CN1402191A - Multiple focussing image fusion method based on block dividing - Google Patents
Multiple focussing image fusion method based on block dividing Download PDFInfo
- Publication number
- CN1402191A CN1402191A CN 02137055 CN02137055A CN1402191A CN 1402191 A CN1402191 A CN 1402191A CN 02137055 CN02137055 CN 02137055 CN 02137055 A CN02137055 A CN 02137055A CN 1402191 A CN1402191 A CN 1402191A
- Authority
- CN
- China
- Prior art keywords
- image
- zone
- region
- piece
- pixel
- 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
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
A method based on block division for the fusion of multiple focused images includes dividing the original focused images into blocks with same sizes, finding out the local contrast of each block to reflect the difference between clear focusing area and fuzzy focusing area, dividing an image into clear region and fuzzy region, defining the blocks adjacent to the clear region and fuzzy region as boundary region, expressing said three regions, directly using the clear region as the fused region and using pixel fusion method to fuse the boundary region. Its advantage is high image quality.
Description
Technical field:
The present invention relates to a kind of multi-focus image fusing method of cutting apart based on piece, is a multiple focussing image information fusion method in the information fusion field, all is widely used in systems such as optical imagery, targeted surveillance, safety inspection.
Background technology:
Image fusion technology is the fusion of visual information in the multi-sensor information fusion, it utilizes the different imaging mode of various imaging sensors, for different images provides complementary information, increase amount of image information, reduce the raw image data amount, raising is to the adaptability of environment, and is more reliable to obtain, useful information is for observing or further handling more accurately.It is an emerging technology that combines sensor, signal Processing, Flame Image Process and artificial intelligence etc.In recent years, image co-registration has become a kind of very important and useful graphical analysis and computer vision technique.It has a wide range of applications in fields such as automatic target identification, computer vision, remote sensing, robot, Medical Image Processing and Military Application.
Multiple focussing image as one of image co-registration research contents merges, and it is meant under identical image-forming condition, and a plurality of images that the lens focus target is different can obtain all targets by image co-registration and all focus on fused image clearly.In the fusion method of handling multiple focussing image, representative method is many resolution images fusion method.Its basic thought is exactly obtain respectively they being merged the image multi-resolution representation that computing obtains a fusion on the basis that different resolution represents in that input original image is decomposed, and obtains fused image through multiresolution reconstruct.Yet adopt the multi-resolution image fusion method that the burnt original image of poly is carried out the resulting fusion results of fusion treatment, compare with the clear area of original image, the picture quality of its corresponding region decreases; And compare with the fuzzy region of original image, the picture quality of its corresponding region is improved, this that is to say that the multi-resolution image fusion method promotes the picture quality in image blurring zone to obtain all fusion results of " clear " of target by the picture quality that reduces the clear picture zone.There are deviation to a certain degree in its fusion results and desirable fusion results, and cause losing of some marginal informations in the image.
Summary of the invention:
The objective of the invention is to the deficiency that exists at prior art, a kind of multi-focus image fusing method of cutting apart based on piece is provided, can improve the picture quality after the fusion, reach desirable practical function.
For realizing such purpose, the innovative point of technical solution of the present invention is image is carried out area dividing and makes corresponding fusion treatment.The burnt input original image of poly is being divided on the basis in several equal-sized zone, after input original image is carried out non-down-sampled wavelet decomposition, obtain on the basis of low frequency component, vertical high frequency component, horizontal high fdrequency component and diagonal high-frequency components of image, the mean value of trying to achieve with the high fdrequency component and the absolute value sum of the ratio of low frequency component of each point in the piece zone is as this regional local contrast.Reflect the image focusing clear area and focus on difference between the fuzzy region with this.When carrying out each area dividing of image, at first utilize piece zone local contrast that entire image is divided into clear area and fuzzy region, and then all piece zones that the clear area is adjacent with fuzzy region divide borderline region into, obtain three different area dividing of image and represent with the form of image-region signature with this.For clear area and fuzzy region,, when carrying out fusion treatment, directly choose clear zone as the relevant block zone after merging because input original image is complementary in these two zones.For borderline region, at first ask for the wherein interior low frequency component of the residing neighborhood of each pixel, absolute value sum with the high fdrequency component of pixel and the ratio of low frequency component is the contrast of pixel, choosing the bigger pixel value of each point contrast sum that respective pixel is put in its neighborhood in the input picture borderline region at last is the pixel value of this point after merging, and so adopts this method of choosing based on pixel to handle borderline region.
A kind of multi-focus image fusing method of cutting apart based on piece of the present invention comprises following concrete steps:
1. after burnt input original image is divided into several equal-sized zone with poly, the computed image piece
The local contrast in zone.After input original image is carried out non-down-sampled wavelet decomposition, obtain figure
The low frequency component of picture, vertical high frequency component, horizontal high fdrequency component and diagonal high-frequency components, right
After with the high fdrequency component of each point in the piece zone and the absolute value sum of the ratio of low frequency component try to achieve flat
Average is as this regional local contrast.With this reflect the fuzzy of input original image piece zone and
The difference of readability.Distinguish the clear journey in piece zone by the local contrast size in piece zone
Degree.The readability in piece zone is high more, and its local contrast is big more, otherwise the piece zone is fuzzy more,
Its contrast is more little.Also can adopt the big of the average gradient in piece zone or information entropy that it comprises in addition
The little readability of distinguishing the piece zone.
2. by comparing the size of corresponding region contrast of input picture, entire image can be divided into clearly
Clear zone and fuzzy region.What the piece region contrast was big is clear zone; What contrast was little is mould
Paste piece zone.Yet, because the influence of factors such as actual imaging makes the individual blocks zone to occur
The division of mistake.Do following processing for this reason:
1). according to the size of input picture, make that choosing of image block areas should not be too little, be generally
32*32、32*16、16*32、16*16；
2). image is lined by line scan by the piece zone, find out as yet the not piece zone of ownership;
3). with this zone is centre retrieves neighbor around it, belong to together a class block type with them
Merge;
4). the zone with new merging is the center, execution in step 2), the neighborhood in retrieving novel zone is up to the district
The territory can not be expanded;
5). return step 1), up to the piece zone of not finding not have ownership;
6). it comprises the quantity of piece to obtaining each zone calculating, when less than a certain number of (3 or 5),
Change the affiliated type in piece zone in this zone, think that these piece zones are wrong choosing
The piece zone; When a certain number of, the affiliated type in piece zone does not become in this zone
Change.
And then all piece zones that the clear area is adjacent with fuzzy region divide borderline region into, obtain three different area dividing of image with this.After above-mentioned processing, just can obtain merging required image-region signature.
3. after obtaining the image-region signature, can carry out melting of image block respectively at dissimilar zones
Close processing.For clear area and fuzzy region, because input original image is mutual in these two zones
Mend, i.e. the fuzzy region of the clear area correspondence image B of image A, otherwise, image A
The clear area of fuzzy region correspondence image B.When carrying out fusion treatment, directly choose clear
The zone is as the relevant block zone after merging.For borderline region, the picture based on contrast has been proposed
Element is chosen fusion method and is carried out fusion treatment.Specific as follows:
1). to each pixel in the borderline region, ask for the low frequency component in its residing neighborhood;
2). obtain the contrast of this pixel.Cutting off of the ratio of the high fdrequency component of this pixel and low frequency component
To the value sum is the contrast of this pixel;
3). choose in the borderline region in the input picture each point contrast sum in the respective pixel vertex neighborhood
Big pixel value is this pixel value after merging.
4). consider the correlativity of neighbor, to the arbitrary pixel in the image, if its adjacent pixels is equal
Be selected from another input original image, choosing of this pixel will be identical with choosing of neighbor so.Image interfusion method of the present invention has following beneficial effect:
On the basis that image-region is divided, carry out different fusion treatment at different zones.For clear area and fuzzy region, directly choose the clear area as merging corresponding zone, back, this makes does not introduce any deviation in this part zone after merging, and can access the optimal image effect; Borderline region for image, employing is chosen fusion method based on the pixel of contrast and is carried out fusion treatment, can keep image edge information well like this, natural in combination, level and smooth with the clear area of image simultaneously, whole fused image more approaches desirable fusion results.The multi-focus image fusing method that employing is cut apart based on piece has improved the fused image quality greatly, for the subsequent treatment and the significant and practical value of image demonstration of application system.
Description of drawings:
Fig. 1 the present invention is based on the multi-focus image fusing method synoptic diagram that piece is cut apart.
As shown in the figure, in that the burnt input original image of poly is being divided on the basis in several equal-sized zone,, entire image can be divided into clear area, fuzzy region and borderline region by comparing the size of corresponding region contrast of input picture.After obtaining the image-region signature, carry out the fusion treatment of image block respectively at dissimilar zones.
Fig. 2 is for focusing on different piece area image contrast figure.
Wherein, Fig. 2 (a), Fig. 2 (b) are first group of contrast figure; Fig. 2 (c), Fig. 2 (d) are second group of contrast figure; Fig. 2 (e), Fig. 2 (f) are the 3rd group of contrast figure; Fig. 2 (g), Fig. 2 (h) are the 4th group of contrast figure.
Fig. 3 is image co-registration result contrast.
Wherein, Fig. 3 (a), Fig. 3 (b) is an input original image; Fig. 3 (c) is the fused image of 32*32 for the piece area size; Fig. 3 (d) is the fused image of 16*16 for the piece area size; Fig. 3 (e) is a laplacian pyramid algorithm fused image; Fig. 3 (f) is a discrete wavelet transformer scaling method fused image.
Embodiment:
In order to understand technical scheme of the present invention better, embodiments of the present invention are further described below in conjunction with accompanying drawing.
A kind of fusion method synoptic diagram of cutting apart multi-focus image fusing method based on piece that Fig. 1 proposes for the present invention.The concrete implementation detail of each several part is as follows:
1. the local contrast of image block areas
After the burnt input original image of poly is divided into several equal-sized, adopt piece zone local contrast to reflect difference between image focusing clear area and the focusing fuzzy region.
The definition of picture contrast D:
D=(L-L
B)/L
B(1) wherein, L is the local luminance of image, and it is equivalent to the image local gray scale; L
BBe local background's brightness of image, it is equivalent to the image local low frequency component; L-L so
BBe equivalent to the image local high fdrequency component.Here utilize wavelet decomposition to obtain the image block areas local contrast.Suppose that original image is f, when image f is carried out wavelet decomposition, do not carry out down-sampling after the filtering, remain unchanged with picture content and original image size after guaranteeing to decompose, so that graphical analysis.If A
1f, D
1f, D
2fAnd D
3fBe respectively low frequency component, vertical high frequency component, horizontal high fdrequency component and the diagonal high-frequency components of image.The piece zone local contrast Ci of image may be defined as:
Wherein i is an image block; n
iBe the pixel count in the image block; Fig. 2 has provided four groups and has focused on different piece areal maps, and table 1 provides its corresponding contrast results.
Table 1: the contrast results of image block
| First group | Second group | The 3rd group | The 4th group | ||
| The piece contrast | ??0.0604 | ??0.0886 | ??0.0465 | ??0.1464 | Clear |
| ??0.0187 | ??0.0286 | ??0.0239 | ??0.0421 | Blurred block |
2. each dividing region of image
If C
i XLocal contrast for the i piece zone of image X; I
i X(x y) is grey scale pixel value in the i piece zone of image X.So image A is had:
In like manner, have for image B:
For the piece zone that gray-scale value equals 0, it is expressed as the blurred block zone; It is clear zone that gray-scale value equals 1 piece region representation.Clear these piece zones adjacent with blurred block are become 2 with its gray-scale value, be decided to be the boundary block zone.Thus, can access piece image zone marker figure.With regard to generalized case, come the zoning if so, the boundary block zone overwhelming majority is distributed in the intersection of clear area and fuzzy region, but because the influence of factors such as actual imaging, make indivedual boundary block areal distribution in clear area and fuzzy region, area dividing is undesirable.For this reason, need carry out to determine borderline region again after some processing.
When at first the size of image block being chosen, from a large amount of emulation as can be known, image block should not be chosen too for a short time (for entire image), otherwise can increase the existence of boundary block zone in clear area and fuzzy region, the mistake that piece occurs is chosen, and fused image has tangible blocking effect; If it is too big that image block is chosen, borderline region can become greatly so, and the fusion results that this can influence image makes its syncretizing effect that raising by a relatively large margin can not be arranged.The size of image block generally is chosen for: 32*32,32*16,16*32,16*16.
If any one the piece zone in the entire image or several their all adjacent block zones, homogeneous blocks zone that link to each other are another kind of zone, stipulate that so the such piece zone or the type in some continuous homogeneous blocks zone change, and will become the type in adjacent block zone.That is to say to have only above a certain number of continuous homogeneous blocks zone to constitute the class one zone territory, otherwise will be considered to choose wrong piece zone so.The number of choosing adjacent piece zone is for being no more than 3 (or 5), and the continuous piece zone that surpasses 3 (or 5) piece just is considered to a zone.By after the above-mentioned processing, just can obtain merging required image-region signature like this.
3. the fusion treatment in piece zone
After obtaining the image-region signature, can carry out the fusion treatment of image block respectively at dissimilar zones.
For clear area and fuzzy region, because input original image is complementary in these two zones, i.e. the fuzzy region of the clear area correspondence image B of image A, otherwise, the clear area of the fuzzy region correspondence image B of image A.When carrying out fusion treatment, directly choose clear zone as the relevant block zone after merging.
For borderline region, proposed to choose fusion method on the wavelet decomposition basis when the local contrast in computed image piece zone and handled based on the pixel of contrast.
The first step:, ask for the low frequency component A in its residing neighborhood to each pixel in the borderline region
Z Wherein, n
ZBe the number of pixels in the neighborhood Z; A
1f(m n) is (m, n) low frequency component of pixel.
Second step: obtain (m, n) contrast of pixel.
Wherein, D
1f(m, n), D
2f(m, n) and D
3f(m n) is respectively (m, n) the vertical high frequency component of pixel, horizontal high fdrequency component and diagonal high-frequency components.
The 3rd step: carry out choosing of pixel based on contrast.
Wherein, F (m, n) pixel value for choosing after merging; I
A(m, n), I
B(m n) is the pixel value of input original image; X is (m, neighborhood n).
The 4th step: consider the correlativity of neighbor, to the arbitrary pixel in the image, if its adjacent pixels all is selected from another input original image, choosing of this pixel will be identical with choosing of neighbor so.
Figure 3 shows that the fusion results contrast of cutting apart multi-focus image fusing method and wavelet transform fusion and laplacian pyramid fusion method based on piece; Table 2 is corresponding fusion results performance evaluation.
Table 2: image co-registration is index evaluation as a result
| ????P | ??32*32 | ??32*16 | ??16*32 | ??16*16 | ????LP | ??DWT |
| Average error | ??0.6791 | ??0..6776 | ??0.7.051 | ??0.7082 | ??1.8973 | ??2.4380 |
| Total information | ??6.3063 | ??6.3245 | ??6.2718 | ??6.2793 | ??3.6716 | ??3.8991 |
Claims (1)
1, a kind of multi-focus image fusing method of cutting apart based on piece is characterized in that comprising following concrete steps:
1) after input original image is carried out non-down-sampled wavelet decomposition, obtain low frequency component, vertical high frequency component, horizontal high fdrequency component and the diagonal high-frequency components of image, the mean value that the high fdrequency component and the absolute value sum of the ratio of low frequency component of each point in the piece zone are tried to achieve is as this regional local contrast;
2) utilize piece zone local contrast that entire image is divided into clear area and fuzzy region, and then all piece zones that the clear area is adjacent with fuzzy region divide borderline region into, obtain three different area dividing of image with this, size by limited images piece zone and regulation have only and surpass the method that a certain number of continuous homogeneous blocks zone could constitute the class one zone territory and eliminate the image block areas that mistake is chosen, and obtain merging required image-region signature;
3) carry out the fusion treatment in piece zone, for clear area and fuzzy region, directly choose clear zone as the relevant block zone after merging, for borderline region, on the wavelet decomposition basis when the local contrast in computed image piece zone, ask for the wherein interior low frequency component of the residing neighborhood of each pixel, absolute value sum with the high fdrequency component of pixel and the ratio of low frequency component is the contrast of pixel, chooses respective pixel in the input picture borderline region at last and puts the pixel value of the bigger pixel value of each point contrast sum in its neighborhood for this point after merging.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021370559A CN1177298C (en) | 2002-09-19 | 2002-09-19 | Multiple focussing image fusion method based on block dividing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021370559A CN1177298C (en) | 2002-09-19 | 2002-09-19 | Multiple focussing image fusion method based on block dividing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1402191A true CN1402191A (en) | 2003-03-12 |
| CN1177298C CN1177298C (en) | 2004-11-24 |
Family
ID=4748866
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB021370559A Expired - Fee Related CN1177298C (en) | 2002-09-19 | 2002-09-19 | Multiple focussing image fusion method based on block dividing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1177298C (en) |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1303432C (en) * | 2003-06-05 | 2007-03-07 | 上海交通大学 | Remote sensing image picture element and characteristic combination optimizing mixing method |
| CN1313972C (en) * | 2003-07-24 | 2007-05-02 | 上海交通大学 | Image merging method based on filter group |
| CN100378751C (en) * | 2005-01-21 | 2008-04-02 | 致伸科技股份有限公司 | Discriminating system and method for discriminating pictures and text |
| CN100414969C (en) * | 2003-08-29 | 2008-08-27 | 三星电子株式会社 | Apparatus and method for improving the quality of a picture having a high illumination difference |
| CN101137034B (en) * | 2006-08-29 | 2010-09-01 | 索尼株式会社 | Image determination by frequency domain processing |
| CN101834980A (en) * | 2009-03-13 | 2010-09-15 | 索尼公司 | Image processing equipment and method, facility for study and method and program |
| CN101887581A (en) * | 2010-06-17 | 2010-11-17 | 东软集团股份有限公司 | Image fusion method and device |
| CN101510309B (en) * | 2009-03-30 | 2010-12-01 | 西安电子科技大学 | Segmentation method for improving water parting SAR image based on compound wavelet veins region merge |
| CN101515366B (en) * | 2009-03-30 | 2010-12-01 | 西安电子科技大学 | Watershed SAR image segmentation method based on complex wavelet extraction mark |
| CN1947151B (en) * | 2004-02-23 | 2011-01-26 | 美国西门子医疗解决公司 | A system and method for toboggan based object segmentation using divergent gradient field response in images |
| CN101465000B (en) * | 2007-12-18 | 2011-10-05 | 索尼株式会社 | Image processing apparatus and method, and program |
| CN101630405B (en) * | 2009-08-14 | 2011-10-12 | 重庆市勘测院 | Multi-focusing image fusion method utilizing core Fisher classification and redundant wavelet transformation |
| CN102393958A (en) * | 2011-07-16 | 2012-03-28 | 西安电子科技大学 | Multi-focus image fusion method based on compressive sensing |
| CN102542545A (en) * | 2010-12-24 | 2012-07-04 | 方正国际软件(北京)有限公司 | Multi-focal length photo fusion method and system and photographing device |
| CN101408973B (en) * | 2007-10-10 | 2012-09-26 | 奇景光电股份有限公司 | Method of image processing and device thereof |
| US8306360B2 (en) | 2007-09-07 | 2012-11-06 | Lite-On Technology Corporation | Device and method for obtaining clear image |
| CN103247042A (en) * | 2013-05-24 | 2013-08-14 | 厦门大学 | Image fusion method based on similar blocks |
| CN104394308A (en) * | 2014-11-28 | 2015-03-04 | 广东欧珀移动通信有限公司 | Method of taking pictures in different perspectives with double cameras and terminal thereof |
| CN104506767A (en) * | 2014-11-27 | 2015-04-08 | 惠州Tcl移动通信有限公司 | Method for generating different focal lengths of same scene by using continuous movement of motor and terminal |
| CN104700382A (en) * | 2012-12-16 | 2015-06-10 | 吴凡 | Multi-focus image file handling method |
| CN104881855A (en) * | 2015-06-10 | 2015-09-02 | 北京航空航天大学 | Multi-focus image fusion method using morphology and free boundary condition active contour model |
| CN104036481B (en) * | 2014-06-26 | 2017-02-15 | 武汉大学 | Multi-focus image fusion method based on depth information extraction |
| CN106683064A (en) * | 2016-12-13 | 2017-05-17 | 西北工业大学 | Multi-focusing image fusion method based on two-dimensional coupling convolution |
| WO2017080237A1 (en) * | 2015-11-15 | 2017-05-18 | 乐视控股（北京）有限公司 | Camera imaging method and camera device |
| CN103795920B (en) * | 2014-01-21 | 2017-06-20 | 宇龙计算机通信科技(深圳)有限公司 | Photo processing method and device |
| CN106874444A (en) * | 2017-02-09 | 2017-06-20 | 北京小米移动软件有限公司 | Image processing method and device |
| CN106997446A (en) * | 2016-01-26 | 2017-08-01 | 手持产品公司 | Enhanced matrix notation error correction method |
| CN107424146A (en) * | 2017-06-28 | 2017-12-01 | 北京理工大学 | A kind of infrared polarization method for objectively evaluating image quality and system |
| CN107451959A (en) * | 2016-05-31 | 2017-12-08 | 宇龙计算机通信科技(深圳)有限公司 | Image processing method and system |
| CN107993218A (en) * | 2018-01-30 | 2018-05-04 | 重庆邮电大学 | Image interfusion method based on algebraic multigrid and watershed segmentation |
| CN109257540A (en) * | 2018-11-05 | 2019-01-22 | 浙江舜宇光学有限公司 | Take the photograph photography bearing calibration and the camera of lens group more |
| CN109300086A (en) * | 2018-08-16 | 2019-02-01 | 南京理工大学 | Image block method based on clarity |
| CN109831664A (en) * | 2019-01-15 | 2019-05-31 | 天津大学 | Fast Compression three-dimensional video quality evaluation method based on deep learning |
| CN109949258A (en) * | 2019-03-06 | 2019-06-28 | 北京科技大学 | A kind of image recovery method and device based on NSCT transform domain |
| CN110610470A (en) * | 2019-09-18 | 2019-12-24 | 长安大学 | Camera multi-focus clear image extraction method based on multi-azimuth gradient comparison |
| CN110619616A (en) * | 2019-09-19 | 2019-12-27 | 广东工业大学 | Image processing method, device and related equipment |
| CN110738628A (en) * | 2019-10-15 | 2020-01-31 | 湖北工业大学 | self-adaptive focus detection multi-focus image fusion method based on WIML comparison graph |
| CN110796624A (en) * | 2019-10-31 | 2020-02-14 | 北京金山云网络技术有限公司 | Image generation method and device and electronic equipment |
| CN112634160A (en) * | 2020-12-25 | 2021-04-09 | 北京小米松果电子有限公司 | Photographing method and device, terminal and storage medium |
| CN114972141A (en) * | 2022-05-13 | 2022-08-30 | 华侨大学 | Double-mode focusing analysis method of re-fuzzy theory |
| CN115037850A (en) * | 2021-03-05 | 2022-09-09 | 电子科技大学 | Image acquisition method, device and equipment based on liquid crystal lens and storage medium |
-
2002
- 2002-09-19 CN CNB021370559A patent/CN1177298C/en not_active Expired - Fee Related
Cited By (59)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1303432C (en) * | 2003-06-05 | 2007-03-07 | 上海交通大学 | Remote sensing image picture element and characteristic combination optimizing mixing method |
| CN1313972C (en) * | 2003-07-24 | 2007-05-02 | 上海交通大学 | Image merging method based on filter group |
| CN100414969C (en) * | 2003-08-29 | 2008-08-27 | 三星电子株式会社 | Apparatus and method for improving the quality of a picture having a high illumination difference |
| CN1947151B (en) * | 2004-02-23 | 2011-01-26 | 美国西门子医疗解决公司 | A system and method for toboggan based object segmentation using divergent gradient field response in images |
| CN100378751C (en) * | 2005-01-21 | 2008-04-02 | 致伸科技股份有限公司 | Discriminating system and method for discriminating pictures and text |
| CN101137034B (en) * | 2006-08-29 | 2010-09-01 | 索尼株式会社 | Image determination by frequency domain processing |
| US8306360B2 (en) | 2007-09-07 | 2012-11-06 | Lite-On Technology Corporation | Device and method for obtaining clear image |
| CN101408973B (en) * | 2007-10-10 | 2012-09-26 | 奇景光电股份有限公司 | Method of image processing and device thereof |
| CN101465000B (en) * | 2007-12-18 | 2011-10-05 | 索尼株式会社 | Image processing apparatus and method, and program |
| CN101834980A (en) * | 2009-03-13 | 2010-09-15 | 索尼公司 | Image processing equipment and method, facility for study and method and program |
| CN101510309B (en) * | 2009-03-30 | 2010-12-01 | 西安电子科技大学 | Segmentation method for improving water parting SAR image based on compound wavelet veins region merge |
| CN101515366B (en) * | 2009-03-30 | 2010-12-01 | 西安电子科技大学 | Watershed SAR image segmentation method based on complex wavelet extraction mark |
| CN101630405B (en) * | 2009-08-14 | 2011-10-12 | 重庆市勘测院 | Multi-focusing image fusion method utilizing core Fisher classification and redundant wavelet transformation |
| CN101887581A (en) * | 2010-06-17 | 2010-11-17 | 东软集团股份有限公司 | Image fusion method and device |
| CN101887581B (en) * | 2010-06-17 | 2012-03-14 | 东软集团股份有限公司 | Image fusion method and device |
| CN102542545A (en) * | 2010-12-24 | 2012-07-04 | 方正国际软件(北京)有限公司 | Multi-focal length photo fusion method and system and photographing device |
| CN102393958A (en) * | 2011-07-16 | 2012-03-28 | 西安电子科技大学 | Multi-focus image fusion method based on compressive sensing |
| CN104700382B (en) * | 2012-12-16 | 2018-08-28 | 吴凡 | A kind of multiple focussing image document handling method |
| CN104700382A (en) * | 2012-12-16 | 2015-06-10 | 吴凡 | Multi-focus image file handling method |
| CN103247042A (en) * | 2013-05-24 | 2013-08-14 | 厦门大学 | Image fusion method based on similar blocks |
| CN103247042B (en) * | 2013-05-24 | 2015-11-11 | 厦门大学 | A kind of image interfusion method based on similar piece |
| CN103795920B (en) * | 2014-01-21 | 2017-06-20 | 宇龙计算机通信科技(深圳)有限公司 | Photo processing method and device |
| CN104036481B (en) * | 2014-06-26 | 2017-02-15 | 武汉大学 | Multi-focus image fusion method based on depth information extraction |
| CN104506767A (en) * | 2014-11-27 | 2015-04-08 | 惠州Tcl移动通信有限公司 | Method for generating different focal lengths of same scene by using continuous movement of motor and terminal |
| CN104506767B (en) * | 2014-11-27 | 2019-08-02 | 惠州Tcl移动通信有限公司 | The method and terminal of same scenery different focal length are generated using motor continuous moving |
| CN104394308B (en) * | 2014-11-28 | 2017-11-07 | 广东欧珀移动通信有限公司 | Method and terminal that dual camera is taken pictures with different visual angles |
| CN104394308A (en) * | 2014-11-28 | 2015-03-04 | 广东欧珀移动通信有限公司 | Method of taking pictures in different perspectives with double cameras and terminal thereof |
| CN104881855A (en) * | 2015-06-10 | 2015-09-02 | 北京航空航天大学 | Multi-focus image fusion method using morphology and free boundary condition active contour model |
| CN104881855B (en) * | 2015-06-10 | 2017-07-28 | 北京航空航天大学 | A kind of multi-focus image fusing method of utilization morphology and free boundary condition movable contour model |
| WO2017080237A1 (en) * | 2015-11-15 | 2017-05-18 | 乐视控股（北京）有限公司 | Camera imaging method and camera device |
| CN106997446A (en) * | 2016-01-26 | 2017-08-01 | 手持产品公司 | Enhanced matrix notation error correction method |
| CN106997446B (en) * | 2016-01-26 | 2022-03-18 | 手持产品公司 | Enhanced matrix symbol error correction method |
| US11449700B2 (en) | 2016-01-26 | 2022-09-20 | Hand Held Products, Inc. | Enhanced matrix symbol error correction method |
| US11727232B2 (en) | 2016-01-26 | 2023-08-15 | Hand Held Products, Inc. | Enhanced matrix symbol error correction method |
| CN107451959A (en) * | 2016-05-31 | 2017-12-08 | 宇龙计算机通信科技(深圳)有限公司 | Image processing method and system |
| CN106683064B (en) * | 2016-12-13 | 2019-07-30 | 西北工业大学 | A kind of multi-focus image fusing method based on two dimension coupling convolution |
| CN106683064A (en) * | 2016-12-13 | 2017-05-17 | 西北工业大学 | Multi-focusing image fusion method based on two-dimensional coupling convolution |
| CN106874444A (en) * | 2017-02-09 | 2017-06-20 | 北京小米移动软件有限公司 | Image processing method and device |
| CN107424146A (en) * | 2017-06-28 | 2017-12-01 | 北京理工大学 | A kind of infrared polarization method for objectively evaluating image quality and system |
| CN107993218B (en) * | 2018-01-30 | 2021-09-07 | 重庆邮电大学 | Image fusion method based on algebraic multiple meshes and watershed segmentation |
| CN107993218A (en) * | 2018-01-30 | 2018-05-04 | 重庆邮电大学 | Image interfusion method based on algebraic multigrid and watershed segmentation |
| CN109300086A (en) * | 2018-08-16 | 2019-02-01 | 南京理工大学 | Image block method based on clarity |
| CN109300086B (en) * | 2018-08-16 | 2022-09-27 | 南京理工大学 | Image blocking method based on definition |
| CN109257540A (en) * | 2018-11-05 | 2019-01-22 | 浙江舜宇光学有限公司 | Take the photograph photography bearing calibration and the camera of lens group more |
| CN109831664A (en) * | 2019-01-15 | 2019-05-31 | 天津大学 | Fast Compression three-dimensional video quality evaluation method based on deep learning |
| CN109949258A (en) * | 2019-03-06 | 2019-06-28 | 北京科技大学 | A kind of image recovery method and device based on NSCT transform domain |
| CN109949258B (en) * | 2019-03-06 | 2020-11-27 | 北京科技大学 | Image restoration method based on NSCT transform domain |
| CN110610470A (en) * | 2019-09-18 | 2019-12-24 | 长安大学 | Camera multi-focus clear image extraction method based on multi-azimuth gradient comparison |
| CN110610470B (en) * | 2019-09-18 | 2022-12-09 | 西安汇智信息科技有限公司 | Camera multi-focus clear image extraction method based on multi-azimuth gradient comparison |
| CN110619616A (en) * | 2019-09-19 | 2019-12-27 | 广东工业大学 | Image processing method, device and related equipment |
| CN110738628A (en) * | 2019-10-15 | 2020-01-31 | 湖北工业大学 | self-adaptive focus detection multi-focus image fusion method based on WIML comparison graph |
| CN110738628B (en) * | 2019-10-15 | 2023-09-05 | 湖北工业大学 | Adaptive focus detection multi-focus image fusion method based on WIML comparison graph |
| CN110796624B (en) * | 2019-10-31 | 2022-07-05 | 北京金山云网络技术有限公司 | Image generation method and device and electronic equipment |
| CN110796624A (en) * | 2019-10-31 | 2020-02-14 | 北京金山云网络技术有限公司 | Image generation method and device and electronic equipment |
| US11836898B2 (en) | 2019-10-31 | 2023-12-05 | Beijing Kingsoft Cloud Network Technology Co., Ltd. | Method and apparatus for generating image, and electronic device |
| CN112634160A (en) * | 2020-12-25 | 2021-04-09 | 北京小米松果电子有限公司 | Photographing method and device, terminal and storage medium |
| CN115037850A (en) * | 2021-03-05 | 2022-09-09 | 电子科技大学 | Image acquisition method, device and equipment based on liquid crystal lens and storage medium |
| CN115037850B (en) * | 2021-03-05 | 2023-10-20 | 电子科技大学 | Image acquisition method, device, equipment and storage medium based on liquid crystal lens |
| CN114972141A (en) * | 2022-05-13 | 2022-08-30 | 华侨大学 | Double-mode focusing analysis method of re-fuzzy theory |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1177298C (en) | 2004-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1177298C (en) | Multiple focussing image fusion method based on block dividing | |
| CN1135500C (en) | Method and apparatus for analyzing image structures | |
| CN112884064B (en) | Target detection and identification method based on neural network | |
| CN1273937C (en) | Infrared and visible light image merging method | |
| CN109978807B (en) | Shadow removing method based on generating type countermeasure network | |
| CN1104816C (en) | Method and apparatus for determining position of TV camera for use in virtual studio | |
| CN106971185B (en) | License plate positioning method and device based on full convolution network | |
| CN111046880A (en) | Infrared target image segmentation method and system, electronic device and storage medium | |
| CN1238505A (en) | Method and apparatus for enhancing discrete pixel images | |
| Seo et al. | Progressive attention networks for visual attribute prediction | |
| WO2019218393A1 (en) | Image pre-processing for accelerating cytological image classification by fully convolutional neural networks | |
| CN106446750A (en) | Bar code reading method and device | |
| CN111382658B (en) | Road traffic sign detection method in natural environment based on image gray gradient consistency | |
| Ibarra-Arenado et al. | Shadow-based vehicle detection in urban traffic | |
| CN103778616A (en) | Contrast pyramid image fusion method based on area | |
| CN1835547A (en) | Image processing device and registration data generation method in image processing | |
| CN106204494A (en) | A kind of image defogging method comprising large area sky areas and system | |
| CN111008979A (en) | Robust night image semantic segmentation method | |
| CN114627090A (en) | Convolutional neural network optical lens defect detection method based on attention mechanism | |
| Mathavan et al. | Fast segmentation of industrial quality pavement images using laws texture energy measures and k-means clustering | |
| CN1122237C (en) | Property determine method | |
| CN115620259A (en) | Lane line detection method based on traffic off-site law enforcement scene | |
| CN116342877A (en) | Semantic segmentation method based on improved ASPP and fusion module in complex scene | |
| CN110807494B (en) | Quick positioning method for repeated textures in industrial vision | |
| CN1272746C (en) | Multiple focus image fusing method based inseparable small wave frame change |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C19 | Lapse of patent right due to non-payment of the annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |