CN104112124A - Image identification based indoor positioning method and device - Google Patents
Image identification based indoor positioning method and device Download PDFInfo
- Publication number
- CN104112124A CN104112124A CN201410335949.7A CN201410335949A CN104112124A CN 104112124 A CN104112124 A CN 104112124A CN 201410335949 A CN201410335949 A CN 201410335949A CN 104112124 A CN104112124 A CN 104112124A
- Authority
- CN
- China
- Prior art keywords
- mark
- identification information
- picture
- mark picture
- proper vector
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses an image identification based indoor positioning method. The image identification based indoor positioning method includes acquiring a current position identifier picture shot by a user and preprocessing the identifier picture; matching the preprocessed identifier picture with identifier information of each identifier in an identifier information base and identifying the identifier picture; determining the position where the shot position identifier picture is located in a map system according to position information of the identification result in the current map system, and presenting the position where the shot position identifier picture is located in the map system. The invention further discloses an image identification based indoor positioning device. By means of the image identification based indoor positioning method and device, positioning of most indoor areas can be achieved, workload is small, the universality is high, and practical values are high.
Description
Technical field
The present invention relates to geographic positioning technology field, relate in particular to a kind of indoor orientation method and device based on image recognition.
Background technology
Along with the people are at the colouring day by day of living, people are increasing uses large-scale indoor public places to carry out comings and goings, for example: remove megastore, supermarket shopping, or hold a meeting in large-scale commercial hotel, and put on an exhibition, visit various themes museum etc. in various large-scale meeting-place.Due to the layout difference in various indoor places, space is larger in addition, sometimes wants to determine that own current residing position is not easy especially.
Be example taking megastore, want in market, to determine own residing position, traditional way is the market floor level schematic diagram of putting up in market by checking.In floor map, provide the shop distributed intelligence of this floor of market, sometimes also can directly demonstrate the residing position of this floor map with small red dot or little Red Star; In the case of the small red dot or little Red Star that do not mark current location, client can only compare according to retail shop's information and floor map around, thereby judges self current present position.
But traditional floor map also exists inconvenience part, such as, the floor level schematic diagram limited amount that can put up, can not meet the needs of client's inquiring position information whenever and wherever possible in market; Or floor level schematic diagram may exist stained situation, in the case of not to being undertaken by stained floor level schematic diagram replacing in time, the positional information accuracy that this figure can provide will reduce.In addition, once market retail shop changes to some extent, original floor level schematic diagram, by ineffective, need to be made new map again.
Traditional indoor orientation method, can not be well for client provides accurately and service easily due to the restriction of factors.But the location technology that current some are comparatively general, due to the restriction of actual conditions, is not well positioned to meet the demand of indoor positioning.For example, as the widest GPS location technology of range of application, due to signal propagation problem, cannot apply in indoor positioning; And the accuracy limitations of GPS location technology, causes realizing 10 meters with interior accurate location; For Wifi and iBeacon indoor positioning technology, although under experiment condition, Wifi and iBeacon indoor positioning have all reached good effect, but in actual applications, because the factor that has various environmental interference can affect positioning precision, and need to carry out in advance infrastructure laying, cause these two kinds of indoor positioning technology still at the experimental stage, remain in certain distance apart from practical application.
Therefore,, in current indoor positioning technology, inquiring position information is the current indoor topmost problem existing in technology that is decided to be whenever and wherever possible.
Summary of the invention
In view of this, the embodiment of the present invention expects to provide a kind of indoor orientation method and device based on image recognition, utilize the camera head in intelligent terminal and terminal just can realize the accurate location to indoor most areas, and have the advantages that workload is little, popularization is strong, practical value is high.
For achieving the above object, technical scheme of the present invention is achieved in that
The embodiment of the present invention provides a kind of indoor orientation method based on image recognition, and described method comprises:
Obtain the current location mark picture that user takes, and described mark picture is carried out to pre-service;
Pretreated mark picture is mated with the identification information of the each mark in identification information storehouse, described mark picture is identified;
Positional information according to recognition result in current map system, determines station location marker present position in map system of described shooting, and presents on map.
In such scheme, described mark picture is carried out pre-service and comprised: described mark picture is carried out to picture cutting and gray processing processing.
In such scheme, described method also comprises: described pretreated mark picture is carried out to yardstick invariant features conversion Sift proper vector and extract.
In such scheme, in described identification information storehouse, store described identification information in the mode of picture or Sift proper vector.
In such scheme, described pretreated mark picture is mated with the identification information of each mark in identification information storehouse, described mark picture is identified and comprised:
Calculating respectively the Sift proper vector of described mark picture counted with the mating of Sift proper vector of the identification information of the each mark in identification information storehouse;
According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number.
In such scheme, described identification information storehouse comprises the identification information under the multiple shooting angle of different identification;
Accordingly, described calculating respectively mating to count and comprising of the Sift proper vector of described mark picture and the Sift proper vector of the identification information of the each mark in identification information storehouse: calculate respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse Sift proper vector mate the summation of counting; The mark that described basis and described mark picture match point number are maximum determines that the recognition result of described mark picture comprises: according to counting mark that summation is the highest and determine the recognition result of described mark picture with described mark picture match point.
In such scheme, the summation of counting of mating of the described Sift proper vector of calculating respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse comprises:
Obtain the angle of the captured mark picture of user's input;
Determine according to the shooting angle of user's input the sequencing that described mark picture mates with the identification information of multiple shooting angle in identification information storehouse;
Calculate respectively in the Sift proper vector and identification information storehouse of described mark picture the summation that the coupling of the Sift proper vector of the multiple identification informations under the multiple shooting angle of each mark is counted according to sequencing.
The embodiment of the present invention also provides a kind of indoor positioning device based on image recognition, and described device comprises: image pretreatment unit, images match unit, position determination unit; Wherein,
Described image pretreatment unit, the current location mark picture of taking for obtaining user, and described mark picture is carried out to pre-service;
Described images match unit, for pretreated mark picture is mated with the identification information of each mark in identification information storehouse, identifies described mark picture;
Described position determination unit, for the positional information at current map system according to recognition result, determines station location marker present position in map system of described shooting, and presents on map.
In such scheme, described image pretreatment unit carries out pre-service to mark picture and comprises: described mark picture is carried out to picture cutting and gray processing processing.
In such scheme, described device also comprises proper vector extraction unit, extracts for described pretreated mark picture being carried out to yardstick invariant features conversion Sift proper vector.
In such scheme, described images match unit mates pretreated mark picture with the identification information of each mark in identification information storehouse, described mark picture is identified and comprised:
Described images match unit calculates respectively the Sift proper vector of described mark picture and counts with the mating of Sift proper vector of the identification information of the each mark in identification information storehouse;
According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number.
In such scheme, in the time that identification information storehouse comprises the identification information under the multiple shooting angle of different identification,
Described images match unit calculates respectively mating to count and comprising of the Sift proper vector of described mark picture and the Sift proper vector of the identification information of the each mark in identification information storehouse: calculate respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse Sift proper vector mate the summation of counting; The mark that described basis and described mark picture match point number are maximum determines that the recognition result of described mark picture comprises: according to counting mark that summation is the highest and determine the recognition result of described mark picture with described mark picture match point.
In such scheme, the summation of counting of mating that described images match unit calculates respectively the Sift proper vector of the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse comprises:
Obtain the angle of the captured mark picture of user's input;
Determine according to the shooting angle of user's input the sequencing that described mark picture mates with the identification information of multiple shooting angle in identification information storehouse;
Calculate respectively the summation of counting of mating of the Sift proper vector of described mark picture and the Sift proper vector of the multiple identification informations under the multiple shooting angle of each mark in identification information storehouse according to sequencing.
The indoor orientation method based on image recognition and device that the embodiment of the present invention provides, first obtain the current location mark picture that user takes, and described mark picture carried out to pre-service; Then described pretreated mark picture is mated with the identification information of the each mark in identification information storehouse, described mark picture is identified; Finally the positional information in current map system according to recognition result, determines station location marker present position in map system of described shooting, and presents on map.So, utilize identification information and the electronic chart that user terminal obtains to determine user self present position, and do not rely on floor maps on paper, thus can avoid cannot identification present position due to the stained client of causing of floor level schematic diagram situation; In the time that indoor arrangement changes, only need to upgrade the various position relationships that are identified in electronic chart, greatly reduce the labor capacity that substitutes floor level schematic diagram, have the advantages that workload is little, popularization is strong, practical value is high.
Brief description of the drawings
Fig. 1 is the indoor orientation method schematic flow sheet of the embodiment of the present invention one based on image recognition;
Fig. 2 is the identification information that the multiple shooting angle of the same mark of the embodiment of the present invention takes;
Fig. 3 is embodiment of the present invention identification information storehouse project organization schematic diagram;
Fig. 4 is the indoor orientation method schematic flow sheet of the embodiment of the present invention two based on image recognition;
Fig. 5 is the indoor positioning apparatus structure schematic diagram of the embodiment of the present invention based on image recognition.
Embodiment
In the embodiment of the present invention, first obtain the current location mark picture that user takes, and described mark picture is carried out to pre-service; Then described pretreated mark picture is mated with the identification information of the each mark in identification information storehouse, described mark picture is identified; Finally the positional information in current map system according to recognition result, determines station location marker present position in map system of described shooting, and presents on map.
Wherein, current map system refers to complete topomap place, the current place of user, electronization, as: when place, the current place of user is market, supermarket, hotel, museum, museum, library, current map system comprises respectively the complete topomap in market, supermarket, hotel, museum, museum, library and the layering topomap of each layer.
Here described, mark picture is carried out pre-service and comprised: described mark picture is carried out to picture cutting and gray processing processing; After mark picture is carried out to pre-service, described method also comprises: described pretreated mark picture is carried out to yardstick invariant features conversion (Sift, Scale-invariant feature transform) proper vector and extract; Correspondingly, in described identification information storehouse, store described identification information in the mode of picture or Sift proper vector.Here, described identification information is the pre-stored information relevant to mark picture recognition, can be the picture relevant to mark picture recognition or with identifying picture recognition the Sift proper vector of relevant picture.For example, for in the time that the described indoor orientation method based on image recognition is applied to the location in market, the mark picture that described user takes can be the logo of retail shop of user's shooting, described identification information can be the logo of the retail shop picture of taking under multiple shooting angle of pre-stored described retail shop, or these pictures is carried out to the Sift proper vector obtaining after Sift proper vector is extracted.
Concrete, described described pretreated mark picture is mated with the identification information of each mark in identification information storehouse, described mark picture is identified and comprised: the Sift proper vector of calculating respectively described mark picture was counted with the mating of Sift proper vector of the identification information of the each mark in identification information storehouse; According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number.
In the time that described identification information storehouse comprises the identification information under the multiple shooting angle of different identification, described calculating respectively mating to count and comprising of the Sift proper vector of described mark picture and the Sift proper vector of the identification information of the each mark in identification information storehouse: calculate respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse Sift proper vector mate the summation of counting; The mark that described basis and described mark picture match point number are maximum determines that the recognition result of described mark picture comprises: according to counting mark that summation is the highest and determine the recognition result of described mark picture with described mark picture match point.
Concrete, the summation of counting of mating of the described Sift proper vector of calculating respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse comprises: the angle of obtaining the captured mark picture of user's input; Determine according to the shooting angle of user's input the sequencing that described mark picture mates with the identification information of multiple shooting angle in identification information storehouse; According to sequencing, calculate respectively the summation of counting of mating of the Sift proper vector of described mark picture and the Sift proper vector of the multiple identification informations under the multiple shooting angle of each mark in identification information storehouse.
Below in conjunction with drawings and the specific embodiments, the technical scheme of the embodiment of the present invention is realized and being described in further detail.
Fig. 1 is the indoor orientation method schematic flow sheet of the embodiment of the present invention one based on image recognition, and as shown in Figure 1, the indoor orientation method of the embodiment of the present invention based on image recognition comprises the following steps:
Step 101: obtain the current location mark picture that user takes, and described mark picture is carried out to pre-service;
Wherein, described mark picture is carried out pre-service and comprised: described mark picture is carried out to picture cutting and gray processing processing.
In picture due to user's shooting, except when outside the mark of front position, also comprise all the other background parts.Background parts does not only provide the relevant information of current location mark, also can introduce extra interfere information, and recognition result is caused to adverse influence.In order to reduce as far as possible the interference of background, need to carry out cutting to the mark picture of taking, remove with current location and identify irrelevant part, thereby be more absorbed in the processing to position mark.
In addition, because the difference of station location marker is mainly shape facility but not on color, therefore, can carry out gray processing operation to picture, to avoid being worth little color characteristic processing, make subsequent treatment can more be absorbed in the processing of station location marker shape facility, contents processing is more reasonable, and efficiency is higher; And because the pictorial information amount of gray processing reduces, picture can make picture transmitting procedure more smooth.
After image is carried out to pre-service, described method also comprises: described pretreated mark picture is carried out to the extraction of Sift proper vector.
In the image matching technology based on feature, top priority is that extraction is suitable for the feature of explaining picture material and having better stability, and is showed.Sift proper vector has comprised many places characteristic point information in picture, the effectively content characteristic of presentation video.In addition, Sift feature has yardstick unchangeability, rotational invariance and illumination invariant, can successfully manage the impact of irradiating due to take pictures distance, photo angle and light in process is taken pictures in market, there is good scene adaptability, therefore the Sift feature of selective extraction image is carried out images match.
Step 102: described pretreated mark picture is mated with the identification information of the each mark in identification information storehouse, described mark picture is identified;
Concrete, calculate respectively the Sift proper vector of described mark picture and count with the mating of Sift proper vector of the identification information of the each mark in identification information storehouse; According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number.
The identification of mark picture is to determine according to the matching degree of the identification information in mark picture and identification information storehouse, after the Sift proper vector of the identification information in described mark picture and identification information storehouse generates, adopt the Euclidean distance of key point Sift proper vector as the similarity determination tolerance of key point in two width images.Concrete; label taking is known the Sift proper vector of certain key point of picture; and to find out itself and the Sift proper vector of the nearest the first two key point of identification information Euclidean distance; in these two key points; if the nearest distance of Sift proper vector is less than certain proportion threshold value except the following near distance, to think that this is a pair of match point.Coupling is counted more, the matching degree that described mark picture and described identification information are described is higher, more likely belong to a station location marker, thereby captured mark picture is mated with the identification information of all marks in identification information storehouse, think and determine that with the maximum mark of described mark picture match point number described mark picture is recognition result, thereby realize the identification of clapping identification information.
In actual applications, user takes mark picture can not all be positioned at same angle, and this just requires in the time of design identification information storehouse, and same mark need to comprise the identification information that multiple shooting angle takes.As shown in Figure 2, in identification information storehouse, can comprise the identification information under same mark multiple angles: front elevation 2-a, the station location marker 30 that comprises station location marker spent 45 degree right view 2-d and the 45 degree left view 2-e of left view 2-b and 30 degree right view 2-c and station location marker.Fig. 3 is embodiment of the present invention identification information storehouse project organization schematic diagram.
In the time that described identification information storehouse comprises the identification information under the multiple shooting angle of different identification, described calculating respectively mating to count and comprising of the Sift proper vector of described mark picture and the Sift proper vector of the identification information of the each mark in identification information storehouse: calculate respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse Sift proper vector mate the summation of counting; The mark that described basis and described mark picture match point number are maximum determines that the recognition result of described mark picture comprises: according to counting mark that summation is the highest and determine the recognition result of described mark picture with described mark picture match point.
Taking Fig. 2 as example, carrying out identification information when identification, five identification informations couplings that the mark picture that statistics is taken is respectively corresponding with each mark summation of counting, gets and mates mark that the summation of counting is the highest as the recognition result that identifies picture.
In the embodiment of the present invention, in described identification information storehouse, can store described identification information in the mode of picture or Sift proper vector.
Because the Sift proper vector extraction of picture can exist regular hour expense, if directly mark picture is mated with an identification information picture in identification information storehouse, the mark picture of removing taking carries out Sift feature extraction, also needs the identification information picture in identification information storehouse to carry out the extraction of Sift proper vector.In the time that in identification information storehouse, template number is more, the Sift proper vector of extracting identification information picture is consuming time longer, does not allow susceptible to user acceptance.
The identification information of considering the each mark in identification information storehouse can not change, its corresponding Sift proper vector also can not change, therefore storage object in identification information storehouse, can be by original identification information picture, change the corresponding proper vector of identification information picture into, in identifying, only need to extract the mark picture Sift proper vector of taking like this, greatly shortened time overhead.
Consider the identification information in identification information storehouse is mated one by one and still can cause more time overhead, in the embodiment of the present invention, calculate respectively the mating the summation of counting and can adopt Pruning strategy to process of Sift proper vector of the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse, concrete:
First obtain the angle of the captured mark picture of user's input; Then determine according to the shooting angle of user's input the sequencing that described mark picture mates with the identification information of multiple shooting angle in identification information storehouse; Finally calculate respectively in the Sift proper vector and identification information storehouse of described mark picture the summation that the coupling of the Sift proper vector of the multiple identification informations under the multiple shooting angle of each mark is counted according to sequencing.
Still taking Fig. 2 as example, before coupling, complete following processing:
A. first obtain the angle of the shooting mark of user's input;
Here, angle can be divided into front elevation, left view, right view etc., and the embodiment of the present invention is taking input front elevation as example;
B. the mark picture of shooting is mated with the front elevation identification information of all marks in identification information Kuku, and count according to coupling from how to few order, each mark is sorted, the mark of front n% is denoted as to set A; Here, the size of n can determine according to actual conditions, for example, can select front 1/2 mark to be designated as set A, and n value is 50 so;
C. the mark picture of shooting is mated with the left view identification information of all marks in set A, and count according to coupling from many to few order, in pair set A, mark sorts, and the mark of front m% is denoted as to set B; Here, the size of m can determine according to actual conditions, and the value of m can be identical with n, also can be different from n; For example, can select equally front 1/2 mark to be designated as set B, m value is also 50 so;
D. the mark picture of shooting is mated with the right view identification information of all marks in set B, add up each coupling and count.
E. the mark picture that statistics is taken and the template matches of all marks summation of counting, choose match point summation the highest for being designated recognition result;
Here the match point number scale that has neither part nor lot in the identification information of coupling, is 0.
Actual motion shows, has adopted after Pruning strategy, and the recognition time expense of mark picture reduces greatly, can meet the needs of recognition accuracy, also can meet the requirement of real-time.
Step 103: the positional information according to recognition result in current map system, determine station location marker present position in map system of described shooting, and present on map.
After identifying the mark picture of shooting, rely on existing indoor electronic map system, inquire about the coordinate position of this mark, and present on indoor electronic map.
Here, current location is presented on map and do not limit concrete implementation, for example, can in map system, directly demonstrate the residing position of this station location marker with small red dot or little Red Star, also can provide concrete location expression etc.
The indoor orientation method based on image recognition described in the embodiment of the present invention, is applicable to the indoor place altogether of the large-scale public affairs such as market, supermarket, hotel, museum, museum, library; Described station location marker can be the relatively-stationary station location marker in position in above-mentioned large-scale place.For example, in the time that the embodiment of the present invention is applicable to market, the mark picture of described shooting can be the logo of retail shop in market; In the time that the embodiment of the present invention is applied to hotel, the mark picture of described shooting can be the number in room; In the time that the embodiment of the present invention is applied to museum, the mark picture of described shooting can be exhibition booth number or showpiece etc.The locator meams that the embodiment of the present invention provides, can be applied to outdoor positioning equally, for example: using the various theme architecture in theme park as identification information, user can, by taking building at the moment, determine own present position.
Indoor orientation method based on image recognition described in the use embodiment of the present invention, only need to set up identification information storehouse and be identified at position relationship corresponding in current map system, and the shooting by user to current scene, can realize positioning function.In the time of the layout change in described place, corresponding position relationship in only need to upgrading identification information in identification information storehouse and being identified at map system, can realize performance renewal.For example, in the time having new businessman to enter in market, only the logo information updating of new businessman need to be arrived to identification information storehouse, and be identified at position relationship corresponding in map described in adding in map system.
Due to the repeatability of some Code in Hazardous Special Locations tick lableses, in setting up identification information storehouse, do not need to set up respectively corresponding identification information storehouse for all places.Taking most representative market as example, the logo between megastore has very large repeatability, and a lot of brand logo exist in each market, therefore do not need to repeat to set up identification information storehouse and carry out repeated storage.
Below in conjunction with a concrete scene, the realization of the indoor orientation method that the present invention is based on image recognition is described in further detail.Fig. 4 is the indoor orientation method schematic flow sheet of the embodiment of the present invention two based on image recognition, and the embodiment of the present invention is example taking market, comprises the following steps:
Step 400: in model market logo storehouse corresponding to all logo of retail shop;
In actual applications, need to provide its corresponding logo storehouse for every market; Meanwhile, for certain first floor in a certain market, its logo comprising is a subset in logo storehouse, in the time mating, only need mate with this logo subset.As shown in Table 1 and Table 2, wherein, table 1 is the corresponding relation between embodiment of the present invention market and logo storehouse in the foundation in logo storehouse; The brand message that table 2 comprises for the each floor in every market of the embodiment of the present invention.
| Location | Mall?ID |
| Proper vector storehouse 1 | 1 |
| Proper vector storehouse 2 | 2 |
| Proper vector storehouse 3 | 3 |
| Generic features vector storehouse | 4 |
| Generic features vector storehouse | 5 |
Table 1
| Mall?ID | Floor | Brand |
| 1 | Floor1 | AVENIR |
| 1 | Floor1 | CASIO |
| 1 | Floor1 | DIVUM |
| 1 | Floor1 | Elizabeth?Arden |
| 1 | Floor1 | KOSS |
| 1 | Floor1 | MIDO |
| 1 | Floor1 | sasa |
| 1 | Floor1 | TISSOT |
| 1 | Floor1 | ZA |
| 1 | Floor1 | Peace favoured beauty |
| 1 | Floor1 | Herborist |
| 1 | Floor1 | Sweet members of a family's gold |
| 1 | Floor1 | Person of outstanding talent is refined |
| 1 | Floor1 | With beautiful edge |
| 1 | Floor1 | Famous-brand clock center, the prosperous lucky world |
| 1 | Floor1 | Hua Teng |
| 1 | Floor1 | Hui Wenxiang |
| 1 | Floor1 | Good skill |
| 1 | Floor1 | Golden Elephant jewelry |
Table 2
In table 1, for every market all provides its corresponding Logo storehouse, in Location attribute, storing content is the storage information in logo storehouse.For the market that is 4 and 5 for No. ID, unpromising its set up own all logo proper vector storehouses, but corresponding with generic features vector storehouse.This is because the logo between numerous megastore, market in a city has very large repeatability, and a lot of brand logo exist in each market, thereby also do not need to repeat to set up too much logo storehouse deduplication storage.
In table 2, by market ID and floor of living in, can inquire the logo set C that this floor has.In the process of logo identification, corresponding market numbering and floor of living in only need be provided, can obtain the set C in the logo storehouse corresponding with the described floor in described market, and mate.
Step 401: the logo of the retail shop picture that obtains the current present position of user's shooting;
Step 402: the described mark logo of retail shop picture is carried out to pre-service;
Wherein, describedly logo picture is carried out to pre-service comprise: described logo picture is carried out to picture cutting and gray processing processing.
Step 403: described logo picture is carried out to the extraction of Sift proper vector;
Step 404: described logo picture is mated with the information of the each logo in logo storehouse, described logo picture is identified;
In described logo storehouse, can store described logo information in the mode of picture or Sift proper vector.
Concrete, calculate respectively the Sift proper vector of described logo picture and count with the mating of Sift proper vector of the each logo in logo storehouse; According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number.
In the time that described logo storehouse comprises the logo information under the multiple shooting angle of different retail shops, described calculating respectively mating to count and comprising of the Sift proper vector of described logo picture and the Sift proper vector of the logo information of the each retail shop in logo storehouse: calculate respectively the multiple logo information under the Sift proper vector of described logo picture and the multiple shooting angle of each retail shop in logo storehouse Sift proper vector mate the summation of counting; Described basis is mated maximum retail shop of counting and is determined that the recognition result of described logo picture comprises with described logo picture: according to mating the highest retail shop of summation of counting and determine the recognition result of described logo picture with described logo picture.
Here, can adopt Pruning strategy to process, concrete:
First the angle of logo picture is taken in input; Then determine according to the shooting angle of input the sequencing that described logo picture mates with the logo information of multiple shooting angle in logo storehouse; Finally according to sequencing, calculate respectively the Sift proper vector of described logo picture and the multiple logo information under many kinds of shooting angle of each logo of retail shop in logo storehouse Sift proper vector mate the summation of counting.
Still taking Fig. 2 as example, calculate respectively the Sift proper vector of described logo picture and comprise with the summation of counting of mating of the Sift proper vector of the multiple logo information under many kinds of shooting angle of each logo of retail shop in logo storehouse:
A. obtain the angle of the shooting mark of user's input;
Here, angle can be divided into front elevation, left view, right view etc., and the embodiment of the present invention is taking input front elevation as example;
B. the logo picture of shooting is mated with the front elevation of all logo of retail shop in logo storehouse, and count according to coupling from how to few order, each logo of retail shop is sorted, the logo of retail shop of front n% is denoted as to set A; Here, the size of n can be determined according to actual conditions, for example, can select front 1/2 the logo of retail shop to be designated as set A, and n value is 50 so;
C. the logo picture of shooting is mated with the left view of all logo of retail shop in set A, and count according to coupling from many to few order, in pair set A, the logo of retail shop sorts, and the logo of retail shop of front m% is denoted as to set B; Here, the size of m can determine according to actual conditions, and the value of m can be identical with n, also can be different from n; For example, can select equally front 1/2 the logo of retail shop to be designated as set B, m value is also 50 so;
D. the logo picture of shooting is mated with the right view of all logo of retail shop in set B, add up each coupling and count.
The logo information of taking under the logo picture that e. statistics is taken and the multiple shooting angle of each logo of retail shop mate the summation of counting, choose retail shop that match point summation is the highest as recognition result;
Here the match point number scale that has neither part nor lot in the logo information of coupling, is 0.
Step 405: the positional information according to recognition result in current map system, determine the logo of retail shop present position in market of described shooting, and present on map.
After identifying the logo picture of shooting, rely on existing market indoor electronic map system, inquire about the coordinate position of this logo picture, and present on the indoor electronic map in market.
Here, current location is presented on map and do not limit concrete implementation, for example, can in map system, directly demonstrate the residing position of this station location marker with small red dot or little Red Star, also can provide concrete location expression etc.
The embodiment of the present invention also provides a kind of indoor positioning device based on image recognition, and as shown in Figure 5, described device comprises: image pretreatment unit 51, images match unit 52, position determination unit 53; Wherein,
Described image pretreatment unit 51, the current location mark picture of taking for obtaining user, and described mark picture is carried out to pre-service;
Described image pretreatment unit 51 carries out pre-service to mark picture and comprises: described mark picture is carried out to picture cutting and gray processing processing.
In picture due to user's shooting, except when outside the mark of front position, also comprise all the other background parts.Background parts does not only provide the relevant information of current location mark, also can introduce extra interfere information, and recognition result is caused to adverse influence.In order to reduce as far as possible the interference of background, need to carry out cutting to the mark picture of taking, remove with current location and identify irrelevant part, thereby be more absorbed in the processing to position mark.
In addition, because the difference of station location marker is mainly shape facility but not on color, therefore, can carry out gray processing operation to picture, to avoid being worth little color characteristic processing, make subsequent treatment can more be absorbed in the processing of station location marker shape facility, contents processing is more reasonable, and efficiency is higher; And because the pictorial information amount of gray processing reduces, picture can make picture transmitting procedure more smooth.
Described device also comprises proper vector extraction unit 54, extracts for described pretreated mark picture being carried out to yardstick invariant features conversion Sift proper vector.
In the image matching technology based on feature, top priority is that extraction is suitable for the feature of explaining picture material and having better stability, and is showed.Sift proper vector has comprised many places characteristic point information in picture, the effectively content characteristic of presentation video.In addition, Sift feature has yardstick unchangeability, rotational invariance and illumination invariant, can successfully manage the impact of irradiating due to take pictures distance, photo angle and light in process is taken pictures in market, there is good scene adaptability, therefore the Sift feature of selective extraction image is carried out images match.
Described device also comprises identification information storehouse 55 and map system 56, wherein, in described identification information storehouse 55, can store described identification information in the mode of picture or Sift proper vector.
Described images match unit 52, for described pretreated mark picture is mated with the identification information of each mark in identification information storehouse, identifies described mark picture;
Concrete, described images match unit 52 mates pretreated mark picture with the identification information of each mark in identification information storehouse, described mark picture is identified and comprised: described images match unit 52 calculates respectively the Sift proper vector of described mark picture and counts with the mating of Sift proper vector of the identification information of the each mark in identification information storehouse; According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number
The identification of mark picture is to determine according to the matching degree of the identification information in mark picture and identification information storehouse, after the Sift proper vector of the identification information in described mark picture and identification information storehouse generates, adopt the Euclidean distance of key point Sift proper vector as the similarity determination tolerance of key point in two width images.Concrete; label taking is known the Sift proper vector of certain key point of picture; and to find out itself and the Sift proper vector of the nearest the first two key point of identification information Euclidean distance; in these two key points; if the nearest distance of Sift proper vector is less than certain proportion threshold value except the following near distance, to think that this is a pair of match point.Coupling is counted more, the matching degree that described mark picture and described identification information are described is higher, more likely belong to a station location marker, thereby captured mark picture is mated with the identification information of all marks in identification information storehouse, think and determine that with the maximum mark of described mark picture match point number described mark picture is recognition result, thereby realize the identification of clapping identification information.
In actual applications, user takes mark picture can not all be positioned at same angle, and this just requires, in the time of design identification information storehouse, need to comprise the identification information under the multiple shooting angle of different identification.In the time that identification information storehouse comprises the identification information under the multiple shooting angle of different identification, described images match unit 52 calculates respectively mating to count and comprising of the Sift proper vector of described mark picture and the Sift proper vector of the identification information of the each mark in identification information storehouse: calculate respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse Sift proper vector mate the summation of counting; The mark that described basis and described mark picture match point number are maximum determines that the recognition result of described mark picture comprises: according to counting mark that summation is the highest and determine the recognition result of described mark picture with described mark picture match point.
Concrete, the summation of counting of mating that described images match unit 52 calculates respectively the Sift proper vector of the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse comprises:
Obtain the angle of the captured mark picture of user's input;
Here, angle can be divided into front elevation, left view, right view etc., and the embodiment of the present invention is taking input front elevation as example;
Determine according to the shooting angle of user's input the sequencing that described mark picture mates with the identification information of multiple shooting angle in identification information storehouse;
Calculate respectively in the Sift proper vector and identification information storehouse of described mark picture the summation that the coupling of the Sift proper vector of the multiple identification informations under the multiple shooting angle of each mark is counted according to sequencing.
Described position determination unit 53, for the positional information at current map system according to recognition result, determines station location marker present position in map system of described shooting, and presents on map.
After described mark recognition unit 52 identifies the mark picture of shooting, described position determination unit 53 is relied on the indoor map system in existing market, inquires about the coordinate position of this retail shop, and presents on indoor map.
Here, described position determination unit 53 presents current location and does not limit concrete implementation on map, for example, can in map system, directly demonstrate the residing position of this station location marker with small red dot or little Red Star, also can provide concrete location expression etc.
The practical function of the each processing unit in the indoor positioning device based on image recognition shown in Fig. 5, can understand with reference to the associated description of the aforementioned indoor orientation method based on image recognition.The function that it will be appreciated by those skilled in the art that each processing unit in the indoor positioning device based on image recognition shown in Fig. 5 can realize by the program running on various types of processors; Described storage unit also can be realized by various storeies or storage medium.For example a kind of typical implementation method can be: intelligent terminal is responsible for taking the pre-service of mark picture, i.e. cutting and gray processing process; Gray scale picture transfers to background server and processes.Background server can be built with Apache+MySQL+PHP, and MySQL database is wherein used for storing Location table and mall table.
In embodiment provided by the present invention, should be understood that disclosed method and apparatus can be realized in other way.Device embodiment described above is only schematic, for example, the division of described unit, be only that a kind of logic function is divided, when actual realization, can there is other dividing mode, as: multiple unit or assembly can be in conjunction with, maybe can be integrated into another system, or some features can ignore, or do not carry out.In addition, the communication connection each other of shown or discussed each ingredient can be by some interfaces, and the indirect coupling of equipment or unit or communication connection can be wired, wireless or other forms of.
The above-mentioned unit as separating component explanation can or can not be also physically to separate, and the parts that show as unit can be or can not be also physical locations, can be positioned at a place, also can be distributed in multiple network element; Can select according to the actual needs part or all of unit wherein to realize the object of the present embodiment scheme.
In addition, the each functional unit in various embodiments of the present invention can all be integrated in a processing unit, can be also that each unit is distinguished separately as a unit, also can be integrated in a unit two or more unit; Above-mentioned integrated unit both can adopt the form of hardware to realize, and the form that also can adopt hardware to add SFU software functional unit realizes.
One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can complete by the relevant hardware of programmed instruction, aforesaid program can be stored in computer read/write memory medium, this program, in the time carrying out, is carried out the step that comprises said method embodiment; And aforesaid storage medium comprises: various media that can be program code stored such as movable storage device, ROM (read-only memory) (ROM, Read-Only Memory), magnetic disc or CDs.
Or, if the above-mentioned integrated unit of the embodiment of the present invention is realized and during as production marketing independently or use, also can be stored in a computer read/write memory medium using the form of SFU software functional unit.Based on such understanding, the part that the technical scheme of the embodiment of the present invention contributes to prior art in essence in other words can embody with the form of software product, this computer software product is stored in a storage medium, comprises that some instructions are in order to make a computer equipment (can be personal computer, server or the network equipment etc.) carry out all or part of of method described in the present invention each embodiment.And aforesaid storage medium comprises: various media that can be program code stored such as movable storage device, ROM, magnetic disc or CDs.
The present invention be the indoor orientation method based on image recognition recorded in example and device only taking above-described embodiment as example, but be not limited only to this, as long as the method and apparatus that relates to this indoor positioning is all in protection scope of the present invention.
The above, be only preferred embodiment of the present invention, is not intended to limit protection scope of the present invention.
Claims (13)
1. the indoor orientation method based on image recognition, is characterized in that, described method comprises:
Obtain the current location mark picture that user takes, and described mark picture is carried out to pre-service;
Pretreated mark picture is mated with the identification information of the each mark in identification information storehouse, described mark picture is identified;
Positional information according to recognition result in current map system, determines station location marker present position in map system of described shooting, and presents on map.
2. method according to claim 1, is characterized in that, describedly mark picture is carried out to pre-service comprises: described mark picture is carried out to picture cutting and gray processing processing.
3. method according to claim 1, is characterized in that, described method also comprises: described pretreated mark picture is carried out to yardstick invariant features conversion Sift proper vector and extract.
4. method according to claim 1, is characterized in that, in described identification information storehouse, stores described identification information in the mode of picture or Sift proper vector.
5. method according to claim 1, is characterized in that, described pretreated mark picture is mated with the identification information of each mark in identification information storehouse, and described mark picture is identified and comprised:
Calculating respectively the Sift proper vector of described mark picture counted with the mating of Sift proper vector of the identification information of the each mark in identification information storehouse;
According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number.
6. method according to claim 4, is characterized in that, described identification information storehouse comprises the identification information under the multiple shooting angle of different identification;
Accordingly, described calculating respectively mating to count and comprising of the Sift proper vector of described mark picture and the Sift proper vector of the identification information of the each mark in identification information storehouse: calculate respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse Sift proper vector mate the summation of counting;
The mark that described basis and described mark picture match point number are maximum determines that the recognition result of described mark picture comprises: according to counting mark that summation is the highest and determine the recognition result of described mark picture with described mark picture match point.
7. method according to claim 6, it is characterized in that, the summation of counting of mating of the described Sift proper vector of calculating respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse comprises:
Obtain the angle of the captured mark picture of user's input;
Determine according to the shooting angle of user's input the sequencing that described mark picture mates with the identification information of multiple shooting angle in identification information storehouse;
Calculate respectively in the Sift proper vector and identification information storehouse of described mark picture the summation that the coupling of the Sift proper vector of the multiple identification informations under the multiple shooting angle of each mark is counted according to sequencing.
8. the indoor positioning device based on image recognition, is characterized in that, described device comprises: image pretreatment unit, images match unit, position determination unit; Wherein,
Described image pretreatment unit, the current location mark picture of taking for obtaining user, and described mark picture is carried out to pre-service;
Described images match unit, for pretreated mark picture is mated with the identification information of each mark in identification information storehouse, identifies described mark picture;
Described position determination unit, for the positional information at current map system according to recognition result, determines station location marker present position in map system of described shooting, and presents on map.
9. install according to claim 8, it is characterized in that, described image pretreatment unit carries out pre-service to mark picture and comprises: described mark picture is carried out to picture cutting and gray processing processing.
10. install according to claim 8, it is characterized in that, described device also comprises proper vector extraction unit, extracts for described pretreated mark picture being carried out to yardstick invariant features conversion Sift proper vector.
11. install according to claim 8, it is characterized in that, described images match unit mates pretreated mark picture with the identification information of each mark in identification information storehouse, described mark picture is identified and comprised:
Described images match unit calculates respectively the Sift proper vector of described mark picture and counts with the mating of Sift proper vector of the identification information of the each mark in identification information storehouse;
According to determining the recognition result of described mark picture with the maximum mark of described mark picture match point number.
12. according to device described in claim 11, it is characterized in that, and in the time that identification information storehouse comprises the identification information under the multiple shooting angle of different identification,
Described images match unit calculates respectively mating to count and comprising of the Sift proper vector of described mark picture and the Sift proper vector of the identification information of the each mark in identification information storehouse: calculate respectively the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse Sift proper vector mate the summation of counting;
The mark that described basis and described mark picture match point number are maximum determines that the recognition result of described mark picture comprises: according to counting mark that summation is the highest and determine the recognition result of described mark picture with described mark picture match point.
13. according to device described in claim 12, it is characterized in that, the summation of counting of mating that described images match unit calculates respectively the Sift proper vector of the multiple identification informations under the Sift proper vector of described mark picture and the multiple shooting angle of each mark in identification information storehouse comprises:
Obtain the angle of the captured mark picture of user's input;
Determine according to the shooting angle of user's input the sequencing that described mark picture mates with the identification information of multiple shooting angle in identification information storehouse;
Calculate respectively the summation of counting of mating of the Sift proper vector of described mark picture and the Sift proper vector of the multiple identification informations under the multiple shooting angle of each mark in identification information storehouse according to sequencing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410335949.7A CN104112124A (en) | 2014-07-15 | 2014-07-15 | Image identification based indoor positioning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410335949.7A CN104112124A (en) | 2014-07-15 | 2014-07-15 | Image identification based indoor positioning method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104112124A true CN104112124A (en) | 2014-10-22 |
Family
ID=51708909
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410335949.7A Pending CN104112124A (en) | 2014-07-15 | 2014-07-15 | Image identification based indoor positioning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104112124A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184881A (en) * | 2015-08-28 | 2015-12-23 | 宇龙计算机通信科技(深圳)有限公司 | Method, apparatus, server and system for identifying user identity |
| CN105517679A (en) * | 2015-03-25 | 2016-04-20 | 北京旷视科技有限公司 | User location determination |
| CN106153047A (en) * | 2016-08-15 | 2016-11-23 | 广东欧珀移动通信有限公司 | A kind of indoor orientation method, device and terminal |
| CN106650202A (en) * | 2016-09-18 | 2017-05-10 | 中国科学院计算技术研究所 | Date-driven indoor area layout prediction method and system |
| CN106683195A (en) * | 2016-12-30 | 2017-05-17 | 上海网罗电子科技有限公司 | AR scene rendering method based on indoor location |
| CN107144857A (en) * | 2017-05-17 | 2017-09-08 | 深圳市伊特利网络科技有限公司 | Assisted location method and system |
| CN107341829A (en) * | 2017-06-27 | 2017-11-10 | 歌尔科技有限公司 | The localization method and device of virtual reality interactive component |
| CN107566975A (en) * | 2017-09-05 | 2018-01-09 | 合肥工业大学 | A kind of location aware method based on Solr correlations |
| CN107957859A (en) * | 2017-12-12 | 2018-04-24 | 歌尔科技有限公司 | Image presentation method and virtual reality device for virtual reality child teaching |
| CN108170822A (en) * | 2018-01-04 | 2018-06-15 | 维沃移动通信有限公司 | The sorting technique and mobile terminal of a kind of photo |
| CN108734734A (en) * | 2018-05-18 | 2018-11-02 | 中国科学院光电研究院 | Indoor orientation method and system |
| CN109982239A (en) * | 2019-03-07 | 2019-07-05 | 福建工程学院 | Store floor positioning system and method based on machine vision |
| CN110470295A (en) * | 2018-05-09 | 2019-11-19 | 北京智慧图科技有限责任公司 | A kind of indoor walking navigation and method based on AR positioning |
| CN111429385A (en) * | 2020-06-10 | 2020-07-17 | 北京云迹科技有限公司 | Map generation method, device and equipment |
| CN111723232A (en) * | 2020-07-06 | 2020-09-29 | 许广明 | Method and device for positioning and identifying building through Internet of things |
| CN114485605A (en) * | 2020-10-23 | 2022-05-13 | 丰田自动车株式会社 | Position specifying method and position specifying system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655369A (en) * | 2008-08-22 | 2010-02-24 | 环达电脑(上海)有限公司 | System and method of realizing positioning navigation by using image recognition technology |
| US20110090221A1 (en) * | 2009-10-20 | 2011-04-21 | Robert Bosch Gmbh | 3d navigation methods using nonphotorealistic (npr) 3d maps |
| CN102629329A (en) * | 2012-02-28 | 2012-08-08 | 北京工业大学 | Personnel indoor positioning method based on adaptive SIFI (scale invariant feature transform) algorithm |
| CN103198491A (en) * | 2013-01-31 | 2013-07-10 | 北京工业大学 | Indoor visual positioning method |
| CN103424113A (en) * | 2013-08-01 | 2013-12-04 | 毛蔚青 | Indoor positioning and navigating method of mobile terminal based on image recognition technology |
-
2014
- 2014-07-15 CN CN201410335949.7A patent/CN104112124A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655369A (en) * | 2008-08-22 | 2010-02-24 | 环达电脑(上海)有限公司 | System and method of realizing positioning navigation by using image recognition technology |
| US20110090221A1 (en) * | 2009-10-20 | 2011-04-21 | Robert Bosch Gmbh | 3d navigation methods using nonphotorealistic (npr) 3d maps |
| CN102629329A (en) * | 2012-02-28 | 2012-08-08 | 北京工业大学 | Personnel indoor positioning method based on adaptive SIFI (scale invariant feature transform) algorithm |
| CN103198491A (en) * | 2013-01-31 | 2013-07-10 | 北京工业大学 | Indoor visual positioning method |
| CN103424113A (en) * | 2013-08-01 | 2013-12-04 | 毛蔚青 | Indoor positioning and navigating method of mobile terminal based on image recognition technology |
Non-Patent Citations (1)
| Title |
|---|
| 代小红: "《基于机器视觉的数字图像处理与识别研究》", 31 March 2012 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105517679A (en) * | 2015-03-25 | 2016-04-20 | 北京旷视科技有限公司 | User location determination |
| US10657669B2 (en) | 2015-03-25 | 2020-05-19 | Beijing Kuangshi Technology Co., Ltd. | Determination of a geographical location of a user |
| CN105184881A (en) * | 2015-08-28 | 2015-12-23 | 宇龙计算机通信科技(深圳)有限公司 | Method, apparatus, server and system for identifying user identity |
| CN106153047A (en) * | 2016-08-15 | 2016-11-23 | 广东欧珀移动通信有限公司 | A kind of indoor orientation method, device and terminal |
| CN106650202A (en) * | 2016-09-18 | 2017-05-10 | 中国科学院计算技术研究所 | Date-driven indoor area layout prediction method and system |
| CN106650202B (en) * | 2016-09-18 | 2019-03-12 | 中国科学院计算技术研究所 | A kind of the room area layout prediction technique and system of data-driven |
| CN106683195A (en) * | 2016-12-30 | 2017-05-17 | 上海网罗电子科技有限公司 | AR scene rendering method based on indoor location |
| CN107144857A (en) * | 2017-05-17 | 2017-09-08 | 深圳市伊特利网络科技有限公司 | Assisted location method and system |
| CN107341829A (en) * | 2017-06-27 | 2017-11-10 | 歌尔科技有限公司 | The localization method and device of virtual reality interactive component |
| CN107566975A (en) * | 2017-09-05 | 2018-01-09 | 合肥工业大学 | A kind of location aware method based on Solr correlations |
| CN107957859A (en) * | 2017-12-12 | 2018-04-24 | 歌尔科技有限公司 | Image presentation method and virtual reality device for virtual reality child teaching |
| CN108170822A (en) * | 2018-01-04 | 2018-06-15 | 维沃移动通信有限公司 | The sorting technique and mobile terminal of a kind of photo |
| CN110470295B (en) * | 2018-05-09 | 2022-09-30 | 北京智慧图科技有限责任公司 | Indoor walking navigation system and method based on AR positioning |
| CN110470295A (en) * | 2018-05-09 | 2019-11-19 | 北京智慧图科技有限责任公司 | A kind of indoor walking navigation and method based on AR positioning |
| CN108734734A (en) * | 2018-05-18 | 2018-11-02 | 中国科学院光电研究院 | Indoor orientation method and system |
| CN109982239A (en) * | 2019-03-07 | 2019-07-05 | 福建工程学院 | Store floor positioning system and method based on machine vision |
| CN111429385A (en) * | 2020-06-10 | 2020-07-17 | 北京云迹科技有限公司 | Map generation method, device and equipment |
| CN111429385B (en) * | 2020-06-10 | 2021-01-08 | 北京云迹科技有限公司 | Map generation method, device and equipment |
| CN111723232A (en) * | 2020-07-06 | 2020-09-29 | 许广明 | Method and device for positioning and identifying building through Internet of things |
| CN111723232B (en) * | 2020-07-06 | 2023-09-08 | 深圳高速公路集团数字科技有限公司 | Method and device for positioning and identifying building through Internet of things |
| CN114485605A (en) * | 2020-10-23 | 2022-05-13 | 丰田自动车株式会社 | Position specifying method and position specifying system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104112124A (en) | Image identification based indoor positioning method and device | |
| Chen et al. | City-scale landmark identification on mobile devices | |
| CN110017841A (en) | Vision positioning method and its air navigation aid | |
| CN110645986B (en) | Positioning method and device, terminal and storage medium | |
| CN106767812B (en) | A kind of indoor semantic map updating method and system based on Semantic features extraction | |
| CN107131883B (en) | Full-automatic mobile terminal indoor positioning system based on vision | |
| CN103870485B (en) | Method and device for achieving augmented reality application | |
| Ardeshir et al. | GIS-assisted object detection and geospatial localization | |
| CN101593205B (en) | Video-based method for searching three-dimension model | |
| EP2875471A1 (en) | Method of providing image feature descriptors | |
| CN104936283A (en) | Indoor positioning method, server and system | |
| CN103403736A (en) | Dynamic template tracking | |
| CN103778261A (en) | Self-guided tour method based on mobile cloud computing image recognition | |
| CN102317929A (en) | Method and system for image matching | |
| CN103561065A (en) | System and method for achieving 3D virtual advertisement with mobile terminal | |
| CN104897165A (en) | Shot scenery-based navigation method and system thereof | |
| CN102750554A (en) | Information processing device, object recognition method, program, and terminal device | |
| CN108564662A (en) | The method and device that augmented reality digital culture content is shown is carried out under a kind of remote scene | |
| US20230162394A1 (en) | Aligning and Augmenting a Partial Subspace of a Physical Infrastructure with at Least One Information Element | |
| CN110470295A (en) | A kind of indoor walking navigation and method based on AR positioning | |
| CN106354869A (en) | Real-scene image processing method and server based on location information and time periods | |
| CN111504331A (en) | Method and device for positioning panoramic intelligent vehicle from coarse to fine | |
| CN106649747A (en) | Scenic spot identification method and system | |
| US20180247122A1 (en) | Method and system of providing information pertaining to objects within premises | |
| CN106778649A (en) | A kind of image recognition algorithm of judgement sight spot marker |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141022 |
|
| RJ01 | Rejection of invention patent application after publication |