CN108491542A - A kind of trade mark searching system and method based on database building - Google Patents
A kind of trade mark searching system and method based on database building Download PDFInfo
- Publication number
- CN108491542A CN108491542A CN201810291399.1A CN201810291399A CN108491542A CN 108491542 A CN108491542 A CN 108491542A CN 201810291399 A CN201810291399 A CN 201810291399A CN 108491542 A CN108491542 A CN 108491542A
- Authority
- CN
- China
- Prior art keywords
- image
- unit
- trade mark
- memory
- pixel
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of trade mark searching system and method based on database building, including WEB server, search condition analytic unit, image matching unit, database server and isomeric data memory.The present invention pre-processes the mass image data in internet, feature extraction and partitioned storage, it builds in real time and updates the data library, pretreatment and feature extraction are carried out to trade mark to be retrieved, it is matched with the image data in the database of real-time update, it is not limited to the trade mark library of standard, range of search is more comprehensive.Since the image in trade mark to be retrieved and database has carried out pretreatment and feature extraction, matching characteristic point is more accurate, and identification error is small, substantially increases accuracy and the efficiency of retrieval.
Description
Technical field
The present invention relates to a kind of searching systems, and in particular to a kind of trade mark searching system and side based on database building
Method.Belong to picture search technical field.
Background technology
Trade mark is the mark of company, product or service, and it is one to melt with the commercial quality of enterprise, service quality, management
Body plays very important effect in industry and commerce society, is an important attribute of company and product.In order to enable trade mark by
To legal protection, generally require to trademark office's official register.With the quickening of China's expanding economy and globalization process, trade mark
Quantity cumulative year after year.It is the key problem of trade-mark administration to prevent repeated registration or similar brand registration.In order to protect registered trademark
Legitimate rights and interests, hit counterfeit illegal activities for usurping registered trademark, need to retrieve trade mark to be registered, and it is registered
Trade mark be compared, determine both differ or be not similar, just have registration and qualification.
Trademark image demand is continuously increased in recent years, and the retrieval of traditional trade mark is typically based on classification code and with a large amount of people
Power is cost, and retrieval accuracy and efficiency are all extremely low, does not adapt to the requirement of current a large amount of trade mark registrations increasingly.
If carrying out manual retrieval to trade mark to need to expend a large amount of manpowers, the mass data of relative interconnections net, speed right and wrong
It is often slow.And current Internal retrieval, such as Baidu, google etc., or based on keyword.Even some existing bases
In the trade-mark searching method of image, application also mainly searches identical similar trade mark in trade mark library.Figure in trade mark library
Seem by standardized, only includes trade mark, without other backgrounds.And trademark image is clear, proper.And search quotient in internet
In the case of mark, trade mark reappears in the picture, often will produce scale, rotation angle, illumination, the variation at visual angle, shape.
The Internet images are generally compressed more for the ease of transmission, and picture quality is not also high.This is method used in inquiry trade mark library
It is insurmountable.
In addition, the data volume of internet is very big and updates quickly, over time, data when trade mark is retrieved are visited
The amount of asking is increasing, on the one hand can seriously affect retrieval rate, on the other hand can do the retrieval work of many repeatability, cause to provide
Source wastes.
Invention content
The purpose of the present invention is to overcome above-mentioned the deficiencies in the prior art, provide a kind of trade mark inspection based on database building
Cable system and method.
To achieve the above object, the present invention uses following technical proposals：
A kind of trade mark searching system based on database building, including：
WEB server is equipped with man-machine interactive interface, for uploading trade mark to be retrieved, receiving retrieval result and showing；
Search condition analytic unit pre-processes for trade mark to be retrieved, and carries out feature extraction and storage；
Image matching unit, the image stored in the characteristics of image and database server for extracting trade mark to be retrieved
Feature is matched, and matching result is fed back to WEB server；
Database server is carried out for transferring the image data stored in isomeric data memory in real time after pretreatment
Feature extraction and partitioned storage；
Isomeric data memory, for storing image data and real-time update.
As one of preferred technical solution, the search condition analytic unit includes that sequentially connected first image is located in advance
Reason module, the first arithmetic device for executing feature extraction algorithm and the first storage for storing first arithmetic device operation result
Device.
As further preferred one of technical solution, described first image preprocessing module includes Image geometry transform list
Member, image denoising unit, image restoration unit, image enhancing unit and image normalization unit；
Described image geometrical transformation unit determines the gray value of each pixel of correction space using interpolation method three times；
Described image denoising unit is using nonlinear filtering method removal background noise and reduces in image transmitting process
The noise of doping；
Described image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Described image enhancement unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Described image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
As one of preferred technical solution, described image matching unit include execute matching algorithm second arithmetic device with
And the second memory for storing second arithmetic device operation result, second arithmetic device also respectively with first memory and database
Server connects.
As further preferred one of technical solution, the second memory is connect with WEB server.
As one of preferred technical solution, the database server includes sequentially connected second image preprocessing mould
Block, the third arithmetic unit for executing feature extraction algorithm and the third memory for storing third internalarithmetic result, institute
It includes several subregions divided sequentially in time to state third memory.
As further preferred one of technical solution, the second image pre-processing module is connect with isomeric data memory,
Third memory is connect with second arithmetic device.
As further preferred one of technical solution, second image pre-processing module includes Image geometry transform list
Member, image denoising unit, image restoration unit, image enhancing unit and image normalization unit；
Described image geometrical transformation unit determines the gray value of each pixel of correction space using interpolation method three times；
Described image denoising unit is using nonlinear filtering method removal background noise and reduces in image transmitting process
The noise of doping；
Described image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Described image enhancement unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Described image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
A kind of corresponding trade-mark searching method based on database building of above system, including step：
S1. trade mark to be retrieved is uploaded to WEB server；
S2. trade mark to be retrieved is pre-processed, feature extraction and storage；
S3. the image data stored in isomeric data memory is transferred in real time, and feature extraction and subregion are carried out after pretreatment
It stores to database server；
S4. the characteristics of image that step S2 is extracted is stored in the progress of the characteristics of image in database server with step S3
Match, corresponding isomeric data memory original image matching result is fed back into WEB server and shows.
As one of preferred technical solution, pretreated in step S2 the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
As one of preferred technical solution, step S2 extract be characterized in not being illuminated by the light, color, scale and rotationally-varying
The invariant feature of influence, specific extracting method are：
S2-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S2-2. each tomographic image is handled using various features detective operators；
S2-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S2-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
As one of preferred technical solution, feature extraction includes color characteristic and edge shape feature in step S2.
As further preferred one of technical solution, color characteristic is used to describe the scape corresponding to image or image-region
The surface nature of object, extracting method include color histogram, color set, color moment, color convergence vector, color correlogram etc.
Deng.
As further preferred one of technical solution, edge shape feature refers to that its surrounding pixel gray scale is jumpy
The set of those pixels, marginal existence are the most basic features of image, extracting method can between target, background and region
It is realized using any one of following edge detection algorithm：Sobel operator edge detections, Roberts operator edge detections,
Prewitt operator edge detections, Laplacian operator edge detections and Canny operator edge detections.
As one of preferred technical solution, the characteristics of image extracted in step S2 is stored to first memory.
As one of preferred technical solution, pretreated in step S3 the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
As one of preferred technical solution, step S3 extract be characterized in not being illuminated by the light, color, scale and rotationally-varying
The invariant feature of influence, specific extracting method are：
S3-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S3-2. each tomographic image is handled using various features detective operators；
S3-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S3-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
As one of preferred technical solution, feature extraction includes color characteristic and edge shape feature in step S3.
As further preferred one of technical solution, color characteristic is used to describe the scape corresponding to image or image-region
The surface nature of object, extracting method include color histogram, color set, color moment, color convergence vector, color correlogram etc.
Deng.
As further preferred one of technical solution, edge shape feature refers to that its surrounding pixel gray scale is jumpy
The set of those pixels, marginal existence are the most basic features of image, extracting method can between target, background and region
It is realized using any one of following edge detection algorithm：Sobel operator edge detections, Roberts operator edge detections,
Prewitt operator edge detections, Laplacian operator edge detections and Canny operator edge detections.
As one of preferred technical solution, the characteristics of image extracted in step S3 is stored to third memory.
As one of preferred technical solution, the matching result in step S4 is stored to second memory, second memory
It is connect with WEB server.
Beneficial effects of the present invention：
1, the present invention mass image data in internet is pre-processed, feature extraction and partitioned storage, take in real time
It builds and updates the data library, the image data in pretreatment and feature extraction, with the database of real-time update is carried out to trade mark to be retrieved
It is matched, is not limited to the trade mark library of standard, range of search is more comprehensive.Due to the image in trade mark to be retrieved and database
Pretreatment and feature extraction are carried out, matching characteristic point is more accurate, and identification error is small, substantially increases the accuracy of retrieval
And efficiency.
2, the extracted feature of image separately stores in trade mark and database to be retrieved, when carrying out matching operation, Ke Yizhi
It connects and transfers feature and matched with the characteristics of image in database server, improve retrieval accuracy, matching result is individually deposited
Storage is called for WEB server, is avoided repeatability retrieval work, is improved recall precision.
3, the pretreatment and feature extraction of image are very crucial in trade mark and database to be retrieved, and image of the invention is located in advance
It includes Image geometry transform unit, image denoising unit, image restoration unit, image enhancing unit and image normalization to manage module
Unit realizes the denoising and normalization of trademark image, is conducive to the accuracy for improving retrieval.
4, the present invention extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, match pair
As not by image irradiation, color, scale and it is rotationally-varying influenced, retrieval result is more comprehensive.
Description of the drawings
Fig. 1 is the system structure diagram of the present invention；
Wherein, 1 is WEB server, and 2 be search condition analytic unit, and 21 be the first image pre-processing module, and 22 be first
Arithmetic unit, 23 be first memory, and 3 be image matching unit, and 31 be second arithmetic device, and 32 be second memory, and 4 be database
Server, 41 be the second image pre-processing module, and 42 be third arithmetic unit, and 43 be third memory, and 5 store for isomeric data
Device.
Specific implementation mode
The present invention will be further elaborated with reference to the accompanying drawings and examples, it should which explanation, following the description is only
It is not to be defined to its content to explain the present invention.
Embodiment 1：
A kind of trade mark searching system based on database building as shown in Figure 1, including：
WEB server 1 is equipped with man-machine interactive interface, for uploading trade mark to be retrieved, receiving retrieval result and showing；
Search condition analytic unit 2 pre-processes for trade mark to be retrieved, and carries out feature extraction and storage；
The figure stored in image matching unit 3, characteristics of image for extracting trade mark to be retrieved and database server 4
As feature is matched, and matching result is fed back into WEB server 1；
Database server 4, for transferring the image data stored in isomeric data memory 5 in real time, pretreatment is laggard
Row feature extraction and partitioned storage；
Isomeric data memory 5, for storing image data and real-time update.
Wherein, search condition analytic unit 2 includes sequentially connected first image pre-processing module 21, executes feature extraction
The first arithmetic device 22 of algorithm and first memory 23 for storing 22 operation result of first arithmetic device.First image is located in advance
It includes Image geometry transform unit, image denoising unit, image restoration unit, image enhancing unit and image normalizing to manage module 21
Change unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Image matching unit 3 includes executing the second arithmetic device 31 of matching algorithm and being transported for storing second arithmetic device 31
The second memory 32 of result is calculated, second arithmetic device 31 is also connect with first memory 23 and database server 4 respectively.Second
Memory is connect with WEB server.
Database server 4 includes sequentially connected second image pre-processing module 41, executes the of feature extraction algorithm
Three arithmetic units 42 and third memory 43 for storing 42 operation result of third arithmetic unit, the third memory 43 include
Several subregions divided sequentially in time.Second image pre-processing module 41 is connect with isomeric data memory 5, and third is deposited
Reservoir 43 is connect with second arithmetic device 31.Second image pre-processing module 41 includes Image geometry transform unit, image denoising list
Member, image restoration unit, image enhancing unit and image normalization unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
A kind of corresponding trade-mark searching method based on database building of above system, including step：
S1. trade mark to be retrieved is uploaded to WEB server 1；
S2. trade mark to be retrieved is pre-processed, feature extraction and storage；
S3. the image data stored in isomeric data memory 5 is transferred in real time, and feature extraction and subregion are carried out after pretreatment
It stores to database server 4；
S4. the characteristics of image that step S2 is extracted the characteristics of image in database server 4 is stored in step S3 to carry out
Matching, feeds back to WEB server 1 by 5 original image matching result of corresponding isomeric data memory and shows.
It is pretreated in step S2 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S2 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S2-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S2-2. each tomographic image is handled using various features detective operators；
S2-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S2-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S2.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color histogram.Edge shape feature refers to its surrounding pixel
The set of those of gray scale change dramatically pixel, marginal existence are the most basic spies of image between target, background and region
Sign, extracting method are realized using Sobel operator edge detections.
The characteristics of image extracted in step S2 is stored to first memory 23.
It is pretreated in step S3 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S3 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S3-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S3-2. each tomographic image is handled using various features detective operators；
S3-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S3-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S3.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color histogram.Edge shape feature refers to its surrounding pixel
The set of those of gray scale change dramatically pixel, marginal existence are the most basic spies of image between target, background and region
Sign, extracting method are realized using Sobel operator edge detections.
The characteristics of image extracted in step S3 is stored to third memory 43.
Matching result in step S4 is stored to second memory 32, and second memory 32 is connect with WEB server 1.
Embodiment 2：
A kind of trade mark searching system based on database building as shown in Figure 1, including：
WEB server 1 is equipped with man-machine interactive interface, for uploading trade mark to be retrieved, receiving retrieval result and showing；
Search condition analytic unit 2 pre-processes for trade mark to be retrieved, and carries out feature extraction and storage；
The figure stored in image matching unit 3, characteristics of image for extracting trade mark to be retrieved and database server 4
As feature is matched, and matching result is fed back into WEB server 1；
Database server 4, for transferring the image data stored in isomeric data memory 5 in real time, pretreatment is laggard
Row feature extraction and partitioned storage；
Isomeric data memory 5, for storing image data and real-time update.
Wherein, search condition analytic unit 2 includes sequentially connected first image pre-processing module 21, executes feature extraction
The first arithmetic device 22 of algorithm and first memory 23 for storing 22 operation result of first arithmetic device.First image is located in advance
It includes Image geometry transform unit, image denoising unit, image restoration unit, image enhancing unit and image normalizing to manage module 21
Change unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Image matching unit 3 includes executing the second arithmetic device 31 of matching algorithm and being transported for storing second arithmetic device 31
The second memory 32 of result is calculated, second arithmetic device 31 is also connect with first memory 23 and database server 4 respectively.Second
Memory is connect with WEB server.
Database server 4 includes sequentially connected second image pre-processing module 41, executes the of feature extraction algorithm
Three arithmetic units 42 and third memory 43 for storing 42 operation result of third arithmetic unit, the third memory 43 include
Several subregions divided sequentially in time.Second image pre-processing module 41 is connect with isomeric data memory 5, and third is deposited
Reservoir 43 is connect with second arithmetic device 31.Second image pre-processing module 41 includes Image geometry transform unit, image denoising list
Member, image restoration unit, image enhancing unit and image normalization unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
A kind of corresponding trade-mark searching method based on database building of above system, including step：
S1. trade mark to be retrieved is uploaded to WEB server 1；
S2. trade mark to be retrieved is pre-processed, feature extraction and storage；
S3. the image data stored in isomeric data memory 5 is transferred in real time, and feature extraction and subregion are carried out after pretreatment
It stores to database server 4；
S4. the characteristics of image that step S2 is extracted the characteristics of image in database server 4 is stored in step S3 to carry out
Matching, feeds back to WEB server 1 by 5 original image matching result of corresponding isomeric data memory and shows.
It is pretreated in step S2 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S2 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S2-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S2-2. each tomographic image is handled using various features detective operators；
S2-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S2-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S2.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color set.Edge shape feature refers to its surrounding pixel gray scale
The set of those of change dramatically pixel, marginal existence are the most basic features of image between target, background and region,
Extracting method is realized using Roberts operator edge detections.
The characteristics of image extracted in step S2 is stored to first memory 23.
It is pretreated in step S3 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S3 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S3-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S3-2. each tomographic image is handled using various features detective operators；
S3-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S3-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S3.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color set.Edge shape feature refers to its surrounding pixel gray scale
The set of those of change dramatically pixel, marginal existence are the most basic features of image between target, background and region,
Extracting method is realized using Roberts operator edge detections.
The characteristics of image extracted in step S3 is stored to third memory 43.
Matching result in step S4 is stored to second memory 32, and second memory 32 is connect with WEB server 1.
Embodiment 3：
A kind of trade mark searching system based on database building as shown in Figure 1, including：
WEB server 1 is equipped with man-machine interactive interface, for uploading trade mark to be retrieved, receiving retrieval result and showing；
Search condition analytic unit 2 pre-processes for trade mark to be retrieved, and carries out feature extraction and storage；
The figure stored in image matching unit 3, characteristics of image for extracting trade mark to be retrieved and database server 4
As feature is matched, and matching result is fed back into WEB server 1；
Database server 4, for transferring the image data stored in isomeric data memory 5 in real time, pretreatment is laggard
Row feature extraction and partitioned storage；
Isomeric data memory 5, for storing image data and real-time update.
Wherein, search condition analytic unit 2 includes sequentially connected first image pre-processing module 21, executes feature extraction
The first arithmetic device 22 of algorithm and first memory 23 for storing 22 operation result of first arithmetic device.First image is located in advance
It includes Image geometry transform unit, image denoising unit, image restoration unit, image enhancing unit and image normalizing to manage module 21
Change unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Image matching unit 3 includes executing the second arithmetic device 31 of matching algorithm and being transported for storing second arithmetic device 31
The second memory 32 of result is calculated, second arithmetic device 31 is also connect with first memory 23 and database server 4 respectively.Second
Memory is connect with WEB server.
Database server 4 includes sequentially connected second image pre-processing module 41, executes the of feature extraction algorithm
Three arithmetic units 42 and third memory 43 for storing 42 operation result of third arithmetic unit, the third memory 43 include
Several subregions divided sequentially in time.Second image pre-processing module 41 is connect with isomeric data memory 5, and third is deposited
Reservoir 43 is connect with second arithmetic device 31.Second image pre-processing module 41 includes Image geometry transform unit, image denoising list
Member, image restoration unit, image enhancing unit and image normalization unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
A kind of corresponding trade-mark searching method based on database building of above system, including step：
S1. trade mark to be retrieved is uploaded to WEB server 1；
S2. trade mark to be retrieved is pre-processed, feature extraction and storage；
S3. the image data stored in isomeric data memory 5 is transferred in real time, and feature extraction and subregion are carried out after pretreatment
It stores to database server 4；
S4. the characteristics of image that step S2 is extracted the characteristics of image in database server 4 is stored in step S3 to carry out
Matching, feeds back to WEB server 1 by 5 original image matching result of corresponding isomeric data memory and shows.
It is pretreated in step S2 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S2 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S2-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S2-2. each tomographic image is handled using various features detective operators；
S2-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S2-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S2.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color moment.Edge shape feature refers to its surrounding pixel gray scale
The set of those of change dramatically pixel, marginal existence are the most basic features of image between target, background and region,
Extracting method is realized using Prewitt operator edge detections.
The characteristics of image extracted in step S2 is stored to first memory 23.
It is pretreated in step S3 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S3 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S3-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S3-2. each tomographic image is handled using various features detective operators；
S3-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S3-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S3.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color moment.Edge shape feature refers to its surrounding pixel gray scale
The set of those of change dramatically pixel, marginal existence are the most basic features of image between target, background and region,
Extracting method is realized using Prewitt operator edge detections.
The characteristics of image extracted in step S3 is stored to third memory 43.
Matching result in step S4 is stored to second memory 32, and second memory 32 is connect with WEB server 1.
Embodiment 4：
A kind of trade mark searching system based on database building as shown in Figure 1, including：
WEB server 1 is equipped with man-machine interactive interface, for uploading trade mark to be retrieved, receiving retrieval result and showing；
Search condition analytic unit 2 pre-processes for trade mark to be retrieved, and carries out feature extraction and storage；
The figure stored in image matching unit 3, characteristics of image for extracting trade mark to be retrieved and database server 4
As feature is matched, and matching result is fed back into WEB server 1；
Database server 4, for transferring the image data stored in isomeric data memory 5 in real time, pretreatment is laggard
Row feature extraction and partitioned storage；
Isomeric data memory 5, for storing image data and real-time update.
Wherein, search condition analytic unit 2 includes sequentially connected first image pre-processing module 21, executes feature extraction
The first arithmetic device 22 of algorithm and first memory 23 for storing 22 operation result of first arithmetic device.First image is located in advance
It includes Image geometry transform unit, image denoising unit, image restoration unit, image enhancing unit and image normalizing to manage module 21
Change unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Image matching unit 3 includes executing the second arithmetic device 31 of matching algorithm and being transported for storing second arithmetic device 31
The second memory 32 of result is calculated, second arithmetic device 31 is also connect with first memory 23 and database server 4 respectively.Second
Memory is connect with WEB server.
Database server 4 includes sequentially connected second image pre-processing module 41, executes the of feature extraction algorithm
Three arithmetic units 42 and third memory 43 for storing 42 operation result of third arithmetic unit, the third memory 43 include
Several subregions divided sequentially in time.Second image pre-processing module 41 is connect with isomeric data memory 5, and third is deposited
Reservoir 43 is connect with second arithmetic device 31.Second image pre-processing module 41 includes Image geometry transform unit, image denoising list
Member, image restoration unit, image enhancing unit and image normalization unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
A kind of corresponding trade-mark searching method based on database building of above system, including step：
S1. trade mark to be retrieved is uploaded to WEB server 1；
S2. trade mark to be retrieved is pre-processed, feature extraction and storage；
S3. the image data stored in isomeric data memory 5 is transferred in real time, and feature extraction and subregion are carried out after pretreatment
It stores to database server 4；
S4. the characteristics of image that step S2 is extracted the characteristics of image in database server 4 is stored in step S3 to carry out
Matching, feeds back to WEB server 1 by 5 original image matching result of corresponding isomeric data memory and shows.
It is pretreated in step S2 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S2 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S2-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S2-2. each tomographic image is handled using various features detective operators；
S2-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S2-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S2.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color convergence vector.Edge shape feature refers to picture around it
The set of those of plain gray scale change dramatically pixel, marginal existence are the most basic spies of image between target, background and region
Sign, extracting method are realized using Laplacian operator edge detections.
The characteristics of image extracted in step S2 is stored to first memory 23.
It is pretreated in step S3 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S3 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S3-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S3-2. each tomographic image is handled using various features detective operators；
S3-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S3-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S3.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color convergence vector.Edge shape feature refers to picture around it
The set of those of plain gray scale change dramatically pixel, marginal existence are the most basic spies of image between target, background and region
Sign, extracting method are realized using Laplacian operator edge detections.
The characteristics of image extracted in step S3 is stored to third memory 43.
Matching result in step S4 is stored to second memory 32, and second memory 32 is connect with WEB server 1.
Embodiment 5：
A kind of trade mark searching system based on database building as shown in Figure 1, including：
WEB server 1 is equipped with man-machine interactive interface, for uploading trade mark to be retrieved, receiving retrieval result and showing；
Search condition analytic unit 2 pre-processes for trade mark to be retrieved, and carries out feature extraction and storage；
The figure stored in image matching unit 3, characteristics of image for extracting trade mark to be retrieved and database server 4
As feature is matched, and matching result is fed back into WEB server 1；
Database server 4, for transferring the image data stored in isomeric data memory 5 in real time, pretreatment is laggard
Row feature extraction and partitioned storage；
Isomeric data memory 5, for storing image data and real-time update.
Wherein, search condition analytic unit 2 includes sequentially connected first image pre-processing module 21, executes feature extraction
The first arithmetic device 22 of algorithm and first memory 23 for storing 22 operation result of first arithmetic device.First image is located in advance
It includes Image geometry transform unit, image denoising unit, image restoration unit, image enhancing unit and image normalizing to manage module 21
Change unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Image matching unit 3 includes executing the second arithmetic device 31 of matching algorithm and being transported for storing second arithmetic device 31
The second memory 32 of result is calculated, second arithmetic device 31 is also connect with first memory 23 and database server 4 respectively.Second
Memory is connect with WEB server.
Database server 4 includes sequentially connected second image pre-processing module 41, executes the of feature extraction algorithm
Three arithmetic units 42 and third memory 43 for storing 42 operation result of third arithmetic unit, the third memory 43 include
Several subregions divided sequentially in time.Second image pre-processing module 41 is connect with isomeric data memory 5, and third is deposited
Reservoir 43 is connect with second arithmetic device 31.Second image pre-processing module 41 includes Image geometry transform unit, image denoising list
Member, image restoration unit, image enhancing unit and image normalization unit；
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
A kind of corresponding trade-mark searching method based on database building of above system, including step：
S1. trade mark to be retrieved is uploaded to WEB server 1；
S2. trade mark to be retrieved is pre-processed, feature extraction and storage；
S3. the image data stored in isomeric data memory 5 is transferred in real time, and feature extraction and subregion are carried out after pretreatment
It stores to database server 4；
S4. the characteristics of image that step S2 is extracted the characteristics of image in database server 4 is stored in step S3 to carry out
Matching, feeds back to WEB server 1 by 5 original image matching result of corresponding isomeric data memory and shows.
It is pretreated in step S2 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S2 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S2-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S2-2. each tomographic image is handled using various features detective operators；
S2-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S2-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S2.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color correlogram.Edge shape feature refers to its surrounding pixel
The set of those of gray scale change dramatically pixel, marginal existence are the most basic spies of image between target, background and region
Sign, extracting method are realized using Canny operator edge detections.
The characteristics of image extracted in step S2 is stored to first memory 23.
It is pretreated in step S3 that the specific method is as follows：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is removed background noise and is reduced in image transmitting process and adulterated using nonlinear filtering method
Noise；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
Step S3 extract be characterized in not being illuminated by the light, the invariant feature of color, scale and rotationally-varying influence, it is specific to extract
Method is：
S3-1. it is made of multiple frequency ranges, successive bands ruler as gaussian pyramid using sampling and Gaussian convolution structural map
Degree differs 50%, and the multiple sublayers of Gaussian convolution construction are utilized in each frequency range；
S3-2. each tomographic image is handled using various features detective operators；
S3-3. in each frequency range, to each pixel in each sublayer, compare the feature in neighborhood on scale space
The handling result value of detective operators, if the end value on the pixel is maximum value or minimum value in its neighborhood, just by it
As candidate feature point, the frequency range that it occurs, sublayer serial number, the coordinate information in image are recorded；
S3-4. the point repeated in candidate feature point is removed, weak contrast and adjacent edges in candidate feature point are then removed
Point, obtain invariant feature point.
Feature extraction includes color characteristic and edge shape feature in step S3.Color characteristic is for describing image or image
The surface nature of scenery corresponding to region, extracting method are color correlogram.Edge shape feature refers to its surrounding pixel
The set of those of gray scale change dramatically pixel, marginal existence are the most basic spies of image between target, background and region
Sign, extracting method are realized using Canny operator edge detections.
The characteristics of image extracted in step S3 is stored to third memory 43.
Matching result in step S4 is stored to second memory 32, and second memory 32 is connect with WEB server 1.
Above-mentioned, although the foregoing specific embodiments of the present invention is described with reference to the accompanying drawings, not protects model to the present invention
The limitation enclosed, based on the technical solutions of the present invention, those skilled in the art, which need not make the creative labor, to be done
The various modifications or changes gone out are still within protection scope of the present invention.
Claims (10)
1. a kind of trade mark searching system based on database building, which is characterized in that including：
WEB server is equipped with man-machine interactive interface, for uploading trade mark to be retrieved, receiving retrieval result and showing；
Search condition analytic unit pre-processes for trade mark to be retrieved, and carries out feature extraction and storage；
Image matching unit, the characteristics of image stored in the characteristics of image and database server for extracting trade mark to be retrieved
It is matched, and matching result is fed back into WEB server；
Database server carries out feature for transferring the image data stored in isomeric data memory in real time after pretreatment
Extraction and partitioned storage；
Isomeric data memory, for storing image data and real-time update.
2. trade mark searching system according to claim 1, which is characterized in that the search condition analytic unit includes successively
Connection the first image pre-processing module, execute feature extraction algorithm first arithmetic device and for store first arithmetic device fortune
Calculate the first memory of result.
3. trade mark searching system according to claim 1, which is characterized in that described image matching unit includes executing matching
The second arithmetic device of algorithm and second memory for storing second arithmetic device operation result, second arithmetic device also respectively with
First memory is connected with database server.
4. trade mark searching system according to claim 1, which is characterized in that the database server includes being sequentially connected
The second image pre-processing module, execute feature extraction algorithm third arithmetic unit and for storing third internalarithmetic knot
The third memory of fruit, the third memory include several subregions divided sequentially in time.
5. a kind of corresponding trade-mark searching method based on database building of any one of Claims 1 to 4 system, special
Sign is, including step：
S1. trade mark to be retrieved is uploaded to WEB server；
S2. trade mark to be retrieved is pre-processed, feature extraction and storage；
S3. the image data stored in isomeric data memory is transferred in real time, and feature extraction and partitioned storage are carried out after pretreatment
To database server；
S4. the characteristics of image that step S2 is extracted the characteristics of image in database server is stored in step S3 to match,
Corresponding isomeric data memory original image matching result is fed back into WEB server and is shown.
6. trade-mark searching method according to claim 5, which is characterized in that pretreated specific method is such as in step S2
Under：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is made an uproar using what is adulterated in nonlinear filtering method removal background noise and reduction image transmitting process
Sound；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
7. trade-mark searching method according to claim 5, which is characterized in that step S2 extract be characterized in not being illuminated by the light,
The invariant feature of color, scale and rotationally-varying influence.
8. trade-mark searching method according to claim 5, which is characterized in that the characteristics of image extracted in step S2 store to
First memory.
9. trade-mark searching method according to claim 5, which is characterized in that pretreated specific method is such as in step S3
Under：
Image geometry transform unit determines the gray value of each pixel of correction space using interpolation method three times；
Image denoising unit is made an uproar using what is adulterated in nonlinear filtering method removal background noise and reduction image transmitting process
Sound；
Image restoration unit is degenerated using the image caused by Wiener filtering restoring method correction a variety of causes；
Image enhancing unit is selectively reinforced and is inhibited to the information in image using Gabor filtering enhancing method；
Image normalization unit obtains the image with invariance using the normalization algorithm based on image pixel.
10. trade-mark searching method according to claim 5, which is characterized in that step S3 extract be characterized in not being illuminated by the light,
The invariant feature of color, scale and rotationally-varying influence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810291399.1A CN108491542A (en) | 2018-04-03 | 2018-04-03 | A kind of trade mark searching system and method based on database building |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810291399.1A CN108491542A (en) | 2018-04-03 | 2018-04-03 | A kind of trade mark searching system and method based on database building |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108491542A true CN108491542A (en) | 2018-09-04 |
Family
ID=63318306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810291399.1A Withdrawn CN108491542A (en) | 2018-04-03 | 2018-04-03 | A kind of trade mark searching system and method based on database building |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108491542A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111026896A (en) * | 2019-11-15 | 2020-04-17 | 浙江大华技术股份有限公司 | Characteristic value storage and processing method, device and storage device |
| CN114996500A (en) * | 2021-03-02 | 2022-09-02 | 贵州凌恒科技有限公司 | Trademark graph retrieval method |
-
2018
- 2018-04-03 CN CN201810291399.1A patent/CN108491542A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111026896A (en) * | 2019-11-15 | 2020-04-17 | 浙江大华技术股份有限公司 | Characteristic value storage and processing method, device and storage device |
| CN111026896B (en) * | 2019-11-15 | 2023-09-01 | 浙江大华技术股份有限公司 | Feature value storage and processing method, device and storage device |
| CN114996500A (en) * | 2021-03-02 | 2022-09-02 | 贵州凌恒科技有限公司 | Trademark graph retrieval method |
| CN114996500B (en) * | 2021-03-02 | 2024-05-17 | 贵州凌恒科技有限公司 | Trademark graph retrieval method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106845408B (en) | Street garbage identification method under complex environment | |
| CN107688806B (en) | Affine transformation-based free scene text detection method | |
| CN111177446B (en) | Method for searching footprint image | |
| CN107103314B (en) | A kind of fake license plate vehicle retrieval system based on machine vision | |
| Fantoni et al. | Accurate and automatic alignment of range surfaces | |
| US20060126943A1 (en) | Geometric hashing method for model-based recognition of an object | |
| CN108664556A (en) | Parallel search system and method suitable for associated mark | |
| CN108491542A (en) | A kind of trade mark searching system and method based on database building | |
| CN105183857A (en) | Automatic picture training sample extracting method and system | |
| CN110503051B (en) | Precious wood identification system and method based on image identification technology | |
| Wei et al. | Detection of lane line based on Robert operator | |
| Yang et al. | A research of feature-based image mosaic algorithm | |
| Li et al. | RIFT2: Speeding-up RIFT with a new rotation-invariance technique | |
| CN108470073A (en) | Searching system and method suitable for associated mark | |
| CN108416068A (en) | A kind of trade mark parallel search system and method | |
| Tang et al. | Application of SSD framework model in detection of logs end | |
| CN108446405A (en) | A kind of trade mark parallel search system and method based on database building | |
| CN108845997A (en) | A kind of trade mark searching system and method | |
| CN116363655A (en) | Financial bill identification method and system | |
| EP4089644A1 (en) | Image matching system | |
| Ye | Fast and robust registration of multimodal remote sensing images via dense orientated gradient feature | |
| CN108268533A (en) | A kind of Image Feature Matching method for image retrieval | |
| Prakash et al. | Combining novel features for content based image retrieval | |
| Deren et al. | Automatic change detection of geo-spatial data from imagery | |
| Yildirim et al. | Comparison of image matching algorithms on satellite images taken in different seasons |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180904 |