CN102479220A - Image retrieval system and method thereof - Google Patents
Image retrieval system and method thereof Download PDFInfo
- Publication number
- CN102479220A CN102479220A CN2010105665639A CN201010566563A CN102479220A CN 102479220 A CN102479220 A CN 102479220A CN 2010105665639 A CN2010105665639 A CN 2010105665639A CN 201010566563 A CN201010566563 A CN 201010566563A CN 102479220 A CN102479220 A CN 102479220A
- Authority
- CN
- China
- Prior art keywords
- image
- data
- depth
- destination object
- retrieval
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses an image retrieval system and a retrieval method. The image retrieval system comprises a mobile device, wherein the mobile device at least comprises an image capturing unit, a processing unit and an image data server, the image capturing unit is provided with two cameras, the two cameras respectively capture an input image of an object at the same time, the processing unit is coupled with the image capturing unit and is used for obtaining a depth image according to the input image and determining a target object according to characteristic information of the depth image and the input image, the image data server is coupled with the processing unit, receives the target object, retrieves the corresponding target object and obtains retrieval result data, and in addition, the retrieval result data is transmitted to the mobile device. The image retrieval system and the retrieval method have the advantages that the problems that when the mobile device is applied to the image retrieval, the whole image needs to be transmitted to a remote server, and the server needs to carry out mass operation can be solved, the load and the processing time of the server are reduced, and the use affinity and the convenience are improved.
Description
Technical field
The present invention relates to the application of a 3D computer vision image, be particularly to a kind of technical field of utilizing the mobile device pick-up image and carrying out video search.
Background technology
Present mobile device on the market, for example little pen electricity, panel computer, PDA, hand-held mobile device (MID) or intelligent mobile phone etc. all have video acquisition technology, let the user take pictures at any time or to make video recording.On the other hand; Because being widely used of video image, also occurred using correlation technique and product that image that video image captures special object is retrieved this image again at present on the market, but this type of technology mainly is to utilize mobile device or camera; Take 2D photo or image; Be sent to the server of rear end, server again with the technology of photo or its application of image carrying out background removal, characteristic acquisition etc., find out the specific objective object; Then with database in a large amount of image datas of being prestored compare, to find out the data that conform to.Because 2D photo/image is carrying out the sizable operands of operations needs such as background removal, characteristic acquisition, and quite consuming time, also be difficult for correctly finding the specific objective object, and be not suitable for the lower mobile device of resource.
Along with the development of multimedia application and relevant display technique, for producing the also increase of demand that more specifically reaches the display technique of real image (for example three-dimensional or three-dimensional video) more.Generally speaking; Physiologic factor based on beholder's stereoscopic vision; The for example vision difference between beholder's eyes (or so-called binocular parallax binocular parallax), motion parallax etc., the beholder can be perceived as solid or 3-dimensional image with the resultant image that is shown on the screen.
General hand-held mobile device or many of intelligent mobile phones has a camera lens at present; Therefore if will set up degree of depth image with depth information; Then need Same Scene is taken the image of at least two different visual angles; Yet this action is quite inconvenient concerning the user in operation, and manually takes two images often because of hand shake, angle coverage, shooting distance are difficult to precisely grasp, and the degree of depth image of therefore setting up is difficult for precisely usually.
On the other hand, the image retrieval system on the mobile device is many at present uses whole image to carry out data comparison and search with far-end server, retrieves quite consuming time; And accuracy rate is not high; When reason is to use whole image to compare, need reanalyse all objects and the characteristic thereof of whole image, not only cause the burden of far-end server; Also be prone to simultaneously cause system's erroneous judgement because of the indeterminate of destination object, accuracy rate reduces.And the analyses and comparison process is quite consuming time, and the user often need wait for that the time quite of a specified duration just can know the result, does not quite have the compatibility of use and convenience, causes to use wish not high.
Summary of the invention
Therefore the present invention is directed to above-mentioned variety of issue, propose a solution, utilize mobile device to obtain degree of depth image and capture destination object, be sent to the image data server again and retrieve to destination object with two video cameras.Because the degree of depth image that utilizes mobile device to capture; The characteristic information of degree of depth image capable of using is found out destination object fast; And mobile device also need not carry out background removal, characteristic acquisition etc. to the 2D image again; Even the mobile device that resource is lower also can be carried out, mobile device only is sent to the image data server with destination object and retrieves, and its data amount transmitted is low.Therefore; When the present invention can solve mobile device and is applied in video search; Must whole image be sent to far-end server and server must carry out the problem of a large amount of computings, reduce load of server and processing time, and improve compatibility and the convenience that uses.
In view of this, the present invention provides a kind of image retrieval system, and above-mentioned image retrieval system comprises: a mobile device comprises at least: an image acquisition unit, and it has two video cameras, and this pair video camera is simultaneously but respectively to an input of object acquisition image; And a processing unit, it is coupled to image acquisition unit, in order to obtaining a degree of depth image according to above-mentioned input image, and according to a characteristic information of input image and degree of depth image, to determine a destination object; And an image data server, it is coupled to processing unit, the receiving target object, and retrieval obtains result for retrieval data corresponding to destination object, and the result for retrieval data are sent to mobile device.
The present invention more provides a kind of method for retrieving image, and its step comprises: utilize two video cameras of a mobile device, simultaneously but respectively to an input of object acquisition image; Through above-mentioned mobile device, obtain a degree of depth image according to above-mentioned input image, and a characteristic information of above-mentioned input image of foundation and degree of depth image, to determine a destination object; And receive the image information of above-mentioned destination object through an image data server, and retrieval obtains result for retrieval data corresponding to above-mentioned destination object, and above-mentioned result for retrieval data are sent to above-mentioned mobile device.
Description of drawings
Accompanying drawing described herein is used to provide further understanding of the present invention, constitutes the application's a part, does not constitute qualification of the present invention.In the accompanying drawings:
Fig. 1 shows the calcspar according to the image retrieval system of the mobile device of one embodiment of the invention;
Fig. 2 shows the synoptic diagram according to two video camera imaging modes of one embodiment of the invention;
Fig. 3 shows the synoptic diagram of unique point descriptor according to an embodiment of the invention;
Fig. 4 shows that yardstick invariant features conversion method according to an embodiment of the invention calculates the process flow diagram of the image feature of destination object.
Drawing reference numeral:
100～image retrieval system;
110～mobile device;
111～image acquisition unit;
112～processing unit;
113～display unit;
120～image data server;
121～image process unit;
122～presentation content database;
S410, S420, S430, S440, S450, S460～step.
Embodiment
For making the object of the invention, technical scheme and advantage clearer, the embodiment of the invention is explained further details below in conjunction with accompanying drawing.At this, illustrative examples of the present invention and explanation thereof are used to explain the present invention, but not as to qualification of the present invention.
Fig. 1 shows the calcspar according to the image retrieval system of the embodiment of the invention.As shown in Figure 1; The present invention provides a kind of image retrieval system 100 of mobile device; Above-mentioned image retrieval system comprises a mobile device 110 and an image data server 120, and mobile device 110 comprises an image acquisition unit 111 and a processing unit 112 at least.In one embodiment of this invention, mobile device 110 can be hand-held mobile device, PDA, intelligent mobile phone etc., but is not limited thereto.
In one embodiment of this invention; Image acquisition unit 111 is one to have the device of two video cameras (dual camera); It comprises a left video camera and a right video camera, and two video camera simulating human binocular visions are in order to parallel shooting Same Scene; And synchronous indivedual input images of two video cameras about acquisition respectively; Indivedual input images that a left side video camera and right video camera are captured have parallax, use the technology of stereoscopic vision (stereo vision) by this, can obtain a degree of depth image (depth image).The degree of depth generation technique of stereovision technique comprises Block Matching algorithm, Dynamic Programming algorithm, Belief Propogation algorithm and Graph Cuts algorithm etc., but is not limited thereto.Two video cameras can adopt the commercially available product that gets, and its technology that obtains degree of depth image belongs to existing, does not specify at this.Processing unit 112 is coupled to image acquisition unit 111; Can be via existing stereovision technique; After the indivedual image inputs that receive two video cameras, obtain a degree of depth image, and the characteristic information of above-mentioned input image of foundation and degree of depth image; To determine a destination object, detailed ins and outs are of the back.The user also can adopt a range of interest (regions ofinterest) as destination object.Degree of depth image is one to have the image of depth information, and it has the positional information of two-dimensional coordinate (X, Y axle) and the information of depth value (Z axle), so degree of depth image can be expressed as a 3D image.Image data server 120 is coupled to processing unit 112, receives the destination object that processing unit 112 is sent, and retrieval corresponding to destination object to obtain result for retrieval data, then the result for retrieval data are sent to mobile device 110.In the time of further, the result for retrieval data possibly be the data of respective objects object, also possibly be to show not have to meet data retrieved.
In another embodiment of the present invention; Image acquisition unit 111 can be taken continuously; On mobile device 110; The user more can see through one group of specific keys (Fig. 1 does not show), in order to indivedual input images of two video cameras that captured of control image acquisition unit 111, and can select and confirm that tendency to develop gives indivedual input images of two video cameras of processing unit 112.Receive indivedual input images of two video cameras when processing unit 112 after; Promptly obtain a degree of depth image according to above-mentioned two indivedual input images; And calculate the characteristic information of above-mentioned input image and degree of depth image, in order to from above-mentioned degree of depth image, to determine a destination object.
In another embodiment of the present invention, image acquisition unit 111 more can only use a video camera to take input image continuously separately, and in processing unit 112, uses a degree of depth image algorithm, uses producing a degree of depth image.
In one embodiment of this invention, the characteristic information of input image and degree of depth image can be wherein at least one the information of the degree of depth, area, template, profile or characteristic topological relation.And processing unit 112 can select the most shallow object of a degree of depth as destination object according to the depth information of degree of depth image when the decision object; Or, after its normalization, determine its destination object according to the characteristic information of importing image and degree of depth image; Or select all more shallow candidate targets of a degree of depth, and calculate in the input image area of candidate target after degree of depth normalization; The object that selection meets the object area scope that stores in advance is used as destination object; Or be in the comparison input image whether the characteristic that meets a pair of pictograph shape/color/profile that stores in advance to be arranged, with the decision destination object.
One embodiment of two video camera imagings is as shown in Figure 2, O
lAnd O
rBe respectively the horizontal level of left video camera and right video camera, two video camera imaging modes can be tried to achieve with following triangle proportionate relationship:
Wherein T is the horizontal interval distance of two video cameras; Z is the straight line depth distance of the horizontal line mid point of two video cameras to object P; F is the actual focusing degree of depth of video camera; x
lAnd x
rBe respectively a left side and the horizontal level of the right formed image of video camera object of observation P when focal distance f, d is coordinate x
lAnd x
rDistance.
Generally speaking, because video camera or camera be when obtaining the 2D image, different with the distance distance of destination object because of camera lens, object area size or the unique point size also will change thereupon in the taken 2D image is unfavorable for finding out destination object.The present invention is more capable of using under different depth, and the area of destination object and the relation of change in depth calculate among the certain depth Z area A that destination object should comprise automatically
Real, from the 2D image, detecting in all candidate targets then, the object that selection conforms to the destination object area is used as destination object.The relational expression of the degree of depth and area is shown in the equation preface:
A
RealBe real object area, Z
UpWith Z
DownBe detectable maximum of this pair video camera capture device and minimum depth value.A
UpWith A
DownFor respectively at Z
UpWith Z
DownUnder two degree of depth environment, this destination object area size that detects in the 2D image, Z is the depth value of this candidate target object.
In another embodiment of the present invention, through above-mentioned triangle ratio relational expression, when the area size of object is fixed; Distance between photographic subjects object and this video camera is nearer, and in the acquisition 2D picture, the destination object of being seen will be healed greatly; And distance between the two is far away, and in the acquisition 2D picture, the destination object of being seen will be littler; May extend to the calculating of area thus, photographer can adjust the distance (that is object degree of depth Z) of reference object, so that the object area that photographs is a predetermined object area; At this moment, processing unit 112 just can directly capture the most close object of area size as destination object from the 2D image.If destination object has the fraction crested when taking, the information that processing unit 112 still can see through degree of depth image and area is correctly to capture destination object.
In another embodiment of the present invention; General photographer take through being everlasting a destination object the time after, can let destination object occupy the most of ratio in the image, if whole destination object is sent to the image data server; When the comparison characteristic; Still possibly cause sizable burden, at this moment, the user can see through specific keys or the operating function on the mobile device; Use the manual select target object of a square frame to have a part of scope of characteristic or part interested, to be sent to image data server 120.In one embodiment of this invention, image data server 120 sees through a sequence data communication interface, a cable network, a wireless network or a communication network, is coupled to above-mentioned processing unit, receiving above-mentioned destination object, but is not limited thereto.
In one embodiment of this invention; As shown in Figure 1; Image data server 120 more comprises an image process unit 121 and a presentation content database 122, and presentation content database 122 can store corresponding to a plurality of object image datas and corresponding a plurality of object datas thereof in advance, and the object image data can be an image feature of corresponding at least one object that prestores; Areal extent, shape, color, profile of object etc. for example prestore; The object that prestores can be the various objects that may be retrieved, or a certain specific object, similarly is to aim at the butterfly butterfly image database that information is set up is provided.And correspond respectively to the object data of each object image data; Can be respectively corresponding at least one data such as literal, sound, image and film of above-mentioned each object image data; But be not limited thereto, for example introduce closeup photograph of the literal of butterfly, image that butterfly dances in the air and sound, butterfly etc.
In another embodiment of the present invention; Image process unit 121 can be via an aspect ratio to algorithm; The destination object that analysis is determined via processing unit 112; Obtain the image feature of destination object, then the image feature of destination object and the object image data in the presentation content database 122 are compared, to judge whether that one of them conforms to the object image data.When conforming to, image process unit 121 captures the pairing object data of object image data that conforms to, as above-mentioned result for retrieval data from presentation content database 122.Generally speaking, judge whether to conform to, be meant when judging that similarity degree between the two surpasses a preset value, or its difference can think that then it is for conforming to during less than certain limit.
In the time of further; When image process unit 121 and presentation content database 122 stored object image datas carry out aspect ratio to the time; Need to calculate earlier the image feature of destination object; Yet in the 2D image, the image feature of object can change along with position, angle or rotational angle, and this is the character of a kind of noninvariance (non-invariant).In one embodiment of this invention, image process unit 121 utilizes yardstick invariant features conversion (Scale Invariant Feature Transform; Be designated hereinafter simply as SIFT) aspect ratio to algorithm to calculate the image feature of destination object; The object image data with the presentation content database carry out aspect ratio to before; Need to calculate earlier constant (invariant) characteristic of destination object, and the object image data captures corresponding to the image feature of each image in the presentation content database via the SIFT algorithm equally and is stored in advance in the presentation content database.
The image feature extraction comprises SIFT algorithm, template comparison algorithm, SURF algorithm etc. with the comparison mode, but is not limited thereto.
Fig. 4 shows according in one embodiment of the invention, calculates the process flow diagram of the image feature of destination object with yardstick invariant features conversion method, and it is to be used as image feature with the unique point on the image.At first at step S410; In one embodiment of this invention; The SIFT algorithm uses Difference of Gaussian (DoG) wave filter to set up a metric space (scalespace); And in metric space decision a plurality of local extremums (local extrema), local extremum can be the maximal value or the minimum value in zone, in order to as characteristic candidate value (feature candidate).Then at step S420; SIFT calculation rule is distinguished earlier and deleted some can not be as the local extremum of eigenwert; Like the local extremum of low contrast (contrast), or the local extremum of edge (edge), the method also is called accurate positioning feature point (accurate keypoint localization); For instance, distinguish that the method that contrasts low local extremum is to use a 3D quadratic power program to represent:
Wherein D is the result of DoG wave filter; X is a local extremum, and
is a deviate.If the absolute value of
is less than a predetermined value, then
The corresponding area extreme value then is a low correlative value.
At step S430; After the method for utilizing above-mentioned accurate positioning feature point is found out unique point (keypoint); Each unique point is calculated the size and the direction of its gradient; And use the method for a direction histogram (orientation histogram); The method is considered each unique point gradient direction of interior each pixel of a form frame on every side, and the direction of the court of gradient institute of pixels is a main direction (major orientation) at most, and the weight (weight) of each pixel around the unique point; Be the gradient magnitude that a Gaussian distribution (Gaussiandistribution) is multiplied by this pixel again and decide, step S430 also can be described as direction and specifies (orientationassignment).
By above-mentioned steps S410～S430, can obtain position, size and the direction of each unique point, at step S440; Near 8 * 8 form frames each pixel of destination object are cut into the sub-window frame (sub-window) of 2 * 2 sizes, and add up the direction histogram of each 2 * 2 sub-window frame, determine the direction of each 2 * 2 sub-window frame equally according to the method for step S430; Extend to its 4 * 4 corresponding sub-window frames; Therefore, each 4 * 4 sub-window frame can have 8 directions, and available 8 bits are represented; And each pixel has 4 * 8=32 direction; Available 32 bits represent that as shown in Figure 3, this promptly is called area image descriptor (local image descriptor) or unique point descriptor (keypoint descriptor).
When the area image descriptor of obtaining destination object; Can carry out aspect ratio to (feature matching) to each picture in the presentation content database or the pairing unique point descriptor of object; If adopt violence (brute force) comparison method, will quite expend calculation resources and time.In one embodiment of this invention; It is right that the algorithm that adopts K-D Tree at step S450 is carried out aspect ratio to the unique point descriptor of the unique point descriptor of destination object and each picture in the presentation content database; K-D Tree algorithm is made a K-Dtree respectively to the pairing unique point descriptor of each picture in the presentation content database earlier; Come again to carry out K closest value search (k-nearest neighborsearching) with regard to each unique point descriptor with each picture; The k value can be an adjusted value, that is concerning some unique point descriptors, can be set in each data picture k as characteristic; Each the unique point descriptor that can set up each data picture thus to the aspect ratio of other each data pictures to relation; When having new destination object to desire to compare, can be according to the unique point of above-mentioned K-D tree methods analyst destination object, and in presentation content database 122, hunt out object image data fast near destination object; Simultaneously can also significantly lower operand, save search time.
At step S460, according to the picture that hunts out, can in presentation content database 122, find index type (model type indexing) near the picture of destination object, the related data corresponding with it links (data link).Then, image data server 120 can be sent to the object data of the destination object that hunts out mobile device 110.
In one embodiment of this invention; Mobile device 110 more can comprise a display unit 113, and when mobile device 110 received the result for retrieval data that transmit from image data server 120, processing unit 112 can show the result for retrieval data on display unit 113; In the time of further; Can be according to user's selection, with the result for retrieval data in destination object other or the screen corner of display unit 113 or an ad-hoc location, at this moment; When image acquisition unit 111 continues to take continuous image, processing unit 112 more sustainably with continuous image and result for retrieval data presentation on display unit 113.In another embodiment of the present invention, when being a butterfly like destination object, presentation content database 122 can provide species, profile data, binding webpage or other relevant picture of butterfly, in order to the related data as search result, but is not limited thereto.
Method for retrieving image in one embodiment of the invention, it comprises:
Step 2 through mobile device 110, obtains a degree of depth image according to the image of importing, and according to the characteristic information of input image and degree of depth image, determines a destination object then.Wherein, characteristic information can be with the degree of depth, area, template, profile and characteristic topological relation at least one relevant information.
Step 3 through image data server 120 receiving target objects, and is retrieved corresponding to destination object, obtains result for retrieval data, then the result for retrieval data is sent to mobile device 110.Wherein, The image data server more includes presentation content database 122; Store a plurality of object image datas and corresponding object data; The object image data is an image feature of at least one object that prestores, and object data is the data such as literal, sound, image or film of corresponding each object image data.
Mobile device in the above-mentioned steps, image data server and description of Related Art etc. are all as noted earlier, so repeat no more.
Method of the present invention; Or specific kenel or its part, can be contained in tangible media with the kenel of procedure code, like floppy disk, laser disc, hard disk or any other machine-readable (getting) Storage Media like computer-readable; Wherein, When procedure code by machine, when being written into and carrying out like computer, this machine becomes in order to participate in device of the present invention or system.
Method of the present invention, system and device also can see through some transfer mediums with the procedure code kenel; Transmit like electric wire or cable, optical fiber or any transmission kenel; Wherein, When procedure code by machine, such as computer, electronic equipment reception, when being written into and carrying out, this machine becomes in order to participate in device of the present invention or system.When the general service processor is done in fact, the procedure code associative processor provides a class of operation to be similar to the unique apparatus of using particular logic circuit.
The above person of thought; Be merely preferred embodiment of the present invention; When not limiting the scope that the present invention implements with this, the simple equivalent of promptly doing according to claim of the present invention and invention description generally changes and modifies, and all still belongs in the scope that patent of the present invention contains.Arbitrary embodiment of the present invention in addition or claim must not reached whole purposes or advantage or the characteristics that the present invention discloses.In addition, summary part and title only are the usefulness in order to auxiliary patent document search, are not in order to limit interest field of the present invention.
Claims (19)
1. an image retrieval system is characterized in that, said image retrieval system comprises:
One mobile device comprises at least:
One image acquisition unit, it has two video cameras, and the said pair of video camera is simultaneously but respectively to an input of object acquisition image; And
One processing unit, it is coupled to said image acquisition unit, obtains a degree of depth image in order to the said input image of foundation, and a characteristic information of said input image of foundation and degree of depth image, to determine a destination object;
And
One image data server, it is coupled to said processing unit, receives said destination object, and retrieval obtains result for retrieval data corresponding to said destination object, and said result for retrieval data is sent to said mobile device.
2. image retrieval system as claimed in claim 1 is characterized in that, said characteristic information be with the degree of depth, area, template, profile and characteristic topological relation at least one relevant information.
3. image retrieval system as claimed in claim 2; It is characterized in that said characteristic information comprises a depth information at least, said processing unit is more with reference to said depth information; So that said characteristic information is carried out normalization, and determine the said destination object in the said input image according to this.
4. image retrieval system as claimed in claim 1 is characterized in that, said characteristic information is a depth information, and said processing unit said depth information more capable of using determines that the most shallow the most preceding scenery of the degree of depth is said destination object in the said degree of depth image.
5. image retrieval system as claimed in claim 1 is characterized in that, said characteristic information comprises a depth information and an area information at least, and said destination object meets an object of a preset range for its area in said degree of depth image and the degree of depth.
6. image retrieval system as claimed in claim 1 is characterized in that, said image data server sees through a sequence data communication interface, a cable network, a wireless network or a communication network, is coupled to said processing unit, to receive said destination object.
7. image retrieval system as claimed in claim 1; It is characterized in that; Said image data server more comprises a presentation content database; In order to store a plurality of object image datas and corresponding a plurality of object datas thereof, wherein said object image data is an image feature of corresponding at least one object that prestores, and said object data is respectively corresponding at least one data such as literal, sound, image and film of said each object image data.
8. image retrieval system as claimed in claim 7; It is characterized in that; Said image data server more comprises an image process unit, in order to use an aspect ratio to algorithm to analyze said destination object, obtain the image feature of said destination object; And the image feature and the said object image data of said destination object are compared, and whether one of them conforms to said object image data to judge said destination object; And; When one of them conformed to when said destination object and said object image data, said image process unit more captures the object image data that conforms to corresponding to said judgement from said presentation content database said object data was as said result for retrieval data.
9. image retrieval system as claimed in claim 1 is characterized in that said mobile device more comprises a display unit, when said mobile device receives said result for retrieval data, shows said destination object and said result for retrieval data at said display unit.
10. image retrieval system as claimed in claim 9 is characterized in that, when said image acquisition unit continues to take a plurality of continuous image, continues to show said continuous image and said result for retrieval data at said display unit.
11. a method for retrieving image is characterized in that, its step comprises:
Utilize two video cameras of a mobile device, simultaneously but respectively to an input of object acquisition image;
Through said mobile device, obtain a degree of depth image according to said input image, and a characteristic information of said input image of foundation and degree of depth image, to determine a destination object; And
Receive said destination object through an image data server, and retrieval obtains result for retrieval data corresponding to said destination object, and said result for retrieval data are sent to said mobile device.
12. method for retrieving image as claimed in claim 11 is characterized in that, said characteristic information be with the degree of depth, area, template, profile and characteristic topological relation at least one relevant information.
13. method for retrieving image as claimed in claim 12 is characterized in that, said characteristic information comprises a depth information at least, and said method more comprises:
Through said mobile device, with reference to said depth information, so that said characteristic information is carried out normalization, and determine the said destination object in the said input image according to this.
14. method for retrieving image as claimed in claim 11 is characterized in that, said characteristic information is a depth information, and said method more comprises:
Through said mobile device, utilize said depth information, determine that the most shallow the most preceding scenery of the degree of depth is said destination object in the said degree of depth image.
15. method for retrieving image as claimed in claim 11 is characterized in that, said characteristic information comprises a depth information and an area information at least, and said destination object meets an object of a preset range for its area in said degree of depth image and the degree of depth.
16. method for retrieving image as claimed in claim 11; It is characterized in that; Said image data server more comprises a presentation content database; In order to store a plurality of object image datas and corresponding a plurality of object datas thereof, wherein said object image data is an image feature of corresponding at least one object that prestores, and said object data is respectively corresponding at least one data such as literal, sound, image and film of said each object image data.
17. method for retrieving image as claimed in claim 16 is characterized in that, said method more comprises:
Through said image data server; Use an aspect ratio to algorithm to analyze said destination object; Obtain the image feature of said destination object; And the image feature and the said object image data of said destination object are compared, and whether one of them conforms to said object image data to judge said destination object; And when one of them conformed to when said destination object and said object image data, the said object data that from said presentation content database, captures the object image data that conforms to corresponding to said judgement was as said result for retrieval data.
18. method for retrieving image as claimed in claim 11 is characterized in that, said method more comprises:
Through a display unit of said mobile device, when said mobile device receives said result for retrieval data, show said destination object and said result for retrieval data at said display unit.
19. method for retrieving image as claimed in claim 18 is characterized in that, said method more comprises:
When said mobile device continues to take a plurality of continuous image, continue to show said continuous image and said result for retrieval data at said display unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105665639A CN102479220A (en) | 2010-11-30 | 2010-11-30 | Image retrieval system and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105665639A CN102479220A (en) | 2010-11-30 | 2010-11-30 | Image retrieval system and method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102479220A true CN102479220A (en) | 2012-05-30 |
Family
ID=46091868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010105665639A Pending CN102479220A (en) | 2010-11-30 | 2010-11-30 | Image retrieval system and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102479220A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103546803A (en) * | 2012-07-11 | 2014-01-29 | 腾讯科技（深圳）有限公司 | Image processing method, client side and image processing system |
| CN104200189A (en) * | 2014-08-27 | 2014-12-10 | 苏州佳世达电通有限公司 | Barcode scanning device and processing method thereof |
| CN104506768A (en) * | 2014-11-28 | 2015-04-08 | 广东欧珀移动通信有限公司 | Method and device for image selection as well as terminal |
| US20160268571A1 (en) * | 2013-12-26 | 2016-09-15 | Teijin Limited | Non-aqeous secondary cell separator and non-aqueous secondary cell |
| CN108931202A (en) * | 2018-07-13 | 2018-12-04 | Oppo广东移动通信有限公司 | Detection method and device, electronic device, computer equipment and readable storage medium storing program for executing |
| CN108965525A (en) * | 2018-07-13 | 2018-12-07 | Oppo广东移动通信有限公司 | Detection method and device, terminal, computer equipment and readable storage medium storing program for executing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101277454A (en) * | 2008-04-28 | 2008-10-01 | 清华大学 | Method for generating real time tridimensional video based on binocular camera |
| CN101630406A (en) * | 2008-07-14 | 2010-01-20 | 深圳华为通信技术有限公司 | Camera calibration method and camera calibration device |
-
2010
- 2010-11-30 CN CN2010105665639A patent/CN102479220A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101277454A (en) * | 2008-04-28 | 2008-10-01 | 清华大学 | Method for generating real time tridimensional video based on binocular camera |
| CN101630406A (en) * | 2008-07-14 | 2010-01-20 | 深圳华为通信技术有限公司 | Camera calibration method and camera calibration device |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103546803A (en) * | 2012-07-11 | 2014-01-29 | 腾讯科技（深圳）有限公司 | Image processing method, client side and image processing system |
| CN103546803B (en) * | 2012-07-11 | 2016-09-21 | 腾讯科技（深圳）有限公司 | A kind of system of the method for image procossing, client and image procossing |
| US9536285B2 (en) | 2012-07-11 | 2017-01-03 | Tencent Technology (Shenzhen) Company Limited | Image processing method, client, and image processing system |
| US20160268571A1 (en) * | 2013-12-26 | 2016-09-15 | Teijin Limited | Non-aqeous secondary cell separator and non-aqueous secondary cell |
| CN104200189A (en) * | 2014-08-27 | 2014-12-10 | 苏州佳世达电通有限公司 | Barcode scanning device and processing method thereof |
| CN104200189B (en) * | 2014-08-27 | 2017-05-03 | 苏州佳世达电通有限公司 | Barcode scanning device and processing method thereof |
| CN104506768A (en) * | 2014-11-28 | 2015-04-08 | 广东欧珀移动通信有限公司 | Method and device for image selection as well as terminal |
| CN108931202A (en) * | 2018-07-13 | 2018-12-04 | Oppo广东移动通信有限公司 | Detection method and device, electronic device, computer equipment and readable storage medium storing program for executing |
| CN108965525A (en) * | 2018-07-13 | 2018-12-07 | Oppo广东移动通信有限公司 | Detection method and device, terminal, computer equipment and readable storage medium storing program for executing |
| CN108931202B (en) * | 2018-07-13 | 2021-05-07 | Oppo广东移动通信有限公司 | Detection method and apparatus, electronic apparatus, computer device, and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220012495A1 (en) | Visual feature tagging in multi-view interactive digital media representations | |
| EP3099056B1 (en) | Method and apparatus for displaying a light field based image on a user's device, and corresponding computer program product | |
| US20120127276A1 (en) | Image retrieval system and method and computer product thereof | |
| US8928736B2 (en) | Three-dimensional modeling apparatus, three-dimensional modeling method and computer-readable recording medium storing three-dimensional modeling program | |
| US10089762B2 (en) | Methods for navigating through a set of images | |
| CN107197153B (en) | Shooting method and shooting device for photo | |
| CN102479220A (en) | Image retrieval system and method thereof | |
| Nocerino et al. | 3D reconstruction with a collaborative approach based on smartphones and a cloud-based server | |
| CN105023266A (en) | Method and device for implementing augmented reality (AR) and terminal device | |
| CN107430498B (en) | Extending the field of view of a photograph | |
| US20200349187A1 (en) | Method and apparatus for data retrieval in a lightfield database | |
| CN101853268A (en) | Image searching system and image searching method | |
| KR101764424B1 (en) | Method and apparatus for searching of image data | |
| CN103428537A (en) | Video processing method and video processing device | |
| WO2023098737A1 (en) | Three-dimensional reconstruction method, electronic device, and computer-readable storage medium | |
| EP3624050B1 (en) | Method and module for refocusing at least one plenoptic video | |
| KR101334980B1 (en) | Device and method for authoring contents for augmented reality | |
| KR101640020B1 (en) | Augmentated image providing system and method thereof | |
| EP3099078A1 (en) | Method for collecting information on users of 4d light field data, corresponding apparatuses and computer programs | |
| CN117392482A (en) | Multi-mode human face anti-counterfeiting model training and human face anti-counterfeiting recognition method | |
| JP2004213278A (en) | Image processing apparatus, image processing method, image processing program, storage medium, and electronic apparatus |
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: 20120530 |