2 (en) | 2020-11-16 | 2023-08-29 | Huawei Technologies Co., Ltd. | Methods and systems for deblurring blurry images |
| CN114581686A (en) * | 2022-01-26 | 2022-06-03 | 南京富岛油气智控科技有限公司 | Oil pumping working condition fusion reasoning identification method based on indicator diagram Hash search |
| CN117152022A (en) * | 2023-10-25 | 2023-12-01 | 荣耀终端有限公司 | Image processing method and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106548169B (en) | 2019-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106548169B (en) | Fuzzy literal Enhancement Method and device based on deep neural network | |
| CN110929607B (en) | Remote sensing identification method and system for urban building construction progress | |
| CN108960211B (en) | Multi-target human body posture detection method and system | |
| CN108596108B (en) | Aerial remote sensing image change detection method based on triple semantic relation learning | |
| CN110163213B (en) | Remote sensing image segmentation method based on disparity map and multi-scale depth network model | |
| CN113065558A (en) | Lightweight small target detection method combined with attention mechanism | |
| CN111160249A (en) | Multi-class target detection method of optical remote sensing image based on cross-scale feature fusion | |
| CN112967341B (en) | Indoor visual positioning method, system, equipment and storage medium based on live-action image | |
| CN105160310A (en) | 3D (three-dimensional) convolutional neural network based human body behavior recognition method | |
| CN104299006A (en) | Vehicle license plate recognition method based on deep neural network | |
| CN108710893A (en) | A kind of digital image cameras source model sorting technique of feature based fusion | |
| CN114943893B (en) | Feature enhancement method for land coverage classification | |
| CN111222519A (en) | Construction method, method and device of hierarchical colored drawing manuscript line extraction model | |
| CN111414954A (en) | Rock image retrieval method and system | |
| CN113887472A (en) | Remote sensing image cloud detection method based on cascade color and texture feature attention | |
| CN114463619B (en) | Infrared dim target detection method based on integrated fusion features | |
| CN113435407A (en) | Small target identification method and device for power transmission system | |
| CN113610024B (en) | Multi-strategy deep learning remote sensing image small target detection method | |
| CN109741351A (en) | A kind of classification responsive type edge detection method based on deep learning | |
| CN106228553A (en) | High-resolution remote sensing image shadow Detection apparatus and method | |
| CN115115947A (en) | Remote sensing image detection method and device, electronic equipment and storage medium | |
| CN111382685B (en) | Scene recognition method and system based on deep learning | |
| Yuan et al. | Graph neural network based multi-feature fusion for building change detection | |
| CN113628113A (en) | Image splicing method and related equipment thereof | |
| CN113609913A (en) | Pine wood nematode disease tree detection method based on sampling threshold interval weighting |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 401122 5 stories, Block 106, West Jinkai Avenue, Yubei District, Chongqing Applicant after: Chongqing Zhongke Yuncong Technology Co., Ltd. Address before: 401122 Central Sixth Floor of Mercury Science and Technology Building B, Central Section of Huangshan Avenue, Northern New District of Chongqing Applicant before: CHONGQING ZHONGKE YUNCONG TECHNOLOGY CO., LTD. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |