CN110472085A - 3-D image searching method, system, computer equipment and storage medium - Google Patents
3-D image searching method, system, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110472085A CN110472085A CN201910656211.3A CN201910656211A CN110472085A CN 110472085 A CN110472085 A CN 110472085A CN 201910656211 A CN201910656211 A CN 201910656211A CN 110472085 A CN110472085 A CN 110472085A
- Authority
- CN
- China
- Prior art keywords
- image
- dimensional image
- angle
- angle point
- dimensional
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/55—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to technical field of image matching, the present invention provides a kind of 3-D image searching method and system, and the above method includes: the first two dimensional image for obtaining object search, calculates the first angle point of the first two dimensional image；Second angle point of the first angle point the second two dimensional image corresponding with 3-D image each in image data base is matched, target two dimensional image is determined from the second two dimensional image；Calculate the subtense angle between the first two dimensional image and target two dimensional image；Subtense angle is less than 3-D image corresponding to the target two dimensional image of first threshold and is determined as search result.Above-mentioned 3-D image searching method carries out the matching of image by angle point readily identified in two dimensional image according to the first two dimensional image of object search easy to collect, to search out 3-D image and as search result, realizes the search of 3-D image.Moreover, two dimensional image is easy to get in this method, and the angle point as feature is also easy to identify, substantially increases the efficiency of search.
Description
Technical field
The present invention relates to technical field of image matching, specifically, the present invention relates to a kind of 3-D image searching methods, three
Tie up image search system, computer equipment and storage medium.
Background technique
3-D image, the stereo-picture that is otherwise known as, depth image, three-dimensional digitalization model, 3-D image contain really
The three-dimensional coordinate of subject surface and the complete information of gray scale, record the three dimensional space coordinate of each sampled point of body surface.Three
Dimension image makes a big difference with traditional flat image.Traditional flat image can be regarded as the brightness in two-dimensional space point
Cloth, flat image are obtained from being projected in two dimensional image plane as real three-dimensional world.
Information entrained by 3-D image is more, can be used for showing the shape and structure of product three-dimensionally and in all directions
It makes.With advances in technology and the development of business, the 3-D image for being shown to product are just being widely applied, and inevitably increase
It has grown to 3-D image search and has searched the demand of corresponding product.
Information entrained by 3-D image is more, although can pass through the solid of observation three-dimensional image perception 3-D image
Sense, but the two dimensional image with recording three-dimensional image under some visual angle can only be acquired by image collecting devices such as cameras, because
This is usually that can not directly obtain three dimensional space coordinate described in 3-D image, it is difficult to by acquisition 3-D image to the three-dimensional
Image scans for.
Accordingly, there exist can not carry out 3-D image search by image acquisition device 3-D image.
Summary of the invention
The purpose of the present invention is intended at least can solve above-mentioned one of technological deficiency, can not especially pass through image collector
Set the technological deficiency that acquisition 3-D image carries out 3-D image search.
The present invention provides a kind of 3-D image searching method, includes the following steps:
The first two dimensional image of object search is obtained, the first angle point of first two dimensional image is calculated；
By the second angle point of first angle point the second two dimensional image corresponding with 3-D image each in image data base
It is matched, target two dimensional image is determined from second two dimensional image；
Calculate the subtense angle between first two dimensional image and the target two dimensional image；
Subtense angle is less than 3-D image corresponding to the target two dimensional image of first threshold and is determined as search result.
In one embodiment, first two dimensional image for obtaining object search, calculates first two dimensional image
The step of first angle point, comprising:
Two original images of described search object are obtained by picture pick-up device within a preset time interval；Calculate separately institute
State the corresponding third angle point of original image and fourth angle point；Angle point sieve is carried out to the third angle point according to the fourth angle point
Choosing obtains first angle point.
In one embodiment, described by first angle point and 3-D image corresponding second each in image data base
The step of second angle point of two dimensional image is matched, and target two dimensional image is determined from second two dimensional image, comprising:
According to the second angle point and first angle point of the second two dimensional image each in described image database, each institute is calculated
State the first similarity magnitude between the second two dimensional image and first two dimensional image；The first similarity magnitude is chosen to exist
The second two dimensional image in second preset threshold is as the target two dimensional image.
In one embodiment, first calculated between each second two dimensional image and first two dimensional image
The step of similarity magnitude, comprising:
Corresponding second angle point of the second two dimensional image of any one 3-D image is obtained from described image database；It will
First angle point and second angle point are matched, and the angle point pair of same position is extracted；According to the angle point to calculate this
First similarity magnitude of two two dimensional images.
In one embodiment, second X-Y scheme for choosing the first similarity magnitude in the second preset threshold
As the step of as the target two dimensional image, comprising:
Obtain the corresponding visual angle weight in 3-D image of each second two dimensional image；According to second two dimensional image
Visual angle weight and second two dimensional image the first similarity magnitude calculation described in the second two dimensional image the second similarity
Magnitude；Second two dimensional image of the second similarity magnitude in the second preset threshold is chosen as the target X-Y scheme
Picture.
In one embodiment, the subtense angle calculated between first two dimensional image and the target two dimensional image
The step of, comprising:
The coordinate respectively in the target two dimensional image and first two dimensional image is established according to the angle point saturating
Depending on transformation matrix；The subtense angle is calculated according to the perspective transformation matrix.
In one embodiment, subtense angle is less than three-dimensional figure corresponding to the target two dimensional image of first threshold described
After as the step of being determined as search result, comprising:
According to the subtense angle and target two dimensional image visual angle in 3-D image, first two dimensional image is calculated
Observation visual angle；The target three-dimensional image is shown in the observation visual angle.
The present invention also provides a kind of 3-D image search systems, comprising:
Angle point computing module calculates the of first two dimensional image for obtaining the first two dimensional image of object search
One angle point；
Two dimensional image matching module, for by first angle point corresponding with 3-D image each in image data base the
Second angle point of two two dimensional images is matched, and target two dimensional image is determined from second two dimensional image；
Subtense angle computing module, for calculating the visual angle between first two dimensional image and the target two dimensional image
Difference；
Acquiring three-dimensional images module, for subtense angle to be less than to three-dimensional figure corresponding to the target two dimensional image of first threshold
As being determined as search result.
The present invention also provides a kind of computer equipment, including memory, processor and storage on a memory and can located
The computer program run on reason device, the processor are realized three described in any of the above-described embodiment when executing the computer program
The step of tieing up image search method.
The present invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, the computer journey
The step of 3-D image searching method described in any of the above-described embodiment is realized when sequence is executed by processor.
Above-mentioned 3-D image searching method, system, computer equipment and storage medium, according to search pair easy to collect
The first two dimensional image of elephant carries out the matching of image by angle point readily identified in two dimensional image, to search out three-dimensional figure
As and as search result, realize the search of 3-D image.Moreover, two dimensional image is easy to get in this method, and as feature
Angle point be also easy to identify, substantially increase the efficiency of search；The three-dimensional character for considering object search simultaneously, also according to subtense angle
Judgement improve 3-D image search accuracy.The additional aspect of the present invention and advantage will partially give in the following description
Out, these will become apparent from the description below, or practice through the invention is recognized.
Detailed description of the invention
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, in which:
Fig. 1 is the implementation environment figure of the 3-D image searching method provided in one embodiment；
Fig. 2 is the flow chart of 3-D image searching method in one embodiment；
Fig. 3 is the flow chart of 3-D image searching method in another embodiment；
Fig. 4 is the structural schematic diagram of 3-D image search system in one embodiment；
Fig. 5 is the schematic diagram of internal structure of computer equipment in one embodiment.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and for explaining only the invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in specification of the invention
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member
Part is " connected " or when " coupled " to another element, it can be directly connected or coupled to other elements, or there may also be
Intermediary element.In addition, " connection " used herein or " coupling " may include being wirelessly connected or wirelessly coupling.It is used herein to arrange
Diction "and/or" includes one or more associated wholes for listing item or any cell and all combinations.
As shown in FIG. 1, FIG. 1 is the implementation environment figures of the 3-D image searching method provided in one embodiment, in the reality
It applies in environment, including server-side 110 and client 120.
Image data base can be set in server-side 110, can store 3-D image and three-dimensional in image data base
The two dimensional image that image is shown under visual angle, server-side 110 can be built in computer equipment or server apparatus.Visitor
Photographic device can be configured on family end 120, client 120 can send two dimensional image to server-side 110.Client 120 can
Operate in smart phone, tablet computer, laptop, on desktop computer.
In one embodiment, as shown in Fig. 2, Fig. 2 is the flow chart of 3-D image searching method in one embodiment, this
A kind of 3-D image searching method is proposed in embodiment, which can be applied to above-mentioned server-side
In 110, it can specifically include following steps:
Step S210: obtaining the first two dimensional image of object search, calculates the first angle point of the first two dimensional image.
In this step, during being scanned for according to the first two dimensional image to object search, it can be examined according to angle point
It surveys to calculate the first angle point of the first two dimensional image.
Wherein, object search can be the three-dimensional object with three-dimensional structure, be also possible to three dimensional space coordinate
The visual 3-D image of information, search image obtain two dimensional image after capable of being collected on visual angle.3-D image can be by
Observation multi-angle has unlimited, but can reasonably select the one group of visual angle that can cover 3-D image comprehensively, is regarded according to the group
Angle acquires the corresponding multiple two dimensional images of 3-D image.In addition, for some 3-D images with symmetric property, it can be appropriate
Reduce the quantity of the two dimensional image of acquisition.
Wherein, intuitively under overview, angle point has feature below: can be the angle point between profile；For same field
Scape is generally configured with the feature of stability property even if visual angle changes；No matter the pixel of the near zone is in gradient direction
There is large change on upper or its gradient magnitude.Angle point is feature that is intuitive in image and being easy to detect, is also convenient for being closed
Suitable images match.
Specifically, the first two dimensional image of object search is obtained in step S210, calculates first jiao of the first two dimensional image
The step of point, may include:
S211: two original images of object search are obtained by picture pick-up device within a preset time interval.
It, can be with the two dimensional image of Quick Acquisition object search, wherein acquired two by picture pick-up device in this step
Original image belongs to two dimensional image.
S212: the corresponding third angle point of original image and fourth angle point are calculated separately.
S213: angle point screening is carried out to the third angle point according to fourth angle point, obtains the first angle point.
In the mode of above-mentioned the first angle point of calculating, due to having certain similitude between two original images, calculated
Third angle point and fourth angle point between can also have a similitude, therefore angle point screening can be carried out according to relationship between angle point,
Noise caused by excluding because of camera shooting acquisition, or the interference of angle point present in background can be excluded, improve the standard of the first angle point
True property.
Step S220: by the second of the first angle point the second two dimensional image corresponding with 3-D image each in image data base
Angle point is matched, and target two dimensional image is determined from the second two dimensional image.
In this step, is matched and searched in image data base by the first angle point, determining and the first two dimensional image
Similar target two dimensional image.
Wherein, corresponding second two dimensional image of 3-D image is store in image data base, these second two dimensional images can
To be also possible to specially calculate to 3-D image from for showing two dimensional image of the 3-D image under certain visual angle
Two dimensional image under certain visual angle.Since 3-D image has three-dimensional three-dimensional character, entrained information is more, general single view
Two dimensional image under angle is merely able to locally show 3-D image, and the 3-D image generally in image data base can be corresponded to and be deposited
Contain the second two dimensional image under multiple visual angles.
Step S230: the subtense angle between the first two dimensional image and target two dimensional image is calculated.
In this step, can be by visual transform the characteristics of, according to the first angle point and target of the first two dimensional image two dimension
Corresponding second angle point of image can calculate the subtense angle between the first two dimensional image and target two dimensional image.
Step S240: subtense angle is less than 3-D image corresponding to the target two dimensional image of first threshold and is determined as searching for
As a result.
In this step, the accuracy of corners Matching is further judged, whether in error range subtense angle is detected, if visual angle
In error range, i.e., subtense angle is less than first threshold to difference, then is determined as searching by 3-D image corresponding to target two dimensional image
Hitch fruit.
Above-mentioned 3-D image searching method, according to the first two dimensional image of object search easy to collect, by X-Y scheme
Readily identified angle point, carries out the matching of image as in, to search out 3-D image and as search result, realizes three-dimensional figure
The search of picture.Moreover, two dimensional image is easy to get in this method, and the angle point as feature is also easy to identify, substantially increases
The efficiency of search；The three-dimensional character for considering object search simultaneously improves the standard of 3-D image search also according to the judgement of subtense angle
True property.
Corners Matching can be divided into following four step in one of the embodiments: 1. extract detection son: two to
Those pixels (angle point), such as the object edge point of texture-rich etc. for being easiest to identification are found in matched image.2. mentioning
Take description: for the angle point detected, being described with some features mathematically, such as histogram of gradients, part with
Machine binary feature etc..3. matching: judging their corresponding relationships in two images by description of each angle point.4. going
Exterior point: removing the exterior point of erroneous matching, retains point in correctly.Wherein, Euclidean distance matching strategy one: is done to description subvector
Measurement, a simplest strategy are exactly first to set a threshold value (maximum distance), are then returned within this threshold range
Another image in all matchings.Alternatively, matching strategy two: do arest neighbors matching, that is, compare nearest neighbor distance and time
The ratio of nearest neighbor distance, i.e. arest neighbors ratio (NNDR).
In one embodiment, by the first angle point corresponding with 3-D image each in image data base in step S220
Second angle point of two two dimensional images is matched, and from the second two dimensional image the step of determining target two dimensional image, may include:
S221: according to the second angle point and the first angle point of the second two dimensional image each in image data base, each second is calculated
The first similarity magnitude between two dimensional image and the first two dimensional image.
In this step, between the second angle point and the first angle point can be more with matched angle point and similitude is higher, then
One similarity magnitude is higher, and the similarity between corresponding second two dimensional image and the first two dimensional image is higher.
Specifically, the step of the first similarity magnitude between each second two dimensional image and the first two dimensional image is calculated in S221
Suddenly, may include:
A1: corresponding second angle point of the second two dimensional image of any one 3-D image is obtained from image data base.
A2: the first angle point and second angle point are matched, the angle point pair of same position is extracted.
In this step, the angle point pair of each group in the first two dimensional image and the second two dimensional image can be matched one by one.
A3: according to angle point to the first similarity magnitude for calculating second two dimensional image.
It, can be according to all angle points to determining first similarity magnitude in this step.
S222: second two dimensional image of the first similarity magnitude in the second preset threshold is chosen as target X-Y scheme
Picture.
In this step, the second preset threshold can be used for screening the similarity magnitude for meeting precision, meet in order to choose
Second two dimensional image of the second preset threshold improves the accuracy of target two dimensional image.
Specifically, second two dimensional image of the first similarity magnitude in the second preset threshold is chosen in S222 as target
The step of two dimensional image may include:
B1: the corresponding visual angle weight in 3-D image of each second two dimensional image is obtained.
In this step, 3-D image may include second two dimensional image at multiple visual angles, and some of them visual angle is common sight
A possibility that examining visual angle, acquiring the first two dimensional image under common observation visual angle is larger, therefore can also further consider visual angle
Influence to images match.In addition each second two dimensional image can be set according to the common degree and importance at visual angle in advance
Visual angle weight.
B2: according to the first similarity magnitude calculation the two or two of the visual angle weight of the second two dimensional image and the second two dimensional image
Tie up the second similarity magnitude of image.
In this step, influence of the visual angle to images match is considered, according to visual angle weight and the first similarity magnitude calculation the
Two similarity measure values.For example, can be using visual angle weight and the product of the first similarity magnitude as the second similarity magnitude.
B3: second two dimensional image of the second similarity magnitude in the second preset threshold is chosen as target two dimensional image.
The second preset threshold can be in this step, in this step consider visual angle weight after screening meet the similar of precision
Metric, in order to choose accurate target two dimensional image.
Above-mentioned 3-D image searching method, it is similar to progress by angle point between the second two dimensional image and the first two dimensional image
The calculating of metric determines the higher target two dimensional image of accuracy convenient for the comparison of subsequent follow-up similarity magnitude.
In one embodiment, the subtense angle between the first two dimensional image and target two dimensional image is calculated in step S230
Step may include:
S231: perspective transform square is established to the coordinate respectively in target two dimensional image and the first two dimensional image according to angle point
Battle array.
In this step, visual angle difference, perspective transformation matrix can be used between target two dimensional image and the first two dimensional image
Indicate same position mapping relations in target two dimensional image and the first two dimensional image, perspective transformation matrix is can be used to indicate that view
The mapping relations converted between angle.
S232: subtense angle is calculated according to perspective transformation matrix.
Above-mentioned 3-D image searching method, perspective transformation matrix can contain target two dimensional image and the first two dimensional image
Between visual angle incidence relation, can extract by the perspective transformation matrix that converts and represent the ginseng of subtense angle in perspective transformation matrix
Number, or according to the coaptation subtense angle in perspective transformation matrix.
In one embodiment, subtense angle is less than corresponding to the target two dimensional image of first threshold in step S240
3-D image was determined as after the step of search result, comprising:
S251: the visual angle in 3-D image according to subtense angle and target two dimensional image calculates the observation of the first two dimensional image
Visual angle.
In this step, this observation visual angle can be present when the first two dimensional image is practical to be shown in 3-D image
Visual angle；Wherein target two dimensional image is derived from the second two dimensional image of 3-D image in image data base, therefore target two dimension
Image visual angle locating in 3-D image be it is known, can calculate that the observation of the first two dimensional image regards further according to subtense angle
Angle.
S252: target three-dimensional image is shown in observation visual angle.
In this step, target three-dimensional image is placed in the observation visual angle and is shown, the effect shown with to searching
Rope object the first two dimensional image collected is similar, above-mentioned 3-D image searching method, is conducive at same view angle or close visual angle
To comparing observation between object search and target three-dimensional image.
In addition, for when target three-dimensional image may include multiple, it can also be according to the visual angle of each target two dimensional image
Weight is ranked up target three-dimensional image, preferential to show the corresponding higher target three-dimensional image of visual angle priority, i.e., preferentially shows
Show in the 3-D image being easy under viewing angle, convenient for user according to the target three-dimensional figure for quickly understanding displaying under common visual angle
The shape of picture, and it is easy to the common visual angle effect of user to the visual angle of the first two dimensional image, it improves readable.
In another embodiment, as shown in figure 3, Fig. 3 is the process of 3-D image searching method in another embodiment
Scheme, provide a kind of 3-D image searching method in the present embodiment, can specifically include following steps:
S310: obtaining the first two dimensional image of object search, calculates the first angle point of the first two dimensional image.
Two original images of object search are obtained by picture pick-up device within a preset time interval；Calculate separately original graph
As corresponding third angle point and fourth angle point；Angle point screening is carried out to the third angle point according to fourth angle point, obtains the first angle point.
S320: by the second angle point of the first angle point the second two dimensional image corresponding with 3-D image each in image data base
It is matched, target two dimensional image is determined from the second two dimensional image.
Firstly, calculating each second according to the second angle point and the first angle point of the second two dimensional image each in image data base
The first similarity magnitude between two dimensional image and the first two dimensional image.Any one 3-D image is obtained from image data base
Corresponding second angle point of the second two dimensional image；First angle point and second angle point are matched, the angle of same position is extracted
Point pair；According to angle point to the first similarity magnitude for calculating second two dimensional image.
Then, second two dimensional image of the first similarity magnitude in the second preset threshold is chosen as target X-Y scheme
Picture.Obtain the corresponding visual angle weight in 3-D image of each second two dimensional image；According to the visual angle weight of the second two dimensional image and
Second similarity magnitude of first the second two dimensional image of similarity magnitude calculation of the second two dimensional image；Choose the second similarity measure
It is worth the second two dimensional image in the second preset threshold as target two dimensional image.
S330: the subtense angle between the first two dimensional image and target two dimensional image is calculated.
Perspective transformation matrix is established to the coordinate respectively in target two dimensional image and the first two dimensional image according to angle point；Root
Subtense angle is calculated according to perspective transformation matrix.
S340: subtense angle is less than 3-D image corresponding to the target two dimensional image of first threshold and is determined as search knot
Fruit.
S350: the visual angle in 3-D image according to subtense angle and target two dimensional image calculates the observation of the first two dimensional image
Visual angle shows target three-dimensional image in observation visual angle.
Above-mentioned 3-D image searching method, according to the first two dimensional image of object search easy to collect, by X-Y scheme
Readily identified angle point, carries out the matching of image as in, to search out 3-D image and as search result, realizes three-dimensional figure
The search of picture.Moreover, two dimensional image is easy to get in this method, and the angle point as feature is also easy to identify, substantially increases
The efficiency of search；The three-dimensional character for considering object search simultaneously improves the standard of 3-D image search also according to the judgement of subtense angle
True property.
In one embodiment, as shown in figure 4, Fig. 4 is the structural representation of 3-D image search system in one embodiment
Scheme, provides a kind of 3-D image search system, including angle point computing module 410, two dimensional image matching module in the present embodiment
420, subtense angle computing module 430 and acquiring three-dimensional images module 440, in which:
Angle point computing module 410 calculates the first of the first two dimensional image for obtaining the first two dimensional image of object search
Angle point.
Angle point computing module 410, can basis during being scanned for according to the first two dimensional image to object search
Corner Detection calculates the first angle point of the first two dimensional image.
Wherein, object search can be the three-dimensional object with three-dimensional structure, be also possible to three dimensional space coordinate
The visual 3-D image of information, search image obtain two dimensional image after capable of being collected on visual angle.Intuitively under overview, angle
Point has feature below: can be the angle point between profile；For Same Scene, even if visual angle changes, it is generally configured with
The feature of stability property；No matter the pixel of the near zone has larger change on gradient direction or on its gradient magnitude
Change.Angle point is feature that is intuitive in image and being easy to detect, is also convenient for carrying out suitable images match.
Two dimensional image matching module 420, for by the first angle point corresponding with 3-D image each in image data base the
Second angle point of two two dimensional images is matched, and target two dimensional image is determined from the second two dimensional image.
Two dimensional image matching module 420 is matched and is searched in image data base by the first angle point, is determined and the
The similar target two dimensional image of one two dimensional image.
Wherein, corresponding second two dimensional image of 3-D image is store in image data base, these second two dimensional images can
To be also possible to specially calculate to 3-D image from for showing two dimensional image of the 3-D image under certain visual angle
Two dimensional image under certain visual angle.Since 3-D image has three-dimensional three-dimensional character, entrained information is more, general single view
Two dimensional image under angle is merely able to locally show 3-D image, and the 3-D image generally in image data base can be corresponded to and be deposited
Contain the second two dimensional image under multiple visual angles.
Subtense angle computing module 430, for calculating the subtense angle between the first two dimensional image and target two dimensional image.
The characteristics of subtense angle computing module 430 can be by visual transform, according to the first angle point of the first two dimensional image and
Corresponding second angle point of target two dimensional image can calculate the subtense angle between the first two dimensional image and target two dimensional image.
Acquiring three-dimensional images module 440, for being less than subtense angle three corresponding to the target two dimensional image of first threshold
Dimension image is determined as search result.
Further judge that whether the accuracy of corners Matching, acquiring three-dimensional images module 440 detect subtense angle in error model
In enclosing, if subtense angle is in error range, i.e., subtense angle is less than first threshold, then by three-dimensional figure corresponding to target two dimensional image
As being determined as search result.
Above-mentioned 3-D image search system, according to the first two dimensional image of object search easy to collect, by X-Y scheme
Readily identified angle point, carries out the matching of image as in, to search out 3-D image and as search result, realizes three-dimensional figure
The search of picture.Moreover, two dimensional image is easy to get in this method, and the angle point as feature is also easy to identify, substantially increases
The efficiency of search；The three-dimensional character for considering object search simultaneously improves the standard of 3-D image search also according to the judgement of subtense angle
True property.
Specific about 3-D image search system limits the limit that may refer to above for 3-D image searching method
Fixed, details are not described herein.Modules in above-mentioned 3-D image search system can fully or partially through software, hardware and its
Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with
It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding
Operation.
As shown in figure 5, Fig. 5 is the schematic diagram of internal structure of computer equipment in one embodiment.The computer equipment packet
Include processor, non-volatile memory medium, memory and the network interface connected by system bus.Wherein, which sets
Standby non-volatile memory medium is stored with operating system, database and computer program, which is held by processor
When row, processor may make to realize a kind of 3-D image searching method.The processor of the computer equipment for provide calculate and
Control ability supports the operation of entire computer equipment.Computer program can be stored in the memory of the computer equipment, it should
When computer program is executed by processor, processor may make to execute a kind of 3-D image searching method.The computer equipment
Network interface is used for and terminal connection communication.It will be understood by those skilled in the art that structure shown in Fig. 5, only and originally
The block diagram of the relevant part-structure of application scheme, does not constitute the limit for the computer equipment being applied thereon to application scheme
Fixed, specific computer equipment may include perhaps combining certain components or tool than more or fewer components as shown in the figure
There is different component layouts.
In one embodiment it is proposed that a kind of computer equipment, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor are realized in any of the above-described embodiment when executing the computer program
The step of 3-D image searching method.
In one embodiment it is proposed that a kind of storage medium for being stored with computer-readable instruction, is stored thereon with meter
Calculation machine program, the step of 3-D image searching method in any of the above-described embodiment is realized when computer program is executed by processor.
It should be understood that although each step in the flow chart of attached drawing is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, can execute in the other order.Moreover, at least one in the flow chart of attached drawing
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, execution sequence, which is also not necessarily, successively to be carried out, but can be with other
At least part of the sub-step or stage of step or other steps executes in turn or alternately.
The above is only some embodiments of the invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of 3-D image searching method, which comprises the steps of:
The first two dimensional image of object search is obtained, the first angle point of first two dimensional image is calculated；
Second angle point of first angle point the second two dimensional image corresponding with 3-D image each in image data base is carried out
Matching determines target two dimensional image from second two dimensional image；
Calculate the subtense angle between first two dimensional image and the target two dimensional image；
Subtense angle is less than 3-D image corresponding to the target two dimensional image of first threshold and is determined as search result.
2. 3-D image searching method according to claim 1, which is characterized in that described to obtain the 1st of object search
The step of tieing up image, calculating the first angle point of first two dimensional image, comprising:
Two original images of described search object are obtained by picture pick-up device within a preset time interval；
Calculate separately the corresponding third angle point of the original image and fourth angle point；
Angle point screening is carried out to the third angle point according to the fourth angle point, obtains first angle point.
3. 3-D image searching method according to claim 1, which is characterized in that described by first angle point and image
The second angle point of corresponding second two dimensional image of each 3-D image is matched in database, from second two dimensional image
The step of determining target two dimensional image, comprising:
According to the second angle point and first angle point of the second two dimensional image each in described image database, each described the is calculated
The first similarity magnitude between two two dimensional images and first two dimensional image；
Second two dimensional image of the first similarity magnitude in the second preset threshold is chosen as the target two dimensional image.
4. 3-D image searching method according to claim 3, which is characterized in that described to calculate each second X-Y scheme
As the step of with the first similarity magnitude between first two dimensional image, comprising:
Corresponding second angle point of the second two dimensional image of any one 3-D image is obtained from described image database；
First angle point and second angle point are matched, the angle point pair of same position is extracted；
According to the angle point to the first similarity magnitude for calculating second two dimensional image.
5. 3-D image searching method according to claim 3, which is characterized in that described to choose first similarity measure
It is worth the step of the second two dimensional image in the second preset threshold is as the target two dimensional image, comprising:
Obtain the corresponding visual angle weight in 3-D image of each second two dimensional image；
According to the visual angle weight of second two dimensional image and the first similarity magnitude calculation of second two dimensional image
Second similarity magnitude of the second two dimensional image；
Second two dimensional image of the second similarity magnitude in the second preset threshold is chosen as the target two dimensional image.
6. three-dimensional image search method according to claim 4, which is characterized in that described to calculate first two dimensional image
The step of subtense angle between the target two dimensional image, comprising:
Perspective is established to the coordinate respectively in the target two dimensional image and first two dimensional image according to the angle point to become
Change matrix；
The subtense angle is calculated according to the perspective transformation matrix.
7. 3-D image searching method according to claim 1, which is characterized in that it is described by subtense angle less than the first threshold
After the step of 3-D image corresponding to the target two dimensional image of value is determined as search result, comprising:
According to the subtense angle and target two dimensional image visual angle in 3-D image, the sight of first two dimensional image is calculated
Examine visual angle；
The target three-dimensional image is shown in the observation visual angle.
8. a kind of 3-D image search system characterized by comprising
Angle point computing module calculates first jiao of first two dimensional image for obtaining the first two dimensional image of object search
Point；
Two dimensional image matching module is used for first angle point and 3-D image the corresponding 2nd 2 each in image data base
Second angle point of dimension image is matched, and target two dimensional image is determined from second two dimensional image；
Subtense angle computing module, for calculating the subtense angle between first two dimensional image and the target two dimensional image；
Acquiring three-dimensional images module, it is true for subtense angle to be less than to 3-D image corresponding to the target two dimensional image of first threshold
It is set to search result.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes any one of claims 1 to 7 institute when executing the computer program
The step of stating 3-D image searching method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of 3-D image searching method described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910656211.3A CN110472085B (en) | 2019-07-19 | 2019-07-19 | Three-dimensional image searching method, system, computer device and storage medium |
| PCT/CN2019/118153 WO2021012523A1 (en) | 2019-07-19 | 2019-11-13 | Three-dimensional image search method and system, computer device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910656211.3A CN110472085B (en) | 2019-07-19 | 2019-07-19 | Three-dimensional image searching method, system, computer device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110472085A true CN110472085A (en) | 2019-11-19 |
| CN110472085B CN110472085B (en) | 2023-06-20 |
Family
ID=68508326
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910656211.3A Active CN110472085B (en) | 2019-07-19 | 2019-07-19 | Three-dimensional image searching method, system, computer device and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110472085B (en) |
| WO (1) | WO2021012523A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793422A (en) * | 2021-08-13 | 2021-12-14 | 深圳安泰创新科技股份有限公司 | Display control method of three-dimensional model, electronic device and readable storage medium |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116274893B (en) * | 2023-02-24 | 2024-02-06 | 北京瓦特曼智能科技有限公司 | Method, processor, device and storage medium for detecting erosion amount of ladle lining |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110313989A1 (en) * | 2009-12-01 | 2011-12-22 | Rishab Aiyer Ghosh | System and method for customizing search results from user's perspective |
| CN104714986A (en) * | 2013-12-12 | 2015-06-17 | 三纬国际立体列印科技股份有限公司 | Three-dimensional picture searching method and three-dimensional picture searching system |
| EP3023909A1 (en) * | 2014-11-18 | 2016-05-25 | Thomson Licensing | Method and apparatus for ranking 2D candidate images |
| CN107481315A (en) * | 2017-06-29 | 2017-12-15 | 重庆邮电大学 | A kind of monocular vision three-dimensional environment method for reconstructing based on Harris SIFT BRIEF algorithms |
| CN109708649A (en) * | 2018-12-07 | 2019-05-03 | 中国空间技术研究院 | A kind of attitude determination method and system of remote sensing satellite |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004164503A (en) * | 2002-11-15 | 2004-06-10 | Olympus Corp | Three-dimensional model retrieval method, three-dimensional model retrieval device, three-dimensional model retrieval program and three-dimensional model retrieval system |
| CN103116904A (en) * | 2012-10-23 | 2013-05-22 | 北京航空航天大学深圳研究院 | Two-dimensional feature extraction system and two-dimensional feature extraction method of three-dimensional model |
| CN103177098B (en) * | 2013-03-19 | 2015-08-26 | 浙江大学 | A kind of method of the three-dimensional model search based on Freehandhand-drawing figure |
| CN106484692B (en) * | 2015-08-25 | 2019-11-29 | 北京师范大学 | A kind of method for searching three-dimension model |
| CN107247752B (en) * | 2017-05-27 | 2020-05-29 | 西安电子科技大学 | Image retrieval method based on corner descriptor |
-
2019
- 2019-07-19 CN CN201910656211.3A patent/CN110472085B/en active Active
- 2019-11-13 WO PCT/CN2019/118153 patent/WO2021012523A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110313989A1 (en) * | 2009-12-01 | 2011-12-22 | Rishab Aiyer Ghosh | System and method for customizing search results from user's perspective |
| CN104714986A (en) * | 2013-12-12 | 2015-06-17 | 三纬国际立体列印科技股份有限公司 | Three-dimensional picture searching method and three-dimensional picture searching system |
| EP3023909A1 (en) * | 2014-11-18 | 2016-05-25 | Thomson Licensing | Method and apparatus for ranking 2D candidate images |
| CN107481315A (en) * | 2017-06-29 | 2017-12-15 | 重庆邮电大学 | A kind of monocular vision three-dimensional environment method for reconstructing based on Harris SIFT BRIEF algorithms |
| CN109708649A (en) * | 2018-12-07 | 2019-05-03 | 中国空间技术研究院 | A kind of attitude determination method and system of remote sensing satellite |
Non-Patent Citations (1)
| Title |
|---|
| 赵鹏图;达飞鹏;: "基于局部特征的大视角图像匹配", 光学学报 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793422A (en) * | 2021-08-13 | 2021-12-14 | 深圳安泰创新科技股份有限公司 | Display control method of three-dimensional model, electronic device and readable storage medium |
| CN113793422B (en) * | 2021-08-13 | 2024-02-23 | 深圳安泰创新科技股份有限公司 | Display control method for three-dimensional model, electronic device and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021012523A1 (en) | 2021-01-28 |
| CN110472085B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7554575B2 (en) | Fast imaging system calibration | |
| JP6216508B2 (en) | Method for recognition and pose determination of 3D objects in 3D scenes | |
| US7027651B2 (en) | Geometric hashing method for model-based recognition of an object | |
| CN109544599B (en) | Three-dimensional point cloud registration method based on camera pose estimation | |
| JP2018524678A (en) | Business discovery from images | |
| US11965967B2 (en) | Apparatus and method for detecting intersection edges | |
| CN113362441B (en) | Three-dimensional reconstruction method, three-dimensional reconstruction device, computer equipment and storage medium | |
| CN110533654A (en) | The method for detecting abnormality and device of components | |
| WO2019006389A1 (en) | Dense optical flow processing in a computer vision system | |
| CN113393439A (en) | Forging defect detection method based on deep learning | |
| US9924865B2 (en) | Apparatus and method for estimating gaze from un-calibrated eye measurement points | |
| US20230394832A1 (en) | Method, system and computer readable media for object detection coverage estimation | |
| CN110472085A (en) | 3-D image searching method, system, computer equipment and storage medium | |
| CN102708589B (en) | Three-dimensional target multi-viewpoint view modeling method on basis of feature clustering | |
| CN113298871B (en) | Map generation method, positioning method, system thereof, and computer-readable storage medium | |
| JP2018195070A (en) | Information processing apparatus, information processing method, and program | |
| CN112197708B (en) | Measuring method and device, electronic device and storage medium | |
| CN105608423A (en) | Video matching method and device | |
| WO2019006405A1 (en) | Hierarchical data organization for dense optical flow | |
| CN113011359A (en) | Method for simultaneously detecting plane structure and generating plane description based on image and application | |
| CN116823816A (en) | Detection equipment and detection method based on security monitoring static memory | |
| Tabb et al. | Using cameras for precise measurement of two-dimensional plant features: CASS | |
| CN106651950B (en) | Single-camera pose estimation method based on quadratic curve perspective projection invariance | |
| CN109961503A (en) | A kind of image processing method and device, terminal and computer readable storage medium | |
| CN113160414A (en) | Automatic identification method and device for remaining amount of goods, electronic equipment and computer readable medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |