CN101930461A - Digital image visualized management and retrieval for communication network - Google Patents
Digital image visualized management and retrieval for communication network Download PDFInfo
- Publication number
- CN101930461A CN101930461A CN 201010258040 CN201010258040A CN101930461A CN 101930461 A CN101930461 A CN 101930461A CN 201010258040 CN201010258040 CN 201010258040 CN 201010258040 A CN201010258040 A CN 201010258040A CN 101930461 A CN101930461 A CN 101930461A
- Authority
- CN
- China
- Prior art keywords
- image
- digital picture
- primitive
- digital
- present
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a visualized management and retrieval method for a digital image. The method comprises the following steps: establishing a binary bitmap sequence for the digital image by a digital image processing technological system and extracting a plurality of feature primitives from the sequence (the primitives refer to factors in the image which are different from pixels, such as a line segment, a curve, a closed figure, texture and the like); correspondingly extracting RGB, a gray value and other data of pixels in the peripheral range of relevant positions from the original digital image according to the position range of the primitives, extracting a set of texture elements according to the texture range related to the primitives, and indexing the digital image according to a plurality of the primitives, the set of the texture elements and corresponding pixel values; and storing the digital image and the image indexing, relevant text comment, network link analysis and the like for retrieving the digital image. In the invention, indexing of the digital image is established based on the binary bitmap sequence, the primitives, the texture elements, the pixels and the like, thus achieving small data quantity and fast and convenient retrieval and being applicable to digital image management, retrieval and search for a mobile communication network and the Internet.
Description
Affiliated technical field
The present invention discloses the method for a kind of visualized management and key numbers image.The index of digital picture especially of the present invention is based on bitonal bitmap, primitive (primitive, refer to be different from the image key element of pixel, as line segment, curve, closed figure etc.), texel and pixel basis set up, the index number image is accurate, its data volume is very little, network retrieval is quick, is applicable to the search of mobile communications network and internet digital picture, management and retrieval.
The present invention is digital image visualized to be based on graphics (graphics) and the application of iconography (imaging) technical method aspect computational science (computational science), lay particular emphasis on and utilize computer graphics to create visual pattern, thereby help people to understand the digitized video that those take digital appearance form intricate and often in large scale, be applied to digital image acquisition, analysis, index, calculating, classification, storage, retrieval and search in the communication network.
Background technology
At present, known network digital image management and search method be substantially, generously always the network digital image managed and retrieves according to two.
Direction one, the characteristics of image that in the network digital image, extracts, the important component part that the corresponding note literal of image is a characteristics of image, then identify other image with similar image feature, to form one or more feature combinations of image, then obtain the text that is associated with described all images, and at the characteristics of image that is obtained and the text sensible friendship key word of getting the bid, utilize at last and intersect key word and come annotating images from Webpage.As, by character search " jaguar ", obtain the image of jaguar.
Direction two is extracted characteristics of image from digital picture, generate color histogram, direction histogram etc. by color, tone, form and aspect, saturation degree, brightness or the like in the digital picture; Use discrete cosine transform (DCT) technology, use principal component analysis (PCA) technology or the like numerous computer vision techniques to index as the view picture digital picture, when retrieving according to the overall color return results of image.As, by retrieval jaguar coat pattern picture, obtain the image of jaguar; Even jaguar has 9 subspecies and every kind of dot pattern that a distinctness is all arranged on one's body, and picture searching can be discerned wherein difference by the retrieval of entire image overall color, and only returns the result of specific decorative pattern subspecies jaguar.
To above known two kinds of digital image management with search method is simple and brief is interpreted as, crucial literal and picture color indexed search, also have the network image research tool with above two kinds of methods or more multiclass lump together to mix like method and use.But, more than the management and retrieval method of known network digital image, in real network is used, have very big defective.
Crucial literal indexed search image method, the prerequisite of its existence are to have a plurality of important hypothesis in the computational science algorithm principle: promptly all images all have a firm linked, diagram, and each image can reach from the hyperlink of other Web page images, literal; Each image all has its literal note or corresponding article introduction based on webpage; Each image by web page or other form with a plurality of image-related.But in real communication network, and not exclusively be such situation.When an image did not link, it just can not be with other image associations in the network.As a same reason, when an image did not have words identification, it does not have yet can be with the crossing literal of other texts.Just there is defective in crucial literal thumbnail method.
Picture color indexed search image method, in the face of surpassing digital picture 10,000,000,000, enormous amount in the communication network, and along with the explosive growth of the network information, data are also in continuous growth.Will carry out one by one that the entire image index upgrade need expend plenty of time and data space to like this huge data, even if can satisfy the demand in time and space, tackle millions of retrieval and query demand, this also will be a bad dream! Moreover this is not only the problem that software engineering can solve, and also will rely on the development of memory device and technology thereof.
More than known digital image management and search method, also to face many difficult problems such as the obstruction of communication network data, data transmission loss, terminal arithmetic speed, terminal display capabilities.Everything causes difficulty for the digital image management and the retrieval of communication network, occurs wrong result easily.
Summary of the invention
In order to overcome the deficiency of existing known network digital image management and search method, the present invention discloses a kind of digital image visualized management and search method of communication network, this method can be simply, index number image apace, and under the prerequisite of a plurality of feature primitives of index number image, the texel set of thumbnail texture, the color character of index number image related pixel, and the data volume of index number image reduced to very little data volume scope.The present invention generates a plurality of even numerous primitives to the equal index of each digital picture, and also there is similarity in the property of there are differences between these primitives, even homogeneity; This has just formed the primitive database of numerous independent primitive set, and the numerous independent primitive in the primitive database passes through Different Rule, various combination, various computing, thereby can index and each digital picture of annotating hundreds of millions quantity in the communication network.This just looks like to be the formation of explaining cell with DNA, and distinguishing different organic-biological bodies with the combination of different DNA is reasons.To the index of digital picture according to primitive, auxiliary primitive respective pixel numerical value index, and related text note, network linking analysis etc., the present invention can reach fast, accurately, efficiently the management and retrieval of digital picture in the communication network.
The technical solution adopted for the present invention to solve the technical problems is:
The digital image visualized management of communication network of the present invention and retrieval solve the technical scheme that its technical matters adopted: digital picture in the image capturing system collection network and text thereof, peer link, digital picture is because its distinctive file independence, image capturing system may only collect digital picture itself, and literal that is not associated and link.
Image processing system of the present invention carries out computing with the digital picture that image capturing system collects, and sets up the bitonal bitmap sequence according to digital picture.Bitmap (Bitmap) claims raster pattern (Raster Graphics) again, the image that is to use pel array to represent, and each color of pixel information is represented by RGB combination or gray-scale value.Be divided into 1,4,8,16,24 and 32 etc. according to the required data bit of colouring information, the high more color of figure place is abundant more, and corresponding data volume is big more.Wherein use the bitmap of a pixel color of 1 bit representation because a data bit can only be represented two kinds of colors, so be called bitonal bitmap again.Bitonal bitmap of the present invention presents with two kinds of colors of black and white usually.
The bitonal bitmap sequence that the present invention sets up refers to that these bitonal bitmaps are arranged according to the particular sequence form according to a plurality of bitonal bitmaps of different threshold values color ranges (Threshold Level) numerical value foundation.Threshold values refers to the division boundary of black/white in the tone range.
Sequence refers to: on the mathematics, and the object that sequence is formed a line (or incident); Like this, each element is not before other elements, is exactly after other elements.Here, the order between the element is extremely important.Sequence form, the subsequence of a given sequence are to remove some elements from given sequence, and relative position obtains between other elements and do not change.If the item of sequence belongs to a paritially ordered set, then monotone-increasing sequence be exactly wherein each all more than or equal to before item; If each all strict item greater than before, this sequence are exactly strictly monotone increasing.
When image processing system of the present invention was set up the bitonal bitmap of digital picture, image processing system used a large amount of figures (graphics) and image (imaging) versatility synthetically to calculate and processing digital images, and obtained the splendid bitonal bitmap of quality.Figure (graphics) and image (imaging) technology that image processing system of the present invention uses relate to color range (Levels), curve (Curves), color balance (Color Balance), brightness/contrast (Brightness/Cotrast), form and aspect/saturation degree (Hue/Saturation), (Desaturate) discolors, match (Match Color), replace color (Replace Color), selective color (Selective Color), passage mixes (Channel Mixer), filter (Fliter), shade/Gao Guang (Shadow/Highlight), tonal width (Tonal), color correction (Color Correction), black and white is pruned (Black/White Clip), exposure (Exposure), anti-phase (Invert), tone homogenizing (Equalize), gradual change mapping (Gradient Map), tone separates (Posterize), threshold values (Threshold), the real mixture of colours (Hard Mix), technology and method that difference (Difference) or the like is numerous.
After obtaining the bitonal bitmap sequence of digital picture, characteristics of image primitive extraction system of the present invention is extracted primitive according to the bitonal bitmap of digital picture.Primitive: contain some key elements that are different from the pixel aspect among the senior expression-form of image.These key elements are called as primitive (primitives is called pel or primitive again).For example, just to belong to be wherein primitive for the line segment of digital picture and curve.The present invention will can identify, represent, annotate the point, line segment, curve, closed figure, irregular segment etc. of characteristics of image and image important component part as single primitive in the digital picture.
Characteristics of image primitive extraction system of the present invention can be extracted one or more primitives to the single digital image bitonal bitmap, each digital picture bitonal bitmap can extract the concrete quantity of primitive, handle according to the concrete condition of this digital picture bitonal bitmap sequence by characteristics of image primitive extraction system, do not wait from one to tens of, hundreds of, thousands of.The final primitive that obtains is stored into the primitive database.
Characteristics of image texel extraction system of the present invention is according to different threshold values color range numerical value change situations in the digital picture bitonal bitmap sequence, and with the relative relative position of primitive, calculate and extract the texture that single bitonal bitmap presents in the bitonal bitmap sequence, the texture of extraction is actual to be the set of some texels.Texel is the abbreviation of texel, and it is the elementary cell in the computer graphical texture space, is to be arranged by pixel to form as image, and texture is arranged expression by texel.After the texel set of acquisition with the primitive association, be stored into the texel data storehouse.
After obtaining the primitive of digital picture, characteristics of image pixel extraction of the present invention system is according to the pixel of relevant position peripheral extent in the corresponding original digital image of position range of primitive, data such as the rgb value of extraction primitive peripheral extent pixel and gray scale (Grayscale) value, and be stored into the associated pixel database that is associated and mates with the primitive database.
Simultaneously, other attributes extraction systems of characteristics of image of the present invention extract character features such as associated picture key word, keyword, key sentence, sign literal, note literal, supporting text, the meaning of one's words, stem; With own network linking, forward chaining, backward chaining, the associated link of digital picture, etc. hyperlink (URL) feature extraction; And character features and chain feature be stored into other attribute databases.
The present invention carries out index according to digital picture features such as the primitive that extracts, texel, pixel, literal, links to digital picture.Wherein a plurality of primitives of each digital picture, texel set and related pixel index are the key elements of main search, and literal and link index are searched for key element as assisted retrieval.
Each digital picture in the communication network is obtained by image capturing system, and image processing system is set up the bitonal bitmap sequence; Characteristics of image primitive extraction system is extracted one or more primitives according to the bitonal bitmap sequence of image; Characteristics of image texel extraction system is extracted the texel set with the primitive association, data such as the rgb value of the corresponding primitive peripheral extent of characteristics of image pixel extraction system extraction original digital image pixel and gray scale (Grayscale) value; Other attributes extraction systems of characteristics of image extract character features and chain feature; More than the data extracted of each system be stored into corresponding database respectively.
When picture search and Search Requirement were initiated, the present invention carried out digital picture feature extractions such as primitive, texel, pixel, literal, link to the digital picture of initiating search and retrieval equally, with the characteristics of image that obtains this image of index as a result.Index matching system of the present invention and search processing system will be retrieved, mate and calculate in primitive database that the present invention has set up, texel collective database, associated pixel database, other attribute databases with regard to the index of this image, obtain indexed results and corresponding digital picture identical, similar, that be complementary.
To indexed results and the corresponding digital picture that obtains, index levels output system of the present invention, the actual quantity that will be complementary according to primitive in the indexed results and associated pixel, other attributive character, the digital picture of calculating and dividing the search result shows ordering, and output search result's digital picture.
The digital picture of initiating search is by index of the present invention, the primitive that is extracted, texel set and associated pixel, in the retrieval of database, coupling, comparison, computation process, will influence and determine final search result's ordering to the full extent with primitive, the texel set quantity finally similar with associated pixel, that be complementary of other digital pictures.
The invention has the beneficial effects as follows that the present invention is to be based upon on the digital picture bitonal bitmap basis to digital picture index, the ubiquity mode of bitonal bitmap of the present invention is a black-and-white two color.Owing to convert coloured image to bitonal bitmap, dwindled the size of digital picture space storage data capacity greatly, the data volume ratio that it reduces is tens times, hundred times, even thousands of times.Example, a spatial data amount is 2, the digital image of 968Kb, after setting up the bitonal bitmap sequence by image processing system of the present invention, the spatial data amount of the primitive that collects, texel, pixel characteristic file is between 100KB～20KB, and its spatial data amount only is 3.37%～0.67% of a raw data image space data volume ratio.
The present invention is to the leaching process of digital picture primitive, again digital picture is divided into a plurality of, hundreds of primitive units, and has extracted set of representational, as to annotate picture feature a plurality of primitives, texel and data such as entrained rgb value of the pixel in the original digital image and gray scale (Grayscale) value on every side thereof.On the basis of primitive, keep the intrinsic chromatic characteristic of data image like this, obtained a plurality of key characteristics (Key of Image) of data image again, and dwindled data capacity again greatly, improved the usefulness of data retrieval search.
By like this to the index process of single digital image, those take the digitized video of digital appearance form intricate and often in large scale, process the present invention simplifies and has created primitive, texel, key element and the computing method of many visual patterns, and the index that surpasses digital picture 10,000,000,000, enormous amount in the communication network is just become very simple and clear.Therefore, the present invention to the management of digital picture, retrieval, search can be the most quick, the most efficient, present the most intuitively.
Description of drawings
The present invention is further described below in conjunction with drawings and Examples.
Fig. 1 is an architectural schematic of the present invention
1. image capturing systems among the figure, 2. user search retrieving images, 3. image processing system, 4. characteristics of image primitive extraction system, 5. primitive database, 6. characteristics of image texel extraction system, 7. texel data storehouse, 8. characteristics of image pixel extraction system, 9. pixel data storehouse, 10. index matching system, 11. other attribute disposal systems, 12. characteristics of image literal extraction system, 13. characteristics of image link extraction system, 14. other attribute databases, 15. search processing systems, 16. the index levels output system, 17. search number of results word images ordering output shows.
Embodiment
Fig. 1 is the digital image visualized management of communication network of the present invention and the architectural schematic of retrieval.
In Fig. 1, each digital picture in the communication network is obtained by image capturing system (1) or user search retrieving images (2), and image processing system (3) is set up the bitonal bitmap sequence; Characteristics of image primitive extraction system (4) is extracted one or more primitives according to the bitonal bitmap sequence of image, and is stored into primitive database (5); Characteristics of image texel extraction system (6) is extracted the texel set with the primitive association, and is stored into texel data storehouse (7); Data such as the rgb value of the corresponding primitive peripheral extent of characteristics of image pixel extraction system (8) extraction original digital image pixel and gray scale (Grayscale) value, and be stored into pixel data storehouse (9); Other attributes extraction systems (11) of characteristics of image extract literal and chain feature by character features extraction system (12) and chain feature extraction system (13) respectively, and are stored into other attribute databases (14).
When picture search and Search Requirement were initiated, the present invention carried out digital picture feature extractions such as primitive, texel, pixel, literal, link to the digital picture of initiating search and retrieval equally, with the characteristics of image that obtains this image of index as a result.Index matching system of the present invention (10) and search processing system (15) will be retrieved, mate and calculate in (14) at primitive database (5), texel collective database (7), associated pixel database (9), other attribute databases that the present invention has set up with regard to the index of this image, obtain indexed results and corresponding digital picture identical, similar, that be complementary.
To indexed results and the corresponding digital picture that obtains, index levels output system of the present invention (16), the actual quantity that will be complementary according to primitive in the indexed results and associated pixel, other attributive character, the digital picture of calculating and dividing the search result shows ordering, and output search result's digital picture (17).
Claims (6)
1. the present invention discloses the method for digital image visualized management and retrieval in a kind of communication network.Set up the bitonal bitmap sequence (1) of digital picture by the digital image processing techniques system, and therefrom extract a plurality of feature primitive (primitives, refer to be different from the image key element of pixel, as line segment, curve, closed figure, texture etc.), according to data such as corresponding RGB that extracts relevant position peripheral extent pixel in the original digital image of the position range of primitive and gray-scale values, the set that the texture range that while relates to according to primitive is extracted texel, according to the set of a plurality of primitives, texel set and respective pixel numerical value, this digital picture of index (2); Store this digital picture and image index, related text note, network linking analysis etc., be used for retrieval (3) digital picture.Because the index of digital picture of the present invention is set up based on bitonal bitmap sequence, primitive, texel, pixel basis, its data volume is very little, retrieval is quick, is applicable to mobile communications network and internet digital image management, retrieval and search.
2. the digital image visualized management of communication network according to claim 1 and retrieval, it is characterized in that: the digital picture that image capturing system collects is carried out computing at the image processing system of the present invention described in the step (1), set up the bitonal bitmap sequence according to digital picture, refer to:
● the bitonal bitmap sequence of foundation refers to that these bitonal bitmaps are arranged according to the particular sequence form according to a plurality of bitonal bitmaps of different threshold values color ranges (Threshold Level) numerical value foundation;
● bitmap (Bitmap) is divided into 1,4,8,16,24 and 32 etc. according to the required data bit of colouring information, wherein use the bitmap of a pixel color of 1 bit representation because a data bit can only be represented two kinds of colors, be called bitonal bitmap, bitonal bitmap of the present invention presents with two kinds of colors of black and white usually;
● when image processing system of the present invention is set up the bitonal bitmap of digital picture, image processing system uses a large amount of figures (graphics) and image (imaging) versatility synthetically to calculate and processing digital images, relates to color range (Levels), curve (Curves), color balance (Color Balance), brightness/contrast (Brightness/Cotrast), form and aspect/saturation degree (Hue/Saturation), (Desaturate) discolors, match (Match Color), replace color (Replace Color), selective color (Selective Color), passage mixes (Channel Mixer), filter (Fliter), shade/Gao Guang (Shadow/Highlight), tonal width (Tonal), color correction (Color Correction), black and white is pruned (Black/White Clip), exposure (Exposure), anti-phase (Invert), tone homogenizing (Equalize), gradual change mapping (Gradient Map), tone separates (Posterize), threshold values (Threshold), the real mixture of colours (Hard Mix), technology and method that difference (Difference) or the like is numerous.
3. the digital image visualized management of communication network according to claim 1 and retrieval, it is characterized in that: after the bitonal bitmap sequence of the acquisition digital picture described in the step (2), image characteristics extraction of the present invention system extracts the set of a plurality of primitives, texel set and the respective pixel numerical value of digital picture respectively, comprising:
● characteristics of image primitive extraction system of the present invention is extracted primitive (primitive: contain some key elements that are different from the pixel aspect among the senior expression-form of image according to the bitonal bitmap of digital picture, these key elements are called as primitive primitives, are called pel or primitive again; For example, the line segment of digital picture and curve just to belong to be wherein primitive);
● the present invention will can identify, represent, annotate the point, line segment, curve, closed figure, irregular segment etc. of characteristics of image and image important component part as single primitive in the digital picture, and calculate and extract, the final primitive of acquisition is stored into the primitive database;
● characteristics of image primitive extraction system of the present invention can be extracted one or more primitives to the single digital image bitonal bitmap, each digital picture bitonal bitmap can extract the concrete quantity of primitive, handle according to the concrete condition of this digital picture bitonal bitmap sequence by characteristics of image primitive extraction system, do not wait from one to tens of, hundreds of, thousands of;
● characteristics of image texel extraction system of the present invention is according to different threshold values color range numerical value change situations in the digital picture bitonal bitmap sequence, and with the relative relative position of primitive, calculate and extract the texture that single bitonal bitmap presents in the bitonal bitmap sequence, the texture of extraction is actual to be the set of some texels;
● texel is the abbreviation of texel, and it is the elementary cell in the computer graphical texture space, is to be arranged by pixel to form as image, and texture is arranged expression by texel;
● after calculating the texel set of extraction acquisition, be stored into the texel data storehouse with the primitive association;
● after obtaining the primitive of digital picture, characteristics of image pixel extraction of the present invention system is according to the pixel of relevant position peripheral extent in the corresponding original digital image of position range of primitive, data such as the rgb value of extraction primitive peripheral extent pixel and gray scale (Grayscale) value, and be stored into the associated pixel database that is associated and mates with the primitive database;
● the present invention carries out index according to the digital picture features such as set of a plurality of primitives that extract, texel set and respective pixel numerical value to digital picture.
4. the digital image visualized management of communication network according to claim 1 and retrieval, it is characterized in that: in this digital picture of storage described in the step (3) and image index, related text note, network linking analysis etc., be used for retrieval, comprise digital picture:
● other attributes extraction systems of characteristics of image of the present invention extract character features such as associated picture key word, keyword, key sentence, sign literal, note literal, supporting text, the meaning of one's words, stem;
● with own network linking, forward chaining, backward chaining, the associated link of digital picture, etc. hyperlink (URL) feature extraction;
● and character features and chain feature be stored into other attribute databases;
● the present invention carries out index according to digital picture features such as the literal that extracts, links to digital picture;
● wherein a plurality of primitives of each digital picture, texel set and related pixel index are the key elements of main search, and literal and link index are searched for key element as assisted retrieval.
5. the method for the digital image visualized management of communication network according to claim 1 and retrieval comprises:
● digital picture that image capturing system of the present invention is gathered from numerous communication networks such as mobile communications network, internet and text thereof, peer link;
● system of the present invention can discern, gathers, retrieves, search for, multi-lingual and literal such as Chinese, English;
● the reception user initiates to be transferred to the requirement retrieval of system of the present invention, the digital picture of search;
● obtain texts such as the key word relevant, keyword, key sentence, sign literal, note literal, supporting literal, the meaning of one's words, stem with the digital picture of gathering and receiving;
● obtain own network linking, forward chaining, backward chaining, the associated link relevant, etc. hyperlink (URL) with the digital picture of gathering and receiving;
● utilize the text of acquisition and the digital picture that chain feature comes the note index to gather and receive.
6. the method for the digital image visualized management of communication network according to claim 1 and retrieval comprises:
● when picture search and Search Requirement were initiated, the present invention carried out digital picture feature extractions such as primitive, texel, pixel, literal, link to the digital picture of initiating search and retrieval equally, with the characteristics of image that obtains this image of index as a result;
● index matching system of the present invention and search processing system will be retrieved, mate and calculate in primitive database that the present invention has set up, texel collective database, associated pixel database, other attribute databases with regard to the index of this image, obtain indexed results and corresponding digital picture identical, similar, that be complementary;
● to indexed results and the corresponding digital picture that obtains, index levels output system of the present invention, the actual quantity that will be complementary according to primitive in the indexed results and associated pixel, other attributive character, the digital picture of calculating and dividing the search result shows ordering, and output search result's digital picture;
● the digital picture of initiating search is by index of the present invention, the primitive that is extracted, texel set and associated pixel, in the retrieval of database, coupling, comparison, computation process, will influence and determine final search result's ordering to the full extent with primitive, the texel set quantity finally similar with associated pixel, that be complementary of other digital pictures.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010258040 CN101930461A (en) | 2010-08-20 | 2010-08-20 | Digital image visualized management and retrieval for communication network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010258040 CN101930461A (en) | 2010-08-20 | 2010-08-20 | Digital image visualized management and retrieval for communication network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101930461A true CN101930461A (en) | 2010-12-29 |
Family
ID=43369638
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010258040 Pending CN101930461A (en) | 2010-08-20 | 2010-08-20 | Digital image visualized management and retrieval for communication network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101930461A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101945327A (en) * | 2010-09-02 | 2011-01-12 | 郑茂 | Wireless positioning method and system based on digital image identification and retrieve |
| CN102494606A (en) * | 2011-12-02 | 2012-06-13 | 中国科学院上海微系统与信息技术研究所 | Optical detection method of ten-nanometer-magnitude sizes and errors |
| CN103198162A (en) * | 2013-04-28 | 2013-07-10 | 冠捷显示科技（厦门）有限公司 | Image browsing and interacting method |
| CN103459980A (en) * | 2011-03-22 | 2013-12-18 | 哈曼贝克自动系统股份有限公司 | Management of icons for digital maps |
| CN103782324A (en) * | 2011-09-12 | 2014-05-07 | 索尼电脑娱乐公司 | Accelerated texture lookups by using texture coordinate derivatives |
| CN106233278A (en) * | 2014-07-24 | 2016-12-14 | 惠普发展公司有限责任合伙企业 | Segment for image creates view data |
| CN106909548A (en) * | 2015-12-22 | 2017-06-30 | 北京奇虎科技有限公司 | Picture loading method and device based on server |
| CN109582812A (en) * | 2018-11-21 | 2019-04-05 | 安徽云融信息技术有限公司 | A kind of search method for computer picture |
| CN112182276A (en) * | 2020-10-12 | 2021-01-05 | 云南财经大学 | Astronomical mass observation data hybrid retrieval method based on image content and metadata |
| CN112527210A (en) * | 2020-12-22 | 2021-03-19 | 南京中兴力维软件有限公司 | Storage method and device of full data and computer readable storage medium |
| CN113010722A (en) * | 2021-04-18 | 2021-06-22 | 南通大学 | Chronic disease clinical research queue query system and method based on fusion of fundus images |
| CN114896228A (en) * | 2022-04-27 | 2022-08-12 | 西北工业大学 | Industrial data stream cleaning model and method based on multi-stage combination optimization of filtering rules |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101162506A (en) * | 2007-11-02 | 2008-04-16 | 北京大学 | Seal imprint image search method of circular stamp |
| CN101388020A (en) * | 2008-07-07 | 2009-03-18 | 华南师范大学 | Composite image search method based on content |
| US20090263025A1 (en) * | 2008-04-21 | 2009-10-22 | Jilin Li | Image determination apparatus, image search apparatus and computer readable recording medium storing an image search program |
-
2010
- 2010-08-20 CN CN 201010258040 patent/CN101930461A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101162506A (en) * | 2007-11-02 | 2008-04-16 | 北京大学 | Seal imprint image search method of circular stamp |
| US20090263025A1 (en) * | 2008-04-21 | 2009-10-22 | Jilin Li | Image determination apparatus, image search apparatus and computer readable recording medium storing an image search program |
| CN101388020A (en) * | 2008-07-07 | 2009-03-18 | 华南师范大学 | Composite image search method based on content |
Non-Patent Citations (1)
| Title |
|---|
| 《太原理工大学学报》 20040531 尹荣、郑兆瑞、郭海儒 一种基于图元的多级图像检索系统 345-348 1-6 第35卷, 第3期 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101945327A (en) * | 2010-09-02 | 2011-01-12 | 郑茂 | Wireless positioning method and system based on digital image identification and retrieve |
| CN103459980B (en) * | 2011-03-22 | 2017-05-31 | 哈曼贝克自动系统股份有限公司 | The management of the icon of numerical map |
| CN103459980A (en) * | 2011-03-22 | 2013-12-18 | 哈曼贝克自动系统股份有限公司 | Management of icons for digital maps |
| CN103782324A (en) * | 2011-09-12 | 2014-05-07 | 索尼电脑娱乐公司 | Accelerated texture lookups by using texture coordinate derivatives |
| CN102494606B (en) * | 2011-12-02 | 2014-04-16 | 中国科学院上海微系统与信息技术研究所 | Optical detection method of ten-nanometer-magnitude sizes and errors |
| CN102494606A (en) * | 2011-12-02 | 2012-06-13 | 中国科学院上海微系统与信息技术研究所 | Optical detection method of ten-nanometer-magnitude sizes and errors |
| CN103198162B (en) * | 2013-04-28 | 2016-08-31 | 冠捷显示科技（厦门）有限公司 | A kind of picture browsing exchange method |
| CN103198162A (en) * | 2013-04-28 | 2013-07-10 | 冠捷显示科技（厦门）有限公司 | Image browsing and interacting method |
| CN106233278B (en) * | 2014-07-24 | 2019-05-14 | 惠普发展公司有限责任合伙企业 | Image generating device and the method for generating the image for output |
| CN106233278A (en) * | 2014-07-24 | 2016-12-14 | 惠普发展公司有限责任合伙企业 | Segment for image creates view data |
| US10079959B2 (en) | 2014-07-24 | 2018-09-18 | Hewlett-Packard Development Company, L.P. | Creating image data for a tile on an image |
| CN106909548A (en) * | 2015-12-22 | 2017-06-30 | 北京奇虎科技有限公司 | Picture loading method and device based on server |
| CN109582812A (en) * | 2018-11-21 | 2019-04-05 | 安徽云融信息技术有限公司 | A kind of search method for computer picture |
| CN112182276A (en) * | 2020-10-12 | 2021-01-05 | 云南财经大学 | Astronomical mass observation data hybrid retrieval method based on image content and metadata |
| CN112527210A (en) * | 2020-12-22 | 2021-03-19 | 南京中兴力维软件有限公司 | Storage method and device of full data and computer readable storage medium |
| CN113010722A (en) * | 2021-04-18 | 2021-06-22 | 南通大学 | Chronic disease clinical research queue query system and method based on fusion of fundus images |
| CN113010722B (en) * | 2021-04-18 | 2023-10-03 | 南通大学 | Slow clinical research queue query system and method for fusing fundus images |
| CN114896228A (en) * | 2022-04-27 | 2022-08-12 | 西北工业大学 | Industrial data stream cleaning model and method based on multi-stage combination optimization of filtering rules |
| CN114896228B (en) * | 2022-04-27 | 2024-04-05 | 西北工业大学 | Industrial data stream cleaning model and method based on filtering rule multistage combination optimization |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101930461A (en) | Digital image visualized management and retrieval for communication network | |
| CN111858954B (en) | Task-oriented text-generated image network model | |
| CN105069042A (en) | Content-based data retrieval methods for unmanned aerial vehicle spying images | |
| CN106126585B (en) | The unmanned plane image search method combined based on quality grading with perceived hash characteristics | |
| CN103218427B (en) | The extracting method of local description, image search method and image matching method | |
| CN105335469A (en) | Method and device for image matching and retrieving | |
| US9384519B1 (en) | Finding similar images based on extracting keys from images | |
| CN109299303B (en) | Hand-drawn sketch retrieval method based on deformable convolution and depth network | |
| Li et al. | Instant edit propagation on images based on bilateral grid | |
| CN110751271B (en) | Image traceability feature characterization method based on deep neural network | |
| Wang et al. | MindCamera: Interactive sketch-based image retrieval and synthesis | |
| CN113988147A (en) | Multi-label classification method and device for remote sensing image scene based on graph network, and multi-label retrieval method and device | |
| Zhou et al. | Attention transfer network for nature image matting | |
| CN109543525B (en) | Table extraction method for general table image | |
| CN112906517A (en) | Self-supervision power law distribution crowd counting method and device and electronic equipment | |
| KR20110019117A (en) | Semantic based image retrieval method | |
| CN116469172A (en) | Bone behavior recognition video frame extraction method and system under multiple time scales | |
| Fan et al. | Siamese graph convolution network for face sketch recognition: An application using graph structure for face photo-sketch recognition | |
| Cortés-Osorio et al. | Hybrid Algorithm for the detection of Pixel-based digital image forgery using Markov and SIFT descriptors | |
| KR101142163B1 (en) | Semantic based image retrieval method | |
| CN113411550B (en) | Video coloring method, device, equipment and storage medium | |
| Liang et al. | Agnet: Attention guided sparse depth completion using convolutional neural networks | |
| CN103617616A (en) | Affine invariant image matching method | |
| CN110162654A (en) | It is a kind of that image retrieval algorithm is surveyed based on fusion feature and showing for search result optimization | |
| CN114782688A (en) | Natural image segmentation method and system based on tensor subspace clustering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20101229 |