CN110134815A - Image processing method, device, computer equipment and storage medium - Google Patents
Image processing method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110134815A CN110134815A CN201910298047.3A CN201910298047A CN110134815A CN 110134815 A CN110134815 A CN 110134815A CN 201910298047 A CN201910298047 A CN 201910298047A CN 110134815 A CN110134815 A CN 110134815A
- Authority
- CN
- China
- Prior art keywords
- image
- processed
- alternative
- location information
- characteristic
- 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
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
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/587—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
This application involves big data technical field, a kind of image processing method, device, computer equipment and storage medium are provided.Method includes: to extract the characteristic of image to be processed, and search each alternative image comprising characteristic, the feature vector of image to be processed and each alternative image is obtained respectively, according to the feature vector of the feature vector of image to be processed and each alternative image, calculate separately the similitude of image to be processed Yu each alternative image, according to the Similarity measures result of image to be processed and each alternative image, determine carried in each alternative image location information and with the highest target image of the similitude of image to be processed, extract the location information that target image carries, and location information is added to image to be processed, obtain treated image.Since the target image of acquisition carries location information and image similarity to be processed height, using the corresponding location information of target image as the location information of image to be processed, the image for carrying location information is obtained.
Description
Technical field
This application involves big data technical field, more particularly to a kind of image processing method, device, computer equipment and
Storage medium.
Background technique
With the development of big data technology, data analysis plays an important role in more and more fields.Traditional
In big data analysis treatment process, typically the text information based on acquisition is analyzed.
With the diversity of information representation mode, people increasingly like the life by uploaded videos or image recording oneself
Work is daily, sufficiently excavate the hiding information in picture, such as excavate picture shooting position etc., for carrying out the behavior of user and disappearing
The analysis of expense habit is of great significance.
In image taking, can largely carry Exif (Exchangeable image file format, it is commutative
Image file format) image parameter informations such as parameter can directly acquire the shooting of picture for including the picture of Exif parameter
The information such as position, but if picture, through overcompression, the location information of image can be lost, the position due to that can not obtain picture is believed
Breath, leads to not accurately carry out data analysis.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of can be realized to the image for not carrying location information
Carry out image processing method, device, computer equipment and the storage medium of location information addition.
A kind of image processing method, which comprises
The characteristic of image to be processed is extracted, and searches each alternative image comprising the characteristic;
Obtain the feature vector of the image to be processed and the feature vector of each alternative image;
According to the feature vector of the feature vector of the image to be processed and each alternative image, calculate separately it is described to
Handle the similitude of image and each alternative image;
According to the Similarity measures of the image to be processed and each alternative image as a result, determining each alternative image
In carry location information and the highest target image of similitude with the image to be processed;
The location information that the target image carries is extracted, and the location information is added to the image to be processed,
Obtain treated image.
The characteristic for extracting image to be processed in one of the embodiments, and searching includes the characteristic
According to each alternative image include:
Image preprocessing is carried out to image to be processed, extracts the characteristic of the pretreated image to be processed of described image
According to;
The image comprising the characteristic is searched from network image data, and is filtered out in these images and carried position
Each alternative image of confidence breath.
The lookup from network image data includes the image of the characteristic in one of the embodiments, and
It filters out and carries each alternative image of location information in these images and include:
The corresponding data category of characteristic for obtaining the image to be processed, according to the data category determine it is described to
Handle the feature tag of image;
According to the feature tag of the image to be processed, the net that the feature tag is carried in network image data is searched for
Network image;
The location information of the network image that parsing search obtains, and filter out carried in the network image it is described
Each alternative image of location information.
The location information for extracting the target image and carrying in one of the embodiments, and the position is believed
Breath is added to the image to be processed, before the image that obtains that treated, further includes:
The image for not carrying location information in each alternative image is chosen, image set to be processed is constructed;
The location information for extracting the target image and carrying, and the location information is added to the figure to be processed
Picture, obtains that treated that image includes:
Extract the location information that the target image carries, and by the location information be added to the image to be processed with
And each alternative image in the image set to be processed;
Obtain treated each image, and will treated that each described image is updated to the presetting database, it is described pre-
If database is used to store the image for carrying location information.
The characteristic for extracting image to be processed in one of the embodiments, and searching includes the characteristic
According to each alternative image include:
Image preprocessing is carried out to image to be processed, extracts the feature for passing through the pretreated image to be processed of described image
Data;
Presetting database is traversed, each alternative image in the presetting database comprising the characteristic is searched, it is described
The image for carrying location information is stored in presetting database.
The feature vector according to the image to be processed and each alternative image in one of the embodiments,
Feature vector, the similitude for calculating separately the image to be processed and each alternative image include:
The feature vector of the image to be processed is subjected to Jaccard system with the feature vector of each alternative image respectively
Number measuring and calculating determines the corresponding Jaccard coefficient value of each alternative image;
The Similarity measures according to the image to be processed and each alternative image are as a result, determine each described alternative
The target image for meeting preset requirement in image includes:
It is maximum and carry each alternative figure of location information to choose corresponding Jaccard coefficient value in each alternative image
As being used as target image.
It is described in one of the embodiments, to choose corresponding Jaccard coefficient value maximum in the alternative image and take
Image with location information includes: as target image
The image for carrying location information in each alternative image is chosen, image set to be selected is constructed;
According to described wait select corresponding Jaccard coefficient value size in alternative image described in image set, described in selection
Wait select in image set the corresponding maximum image of Jaccard coefficient value as the target image.
A kind of image processing apparatus, described device include:
Characteristic extraction module includes the characteristic for extracting the characteristic of image to be processed, and searching
Each alternative image;
Feature vector obtains module, for obtain the image to be processed feature vector and each alternative image
Feature vector;
Similarity computing module, for according to the feature vector of the image to be processed and the feature of each alternative image
Vector calculates separately the similitude of the image to be processed Yu each alternative image;
Target image determining module, for the Similarity measures knot according to the image to be processed and each alternative image
Fruit, determine carried in each alternative image location information and with the highest target figure of the similitude of the image to be processed
Picture;
Location information adding module, the location information carried for extracting the target image, and by the location information
It is added to the image to be processed, the image that obtains that treated.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
Device performs the steps of when executing the computer program
The characteristic of image to be processed is extracted, and searches each alternative image comprising the characteristic;
Obtain the feature vector of the image to be processed and the feature vector of each alternative image;
According to the feature vector of the feature vector of the image to be processed and each alternative image, calculate separately it is described to
Handle the similitude of image and each alternative image;
According to the Similarity measures of the image to be processed and each alternative image as a result, determining each alternative image
In carry location information and the highest target image of similitude with the image to be processed;
The location information that the target image carries is extracted, and the location information is added to the image to be processed,
Obtain treated image.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
It is performed the steps of when row
The characteristic of image to be processed is extracted, and searches each alternative image comprising the characteristic;
Obtain the feature vector of the image to be processed and the feature vector of each alternative image;
According to the feature vector of the feature vector of the image to be processed and each alternative image, calculate separately it is described to
Handle the similitude of image and each alternative image;
According to the Similarity measures of the image to be processed and each alternative image as a result, determining each alternative image
In carry location information and the highest target image of similitude with the image to be processed;
The location information that the target image carries is extracted, and the location information is added to the image to be processed,
Obtain treated image.
Above-mentioned image processing method, device, computer equipment and storage medium, by the characteristic for extracting image to be processed
According to searching to obtain each alternative image, similar with image to be processed each image can be obtained, and to be processed by obtaining respectively
It is similar to each alternative image to calculate separately image to be processed using feature vector for the feature vector of image and each alternative image
Property, it filters out and carries location information in each alternative image and with the highest image of the similitude of image to be processed as target
Image extracts the location information that target image carries, and the location information as image to be processed is added to image to be processed, due to
The target image of acquisition and image similarity to be processed are high, can be using the corresponding location information of target image as image to be processed
Location information, obtain the image for carrying location information.
Detailed description of the invention
Fig. 1 is the application scenario diagram of image processing method in one embodiment;
Fig. 2 is the flow diagram of image processing method in one embodiment;
Fig. 3 is the flow diagram of image processing method in another embodiment;
Fig. 4 is the flow diagram of the sub-step of step S240 in Fig. 3 in one embodiment;
Fig. 5 is the flow diagram of image processing method in another embodiment;
Fig. 6 is the flow diagram of image processing method in another embodiment;
Fig. 7 is the structural block diagram of image processing apparatus in one embodiment;
Fig. 8 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Image processing method provided by the present application can be applied in application environment as shown in Figure 1.Wherein, terminal 102
It is communicated with server 104 by network by network.Terminal 102 will carry out the to be processed of processing location information analysis
Image is sent to server 104, and server 104 extracts the characteristic of image to be processed, and searches each comprising characteristic
Alternative image obtains the feature vector of image to be processed and the feature vector of each alternative image, according to the spy of image to be processed
The feature vector for levying vector and each alternative image, calculates separately the similitude of image to be processed Yu each alternative image, according to wait locate
Similarity measures of reason image and each alternative image as a result, determine carried in each alternative image location information and with figure to be processed
The highest image of the similitude of picture extracts the location information that target image carries as target image, and location information is added
To image to be processed, the image that obtains that treated, by treated, the image added with location information pushes to end to server 104
End 102.Wherein, terminal 102 can be, but not limited to be various personal computers, laptop, smart phone, tablet computer and
Portable wearable device, server 104 can use the server cluster of the either multiple server compositions of independent server
To realize.
In actual application, get it should be understood that location information image when, first to image carry out parameter
Analysis can parse the location information in the Exif parameter of image carrying, then return corresponding if image carries Exif parameter
Location information does not carry Exif parameter, is then sent to the image as image to be processed if image is managed through overcompression etc.
Server is handled.
In one embodiment, as shown in Fig. 2, providing a kind of image processing method, it is applied in Fig. 1 in this way
It is illustrated for server, comprising the following steps:
Step S200 extracts the characteristic of image to be processed, and searches each alternative image comprising characteristic.
Image to be processed refers to after overcompression or extended formatting processing, and the attribute of digital photograph is recorded not comprising Exif etc.
The parameter of information and photographed data cannot directly obtain the image of location information by Parameter analysis of electrochemical.Characteristic is from one group
That selects in given feature can effectively identify the minimal feature subset of target, and feature such as color, texture are extracted from image
Or the characteristics such as shape, different features can choose specific extracting method, color histogram such as based on color characteristic,
Gray level co-occurrence matrixes and wavelet transformation based on texture, the FInite Element based on shape, rotation function and wavelet descriptor etc..To
Characteristic in processing image can characterize the stationarity elements such as landscape or building in image.According to characteristic, depositing
It contains and searches the image comprising same characteristic features data in the presetting database of great amount of images, alternately image, alternative image are
Refer to for obtaining location information, and can be using the location information as the image of the location information of image to be processed.It can use
Images match model carries out the acquisition of alternative image, and specifically, the characteristic input of image to be processed is trained in advance
Images match model, using preparatory trained images match model, matching includes identical spy from existing image data
The image of data is levied, and alternately image, the training process of images match model include: that acquisition carries characteristic for output
Image pattern collection, identical label will be marked comprising the sample image of same characteristic features data, sample image will be inputted into initial graph
As Matching Model is trained, according to the match condition of label in training result, the evaluation ginseng of initial pictures Matching Model is obtained
Number, wherein evaluation parameter includes recall rate, accuracy rate and accurate rate.It is right when evaluation parameter is not up to preset requirement range
Initial pictures Matching Model after training carries out parameter adjustment, when the parameter value of evaluation parameter reaches preset requirement range, instruction
Practicing the initial pictures Matching Model completed is preset images match model.It in other embodiments, can also be by net
The lookup of image data present in network obtains multiple alternative images.
Step S300 obtains the feature vector of image to be processed and the feature vector of each alternative image.
Feature vector refers to the data mode of vector mode expression image, by extracting each dimension component of image
Feature, each dimension are indicated by one group of attribute, to the orientation and radian of attribute further described including edge, side and line
Length and curvature, the size in region.Feature vector can be single vector, be also possible to the Vector Groups of multiple vector compositions.?
In embodiment, image to be processed can also be divided into multiple subgraphs, obtain the corresponding feature vector of each subgraph, utilized
The subgraph of image to be processed is compared with numerous images, can increase the success rate of images match.
Step S400 is calculated separately according to the feature vector of the feature vector of image to be processed and each alternative image wait locate
Manage the similitude of image and each alternative image.
The similitude of image refers to the characterization parameter of the consistency of picture material, and the similitude of image can be by calculating two
The similarity factor value of the feature vector of a image determines, such as Jaccard coefficient, Jaccard coefficient distance etc..In embodiment
In, it is first determined image category of image to be processed, such as landscape class, Building class etc. are found mutually similar according to image category
Other image calculates the similitude of image to be processed Yu each image of same category one by one.
Step S500, according to the Similarity measures of image to be processed and each alternative image as a result, determining in each alternative image
Carry location information and with the highest target image of the similitude of image to be processed.
Location information can be the location information that the self-contained Exif parameter etc. of image includes, and be also possible to mark in advance
Position data label etc..Location information will be carried and with the highest image of the similitude of image to be processed as target figure
Picture, the location information of available image to be processed.In embodiment, it may be implemented to obtain the image work for carrying location information
It directly will be similar to image to be processed after calculating the similitude of each alternative image and image to be processed for alternative image
Highest image is spent as target image, saves the data processing amount of Similarity measures, in other embodiments, can also first be obtained
Take alternative image, calculate the similitude of each alternative image and image to be processed, when with the highest figure of the similarity of image to be processed
As using the image as target image, not taken when with the highest image of the similarity of image to be processed when carrying location information
When band location information, according to the sequence from high to low of similarity, sequence in the alternative image of arrangement is selected to carry position letter
It ceases and sorts near preceding image as target image.
Step S600 extracts the location information that target image carries, and location information is added to image to be processed, obtains
Treated image.
When target image carries Exif parameter, the taking location information in Exif parameter is extracted, when target image is taken
When with position data label, corresponding position data label is obtained, the location information of extraction can be used as data label addition
To image to be processed, the image for carrying location information is obtained, the corresponding shooting geographical location of image to be processed is determined, is convenient for
Subsequent progress data analysis, realizes that the information of image data is deeply excavated, and is conducive to the quality for improving analysis of image data.
Above-mentioned image processing method searches to obtain each alternative image by extracting the characteristic of image to be processed, can
To obtain similar with image to be processed each image, and the feature by obtaining image to be processed and each alternative image respectively to
Amount, the similitude of image to be processed Yu each alternative image is calculated separately using feature vector, filters out and takes in each alternative image
With location information and with the highest image of the similitude of image to be processed as target image, the position that target image carries is extracted
Confidence breath, the location information as image to be processed are added to image to be processed, due to the target image and figure to be processed of acquisition
As similarity height, can obtain carrying position using the corresponding location information of target image as the location information of image to be processed
The image of confidence breath.
In one embodiment, as shown in figure 3, step S200, extracts the characteristic of image to be processed, and search and include
Each alternative image of characteristic includes:
Step S220 carries out image preprocessing to image to be processed, the spy of the image to be processed after extracting image preprocessing
Levy data.
Step S240 searches the image comprising characteristic from network image data, and filters out and take in these images
Each alternative image with location information.
Image preprocessing refers in image analysis, is carried out before carrying out feature extraction, segmentation and matching to input picture
Processing, eliminate unrelated information in image, restore useful real information, enhance detectability and maximum limit for information about
Degree ground simplifies data, to improve the data handling procedure of the reliability of feature extraction, image segmentation, matching and identification.Network
Image data can be the image of the storages such as each webpage or platform, there are a part of image be by compression of images in these images
Processing, location information can not be obtained, when selecting alternative image, needs to screen image, obtains including position letter
The image of breath alternately image.
In one embodiment, as shown in figure 4, step S240, is searched from network image data comprising characteristic
Image, and filter out and carry each alternative image of location information in these images and include:
Step S242 obtains the corresponding data category of characteristic of image to be processed, determines the feature of image to be processed
Label.
Step S244 is searched in network image data according to the feature tag of image to be processed and is carried feature tag
Network image.
Step S246, the location information for the network image that parsing search obtains, and filter out and carry position in network image
Each alternative image of confidence breath.
Image in network image data includes the image for carrying the location informations such as Exif parameter or position data label
The image of any location information is not carried.According to the data category of the characteristic of image to be analyzed, image to be analyzed is determined
Feature tag, feature tag includes image attributes and attribute subdivision classification, for example, the image attributes of image to be analyzed can be with
It is landscape, building etc., landscape subdivision classification may include forest, lofty mountains, lake, beach etc., and building subdivision classification may include
Various architectural styles etc..According to the feature tag of image to be processed, image search expression is constructed, search carries same characteristic features
The network image of label, the location information for the network image that parsing search obtains, and filter out and carry position in network image
The alternative image of information.
In one embodiment, as shown in figure 5, step S600, the location information that target image carries is extracted, and by position
Information is added to image to be processed, before the image that obtains that treated, further includes:
Step S580 chooses the image for not carrying location information in each alternative image, constructs image set to be processed.
Step S600 extracts the location information that target image carries, and location information is added to image to be processed, obtains
Treated, and image includes:
Step S620, extract target image carry location information, and by location information be added to image to be processed and
Each alternative image in image set to be processed.
Step S640, obtains treated each image, and will treated each image update to presetting database, present count
It is used to store the image for carrying location information according to library.
Location information is added to the alternative image of image to be analyzed concentration, determines the location information of each alternative image, it will
The location information of image to be processed is labeled as the image to be analyzed and concentrates the position data label of each alternative image, and will carry
The image update of position data label realizes the expansion of image data in presetting database to presetting database.Work as preset data
When the amount of images in library reaches the requirement of setting, can directly by predetermined data stream carry out target image lookup, without into
Row carries the screening of the image of location information, reduces data processing amount.
In one embodiment, step S200 extracts the characteristic of image to be processed, and searches comprising characteristic
Each alternative image includes:
Image preprocessing is carried out to image to be processed, extracts the characteristic of the image to be processed after image preprocessing
According to.
Traverse presetting database, search include in presetting database characteristic each alternative image, in presetting database
It is stored with the image for carrying location information.
Presetting database is for storing the image for carrying location information, by constructing presetting database, to preset data
Image in library is searched, the screening of the image without not carried location information, is conducive to improve data-handling efficiency.
In one embodiment, as shown in fig. 6, step S400, according to the feature vector of image to be processed and each alternative figure
The feature vector of picture, the similitude for calculating separately image to be processed and each alternative image include:
The feature vector of image to be processed is carried out Jaccard with the feature vector of each alternative image respectively by step S420
Coefficient measuring and calculating, determines the corresponding Jaccard coefficient value of each alternative image.
Step S500, according to the Similarity measures of image to be processed and each alternative image as a result, determining in each alternative image
The target image for meeting preset requirement includes:
Step S520 chooses corresponding Jaccard coefficient value maximum in each alternative image and carries each of location information
Alternative image is as target image.
The feature vector of feature vector and each alternative image to image to be processed carries out the measuring and calculating of Jaccard coefficient, determines
Similitude in each alternative image with image to be processed, the calculation of Jaccard coefficient are as follows: two groups of characteristic vector data intersections
The ratio of number and two groups of characteristic vector data union numbers.Jaccard coefficient value is bigger, and image similarity is higher, chooses
Alternative image in the maximum data of Jaccard coefficient value can be obtained with image similarity to be processed most as target image
High image.
In one embodiment, step S520 chooses corresponding Jaccard coefficient value maximum in alternative image and carries
The image of location information includes: as target image
The image for carrying location information in each alternative image is chosen, image set to be selected is constructed.
According to wait select in image set corresponding Jaccard coefficient value size in alternative image, image set to be selected is selected
In the corresponding maximum image of Jaccard coefficient value as target image.
In the selection course of target image, can first it be ranked up by the size of Jaccard coefficient value, in sequence
Successively, it selects to carry location information and the forward image that sorts is as target image, can also first carry out carrying positional number
According to image screening, select screening after image in the maximum image of Jaccard coefficient value as target image.
It should be understood that although each step in the flow chart of Fig. 2-6 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, these steps can execute in other order.Moreover, at least one in Fig. 2-6
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, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in fig. 7, providing a kind of image processing apparatus, comprising:
Characteristic extraction module 200 for extracting the characteristic of image to be processed, and is searched comprising characteristic
Each alternative image;
Feature vector obtains module 300, for obtaining the feature vector of image to be processed and the feature of each alternative image
Vector;
Similarity computing module 400, for the feature vector according to the feature vector of image to be processed and each alternative image,
Calculate separately the similitude of image to be processed Yu each alternative image;
Target image determining module 500, for the Similarity measures according to image to be processed and each alternative image as a result, really
Carried in fixed each alternative image location information and with the highest target image of the similitude of image to be processed;
Location information adding module 600 for extracting the location information of target image carrying, and location information is added to
Image to be processed obtains treated image.
In one embodiment, characteristic extraction module 200 is also used to carry out image preprocessing to image to be processed,
The characteristic of image to be processed after extracting image preprocessing searches the figure comprising characteristic from network image data
Picture, and filter out each alternative image that location information is carried in these images.
In one embodiment, characteristic extraction module 200, the characteristic for being also used to obtain image to be processed are corresponding
Data category, determine the feature tag of image to be processed, according to the feature tag of image to be processed, search for network image data
In carry the network image of feature tag, the location information for the network image that parsing search obtains, and filter out network image
In carry each alternative image of location information.
In one embodiment, image processing apparatus further includes image set building module to be processed, each alternative for choosing
The image for not carrying location information in image, constructs image set to be processed, and location information adding module 600 is also used to extract mesh
The location information that logo image carries, and location information is added to each alternative figure in image to be processed and image set to be processed
Picture, obtains treated each image, and will treated each image update to presetting database, presetting database is taken for storing
Image with location information.
In one embodiment, characteristic extraction module 200 is also used to carry out image preprocessing to image to be processed,
The characteristic of the image to be processed after image preprocessing is extracted, presetting database is traversed, searches and is wrapped in presetting database
Each alternative image containing characteristic is stored with the image for carrying location information in presetting database.
In one embodiment, similarity computing module 400 are also used to the feature vector of image to be processed respectively and respectively
The feature vector of alternative image carries out the measuring and calculating of Jaccard coefficient, determines the corresponding Jaccard coefficient value of each alternative image;Target
Image determining module 500 is also used to choose corresponding Jaccard coefficient value maximum in each alternative image and carries location information
Each alternative image as target image.
In one embodiment, target image determining module 500, be also used to choose carried in each alternative image position letter
The image of breath constructs image set to be selected, according to wait select in image set, corresponding Jaccard coefficient value is big in alternative image
It is small, it selects wait select in image set the corresponding maximum image of Jaccard coefficient value as target image.
Above-mentioned image processing apparatus searches to obtain each alternative image by extracting the characteristic of image to be processed, can
To obtain similar with image to be processed each image, and the feature by obtaining image to be processed and each alternative image respectively to
Amount, the similitude of image to be processed Yu each alternative image is calculated separately using feature vector, filters out and takes in each alternative image
With location information and with the highest image of the similitude of image to be processed as target image, the position that target image carries is extracted
Confidence breath, the location information as image to be processed are added to image to be processed, due to the target image and figure to be processed of acquisition
As similarity height, can obtain carrying position using the corresponding location information of target image as the location information of image to be processed
The image of confidence breath.
Specific about image processing apparatus limits the restriction that may refer to above for image processing method, herein not
It repeats again.Modules in above-mentioned image processing apparatus can be realized fully or partially through software, hardware and combinations thereof.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, the corresponding operation of the above modules is executed in order to which processor calls.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 8.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing product analysis data.The network interface of the computer equipment is used to pass through with external terminal
Network connection communication.To realize a kind of image processing method when the computer program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 8, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, which is stored with
Computer program, the processor perform the steps of when executing computer program
The characteristic of image to be processed is extracted, and searches each alternative image comprising characteristic;
Obtain the feature vector of image to be processed and the feature vector of each alternative image;
According to the feature vector of the feature vector of image to be processed and each alternative image, image to be processed and each is calculated separately
The similitude of alternative image;
Position is carried as a result, determining in each alternative image according to the Similarity measures of image to be processed and each alternative image
Information and with the highest target image of the similitude of image to be processed;
The location information that target image carries is extracted, and location information is added to image to be processed, treated for acquisition
Image.
In one embodiment, it is also performed the steps of when processor executes computer program
Image preprocessing is carried out to image to be processed, the characteristic of the image to be processed after extracting image preprocessing;
The image comprising characteristic is searched from network image data, and is filtered out and carried position letter in these images
Each alternative image of breath.
In one embodiment, it is also performed the steps of when processor executes computer program
The corresponding data category of characteristic for obtaining image to be processed, the spy of image to be processed is determined according to data category
Levy label;
According to the feature tag of image to be processed, the network image that feature tag is carried in network image data is searched for;
The location information for the network image that parsing search obtains, and filter out and carry each of location information in network image
Alternative image.
In one embodiment, it is also performed the steps of when processor executes computer program
The image for not carrying location information in each alternative image is chosen, image set to be processed is constructed;
The location information that target image carries is extracted, and location information is added to image to be processed and image to be processed
Each alternative image concentrated;
Obtain treated each image, and will treated each image update to presetting database, presetting database is used for
Storage carries the image of location information.
In one embodiment, it is also performed the steps of when processor executes computer program
Image preprocessing is carried out to image to be processed, extracts the characteristic of the image to be processed after image preprocessing
According to;
Traverse presetting database, search include in presetting database characteristic each alternative image, in presetting database
It is stored with the image for carrying location information.
In one embodiment, it is also performed the steps of when processor executes computer program
The feature vector of image to be processed is subjected to the measuring and calculating of Jaccard coefficient with the feature vector of each alternative image respectively,
Determine the corresponding Jaccard coefficient value of each alternative image;
It is maximum and carry each alternative image of location information and make to choose corresponding Jaccard coefficient value in each alternative image
For target image.
In one embodiment, it is also performed the steps of when processor executes computer program
The image for carrying location information in each alternative image is chosen, image set to be selected is constructed;
According to wait select in image set corresponding Jaccard coefficient value size in alternative image, image set to be selected is selected
In the corresponding maximum image of Jaccard coefficient value as target image.
The above-mentioned computer equipment for realizing image processing method is looked by extracting the characteristic of image to be processed
Find each alternative image, can obtain each image similar with image to be processed, and by obtain respectively image to be processed with
The feature vector of each alternative image calculates separately the similitude of image to be processed Yu each alternative image using feature vector, screening
Location information is carried in each alternative image out and with the highest image of the similitude of image to be processed as target image, mentioned
The location information for taking target image to carry, the location information as image to be processed is added to image to be processed, due to acquisition
Target image and image similarity to be processed are high, can be using the corresponding location information of target image as the position of image to be processed
Information obtains the image for carrying location information.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
The characteristic of image to be processed is extracted, and searches each alternative image comprising characteristic;
Obtain the feature vector of image to be processed and the feature vector of each alternative image;
According to the feature vector of the feature vector of image to be processed and each alternative image, image to be processed and each is calculated separately
The similitude of alternative image;
Position is carried as a result, determining in each alternative image according to the Similarity measures of image to be processed and each alternative image
Information and with the highest target image of the similitude of image to be processed;
The location information that target image carries is extracted, and location information is added to image to be processed, treated for acquisition
Image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Image preprocessing is carried out to image to be processed, the characteristic of the image to be processed after extracting image preprocessing;
The image comprising characteristic is searched from network image data, and is filtered out and carried position letter in these images
Each alternative image of breath.
In one embodiment, it is also performed the steps of when computer program is executed by processor
The corresponding data category of characteristic for obtaining image to be processed, the spy of image to be processed is determined according to data category
Levy label;
According to the feature tag of image to be processed, the network image that feature tag is carried in network image data is searched for;
The location information for the network image that parsing search obtains, and filter out and carry each of location information in network image
Alternative image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
The image for not carrying location information in each alternative image is chosen, image set to be processed is constructed;
The location information that target image carries is extracted, and location information is added to image to be processed and image to be processed
Each alternative image concentrated;
Obtain treated each image, and will treated each image update to presetting database, presetting database is used for
Storage carries the image of location information.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Image preprocessing is carried out to image to be processed, extracts the characteristic of the image to be processed after image preprocessing
According to;
Traverse presetting database, search include in presetting database characteristic each alternative image, in presetting database
It is stored with the image for carrying location information.
In one embodiment, it is also performed the steps of when computer program is executed by processor
The feature vector of image to be processed is subjected to the measuring and calculating of Jaccard coefficient with the feature vector of each alternative image respectively,
Determine the corresponding Jaccard coefficient value of each alternative image;
It is maximum and carry each alternative image of location information and make to choose corresponding Jaccard coefficient value in each alternative image
For target image.
In one embodiment, it is also performed the steps of when computer program is executed by processor
The image for carrying location information in each alternative image is chosen, image set to be selected is constructed;
According to wait select in image set corresponding Jaccard coefficient value size in alternative image, image set to be selected is selected
In the corresponding maximum image of Jaccard coefficient value as target image.
The above-mentioned computer readable storage medium for realizing image processing method, by the feature for extracting image to be processed
Data search to obtain each alternative image, can obtain similar with image to be processed each image, and by being obtained respectively wait locate
The feature vector for managing image and each alternative image, it is similar to each alternative image to calculate separately image to be processed using feature vector
Property, it filters out and carries location information in each alternative image and with the highest image of the similitude of image to be processed as target
Image extracts the location information that target image carries, and the location information as image to be processed is added to image to be processed, due to
The target image of acquisition and image similarity to be processed are high, can be using the corresponding location information of target image as image to be processed
Location information, obtain the image for carrying location information.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Instruct relevant hardware to complete by computer program, computer program to can be stored in a non-volatile computer readable
It takes in storage medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, this Shen
Please provided by any reference used in each embodiment to memory, storage, database or other media, may each comprise
Non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
Above embodiments only express the several embodiments of the application, and the description thereof is more specific and detailed, but can not
Therefore it is construed as limiting the scope of the patent.It should be pointed out that for those of ordinary skill in the art,
Under the premise of not departing from the application design, various modifications and improvements can be made, these belong to the protection scope of the application.
Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of image processing method, which comprises
The characteristic of image to be processed is extracted, and searches each alternative image comprising the characteristic;
Obtain the feature vector of the image to be processed and the feature vector of each alternative image;
According to the feature vector of the feature vector of the image to be processed and each alternative image, calculate separately described to be processed
The similitude of image and each alternative image;
It is taken according to the Similarity measures of the image to be processed and each alternative image as a result, determining in each alternative image
With location information and with the highest target image of similitude of the image to be processed;
The location information that the target image carries is extracted, and the location information is added to the image to be processed, is obtained
Treated image.
2. the method according to claim 1, wherein the characteristic for extracting image to be processed, and searching
Each alternative image comprising the characteristic includes:
Image preprocessing is carried out to image to be processed, extracts the characteristic of the pretreated image to be processed of described image;
The image comprising the characteristic is searched from network image data, and is filtered out and carried position letter in these images
Each alternative image of breath.
3. according to the method described in claim 2, it is characterized in that, described search from network image data includes the feature
The image of data, and filter out and carry each alternative image of location information in these images and include:
The corresponding data category of characteristic for obtaining the image to be processed, determines described to be processed according to the data category
The feature tag of image;
According to the feature tag of the image to be processed, the network that the feature tag is carried in network image data is searched for
Picture;
The location information for the network image that parsing search obtains, and filter out and carry the position in the network image
Each alternative image of information.
4. according to the method in claim 2 or 3, which is characterized in that the position letter for extracting the target image and carrying
Breath, and the location information is added to the image to be processed, before the image that obtains that treated, further includes:
The image for not carrying location information in each alternative image is chosen, image set to be processed is constructed;
The location information for extracting the target image and carrying, and the location information is added to the image to be processed,
Obtain that treated that image includes:
The location information that the target image carries is extracted, and the location information is added to the image to be processed and institute
State each alternative image in image set to be processed;
Obtain treated each image, and will treated that each described image is updated to presetting database, the presetting database
For storing the image for carrying location information.
5. the method according to claim 1, wherein the characteristic for extracting image to be processed, and searching
Each alternative image comprising the characteristic includes:
Image preprocessing is carried out to image to be processed, extracts the characteristic for passing through the pretreated image to be processed of described image
According to;
Presetting database is traversed, each alternative image in the presetting database comprising the characteristic is searched, it is described default
The image for carrying location information is stored in database.
6. the method according to claim 1, wherein the feature vector according to the image to be processed and each
The feature vector of the alternative image, the similitude for calculating separately the image to be processed and each alternative image include:
The feature vector of the image to be processed is subjected to the survey of Jaccard coefficient with the feature vector of each alternative image respectively
It calculates, determines the corresponding Jaccard coefficient value of each alternative image;
The Similarity measures according to the image to be processed and each alternative image are as a result, determine each alternative image
The middle target image for meeting preset requirement includes:
It is maximum and carry each alternative image of location information and make to choose corresponding Jaccard coefficient value in each alternative image
For target image.
7. according to the method described in claim 6, it is characterized in that, described choose corresponding Jaccard in the alternative image
Coefficient value is maximum and carries the image of location information and includes: as target image
The image for carrying location information in each alternative image is chosen, image set to be selected is constructed;
According to described wait select corresponding Jaccard coefficient value size in alternative image described in image set, select described to be selected
The corresponding maximum image of Jaccard coefficient value is selected in image set as the target image.
8. a kind of image processing apparatus, which is characterized in that described device includes:
Characteristic extraction module for extracting the characteristic of image to be processed, and is searched each comprising the characteristic
Alternative image;
Feature vector obtains module, for obtaining the feature vector of the image to be processed and the feature of each alternative image
Vector;
Similarity computing module, for according to the feature of the feature vector of the image to be processed and each alternative image to
Amount, calculates separately the similitude of the image to be processed Yu each alternative image;
Target image determining module, for according to the Similarity measures of the image to be processed and each alternative image as a result,
It determines and carries location information and the highest target image of similitude with the image to be processed in each alternative image;
Location information adding module, the location information carried for extracting the target image, and the location information is added
To the image to be processed, the image that obtains that treated.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 7 the method when executing the computer program.
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 method described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910298047.3A CN110134815A (en) | 2019-04-15 | 2019-04-15 | Image processing method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910298047.3A CN110134815A (en) | 2019-04-15 | 2019-04-15 | Image processing method, device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110134815A true CN110134815A (en) | 2019-08-16 |
Family
ID=67569921
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910298047.3A Pending CN110134815A (en) | 2019-04-15 | 2019-04-15 | Image processing method, device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110134815A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113688266A (en) * | 2021-09-13 | 2021-11-23 | 上海联影医疗科技股份有限公司 | Image priority determining method, image processing device, image processing apparatus, and recording medium |
CN116188805A (en) * | 2023-04-26 | 2023-05-30 | 青岛尘元科技信息有限公司 | Image content analysis method and device for massive images and image information network |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011238057A (en) * | 2010-05-11 | 2011-11-24 | Univ Of Electro-Communications | Image ranking method, program, and storage medium and image display system |
CN103678661A (en) * | 2013-12-24 | 2014-03-26 | 中国联合网络通信集团有限公司 | Image searching method and terminal |
US20150120760A1 (en) * | 2013-10-31 | 2015-04-30 | Adobe Systems Incorporated | Image tagging |
CN104765877A (en) * | 2015-04-27 | 2015-07-08 | 苏州大学 | Photo processing method and system |
CN105243060A (en) * | 2014-05-30 | 2016-01-13 | 小米科技有限责任公司 | Picture retrieval method and apparatus |
CN105354307A (en) * | 2015-11-06 | 2016-02-24 | 腾讯科技(深圳)有限公司 | Image content identification method and apparatus |
CN107633066A (en) * | 2017-09-21 | 2018-01-26 | 北京金山安全软件有限公司 | Information display method and device, electronic equipment and storage medium |
-
2019
- 2019-04-15 CN CN201910298047.3A patent/CN110134815A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011238057A (en) * | 2010-05-11 | 2011-11-24 | Univ Of Electro-Communications | Image ranking method, program, and storage medium and image display system |
US20150120760A1 (en) * | 2013-10-31 | 2015-04-30 | Adobe Systems Incorporated | Image tagging |
CN103678661A (en) * | 2013-12-24 | 2014-03-26 | 中国联合网络通信集团有限公司 | Image searching method and terminal |
CN105243060A (en) * | 2014-05-30 | 2016-01-13 | 小米科技有限责任公司 | Picture retrieval method and apparatus |
CN104765877A (en) * | 2015-04-27 | 2015-07-08 | 苏州大学 | Photo processing method and system |
CN105354307A (en) * | 2015-11-06 | 2016-02-24 | 腾讯科技(深圳)有限公司 | Image content identification method and apparatus |
CN107633066A (en) * | 2017-09-21 | 2018-01-26 | 北京金山安全软件有限公司 | Information display method and device, electronic equipment and storage medium |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113688266A (en) * | 2021-09-13 | 2021-11-23 | 上海联影医疗科技股份有限公司 | Image priority determining method, image processing device, image processing apparatus, and recording medium |
CN116188805A (en) * | 2023-04-26 | 2023-05-30 | 青岛尘元科技信息有限公司 | Image content analysis method and device for massive images and image information network |
CN116188805B (en) * | 2023-04-26 | 2023-08-04 | 青岛尘元科技信息有限公司 | Image content analysis method and device for massive images and image information network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11704357B2 (en) | Shape-based graphics search | |
CN109308678B (en) | Method, device and equipment for repositioning by using panoramic image | |
US20120027311A1 (en) | Automated image-selection method | |
CN109643318B (en) | Content-based searching and retrieval of brand images | |
US20120030575A1 (en) | Automated image-selection system | |
CN102483767A (en) | Object mapping device, method of mapping object, program and recording medium | |
CN110765882B (en) | Video tag determination method, device, server and storage medium | |
CN107590811B (en) | Scene segmentation based landscape image processing method and device and computing equipment | |
CN107392238A (en) | Outdoor knowledge of plants based on moving-vision search expands learning system | |
CN110136198A (en) | Image processing method and its device, equipment and storage medium | |
CN108875797A (en) | A kind of method of determining image similarity, photograph album management method and relevant device | |
US9665773B2 (en) | Searching for events by attendants | |
JP2021530773A (en) | Image-based relationship analysis methods and their systems | |
CN109800319A (en) | Image processing method, device, computer equipment and storage medium | |
CN110134815A (en) | Image processing method, device, computer equipment and storage medium | |
CN111507285A (en) | Face attribute recognition method and device, computer equipment and storage medium | |
US20120027303A1 (en) | Automated multiple image product system | |
CN114049631A (en) | Data labeling method and device, computer equipment and storage medium | |
JP2006119723A (en) | Device and method for image processing | |
CN109711381A (en) | Target identification method, device and the computer equipment of remote sensing images | |
KR20130038547A (en) | System for dual-searching image using region of interest set and method therefor | |
CN105913024A (en) | Android mobile terminal detecting method based on LAP operator for resisting replay attacks | |
Tomei et al. | Image-to-image translation to unfold the reality of artworks: an empirical analysis | |
CN113822899A (en) | Image processing method, image processing device, computer equipment and storage medium | |
CN107220299A (en) | Picture retrieval method and device |
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 |