CN103632143A - Cloud computing combined object identifying system on basis of images - Google Patents

Cloud computing combined object identifying system on basis of images Download PDF

Info

Publication number
CN103632143A
CN103632143A CN201310651359.0A CN201310651359A CN103632143A CN 103632143 A CN103632143 A CN 103632143A CN 201310651359 A CN201310651359 A CN 201310651359A CN 103632143 A CN103632143 A CN 103632143A
Authority
CN
China
Prior art keywords
image data
module
image
cloud computing
thing
Prior art date
Application number
CN201310651359.0A
Other languages
Chinese (zh)
Other versions
CN103632143B (en
Inventor
陈志沅
黄暐皓
Original Assignee
冠捷显示科技(厦门)有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 冠捷显示科技(厦门)有限公司 filed Critical 冠捷显示科技(厦门)有限公司
Priority to CN201310651359.0A priority Critical patent/CN103632143B/en
Publication of CN103632143A publication Critical patent/CN103632143A/en
Application granted granted Critical
Publication of CN103632143B publication Critical patent/CN103632143B/en

Links

Abstract

The invention discloses a cloud computing combined object identifying system on the basis of images. The cloud computing combined object identifying system is characterized in that image data captured by an image capturing module are characterized and processed by a preceding-stage processing module, acquired object image data divided by a dividing module are subjected to similarity comparison computation in a local knowledge base module, and compared and screened acquired object image data are uploaded to a cloud knowledge base module and are identified and analyzed after characteristic values of the acquired object image data are extracted by a representation and description module. The cloud computing combined object identifying system has the advantages that the cloud computing combined object identifying system is high in identification ability and wide in identification scope, while powerful hardware for identifying gestures, a large storage space for storing gesture templates and shortening of operation time of processors are required in the prior art, and accordingly the cloud computing combined object identifying system with simplified hardware, large storage space and short operation time of processors can be widely applied to electronic equipment with the requirement of image identifying and matching.

Description

Article identification in conjunction with cloud computing based on image

Technical field

The invention belongs to image identification field, specifically, relate to a kind of combination cloud computing on electronic equipments such as can being used in recognition of face, behavior identification, gesture identification, object identification article identification based on image.

Background technology

Image recognition technology has been tending towards ripe at present, the speed detecting is fast, accuracy is high, in most of existing systems, by different gestures, express the different meanings, the shape of gesture is very useful for man-machine interaction, but in current system, images of gestures is identified to coupling, hardware device and the storage space that need to have very high configuration, the processor calculating time is long, and for the identification of images of gestures, can only be limited to several patterns of minority, majority can only be used by unit, cannot identify coupling to complicated image, along with the development of image recognition technology and user's demand, realize long-range, intelligentized online share service is trend of the times.

Summary of the invention

In view of this, the invention provides the article identification of a kind of combination cloud computing based on image.This system identification ability is high, identification category extensively and not needs the hardware requirement of high standard.

For realizing above object, the present invention by the following technical solutions: the article identification of a kind of combination cloud computing based on image, apply to carry out on the electronic equipment of image recognition coupling, this system comprises,

Image acquiring module, for capturing video data;

Prime processing module, for receiving the image data that described image acquiring module catches the feature of extracting image data;

Cut apart module, described in cut apart module and prime processing module is extracted to image data after feature divide into and gather thing image data and non-collection thing image data, and the cutting of collections thing image data is stored in to local side base module.

Local side base module, for access, gather thing image data and be added with several and gather thing template, the collection thing template of the collection thing image data of storage and interpolation is done to preliminary similarity contrast conting, and the collection thing image data filtering out is more afterwards uploaded to high in the clouds base module;

Represent and describing module, extract collection thing image data further Eigenvalues analysis in local side base module;

High in the clouds base module, gathers thing template for access, and upgrades and add collection thing template to local side base module;

Identification and parsing module, the collection thing image data that described expression and describing module are done after further Eigenvalues analysis is done similarity contrast conting with the collection thing template in the base module of described high in the clouds, identifies collection thing image data.

Further, after described image acquiring module first leaves in local side knowledge base by the image data of catching, then image data is sent to prime processing module.

Further, the image data feature that described prime processing module is extracted comprises: the combination of one or more in edge, COLOR COMPOSITION THROUGH DISTRIBUTION, material or angle point.

Further, described prime processing module is before extracting image data feature, the image data of first described image acquiring module being caught carries out that size is dwindled or other accelerate in identifications place, such as taking out the processing such as low-frequency data or statistical value side figure, phasor projection.

Further, the collection thing template in described local side base module is upgraded or is added by high in the clouds base module.

Further, described Indigenous knowledge library module is uploaded to high in the clouds base module by the described collection thing image data of cutting apart module cutting by the original ROI image in image acquiring module.

Further, described identification and parsing module are done eigenwert sum operation to the content gathering in thing image data, utilize different eigenwerts to do different descriptions by gathering thing image data.

Further, the collection thing template in described high in the clouds base module contrasts data used, will do corresponding renewal or interpolation according to different image data features.

Further, the collection thing image data identifying in described expression and describing module is stored in the knowledge base of high in the clouds and uses as gathering the contrast of thing template.

Further, described identification and parsing module make an explanation the collection thing image data identifying, and return to the result of explanation.

The present invention adopts above technical scheme, the image data that image acquiring module is caught carries out characterization through prime processing module, cutting apart the collection thing image data that module cuts out, in local side base module, carry out similarity contrast conting, and the collection thing image data filtering out is more afterwards uploaded to high in the clouds base module, by representing and describing module shifts to an earlier date after eigenwert identification and resolves and gathers thing image data; System identification ability provided by the invention is high, identification category is wide, and current techniques need to be done gesture identification by a powerful hardware, also the operation time that needs a very large storage area to store gesture template and need to consume processor, the operation time of hardware of simplifying of the present invention, storage area and processor.Can apply to widely to carry out on the electronic equipment of image recognition coupling.

Accompanying drawing explanation

Now by reference to the accompanying drawings the present invention is further described:

Fig. 1 be the present invention in conjunction with cloud computing the article identification structural representation based on image;

Fig. 2 is article identification local side system processor of the present invention schematic diagram processed;

Fig. 3 is article identification of the present invention high in the clouds system processor schematic diagram processed.

Embodiment

Refer to shown in Fig. 1, the article identification of a kind of combination cloud computing provided by the invention based on image, applies to carry out on the electronic equipment of image recognition coupling, and this system comprises,

Image acquiring module 101, for capturing video data; Prime processing module 102, for receiving the image data that described image acquiring module 101 catches the feature of extracting image data; Cut apart module 103, described in cut apart module 103 and prime processing module 102 is extracted to image data after features divide into and gather thing image data and non-collection thing image data, and the cutting of collections thing image data is stored in to local side base module 104.

Local side base module 104, for access, gather thing image data and be added with several and gather thing template, the collection thing template of the collection thing image data of storage and interpolation is done to preliminary similarity contrast conting, and the collection thing image data filtering out is more afterwards uploaded to high in the clouds base module 106;

Represent and describing module 105, extract collection thing image data further Eigenvalues analysis in local side base module;

High in the clouds base module 106, gathers thing template for access, and upgrades and add collection thing template to local side base module 104;

Identification and parsing module 107, the collection thing image data that described expression and describing module 105 are done after further Eigenvalues analysis is done similarity contrast conting with the collection thing template in described high in the clouds base module 106, identifies collection thing image data.

Referring to Fig. 1 and Fig. 2 is described in detail local side system processing operation.

In step 201, image acquiring module 101 first leaves the image data capturing in local side base module 104 in, while depositing, for the image data that gathers thing and non-collection thing formation, as an image data, store, then transfer to prime processing module 102 to do preliminary image data and process;

In step 202, prime processing module 102 first the image data size receiving is dwindled or other accelerate in identification place, for example, takes out the processing such as low-frequency data, statistical value side figure or phasor projection.While doing preliminary computing in order to hardware itself, reduce its required usefulness;

In step 203, prime processing module 102 is done characterization by image data, as Sobal, Canny Edge ... Deng edge extraction or other eigenwert technology, get its strong feature as comparison foundation as color Histogram etc., the feature of required extraction comprises one or more the combination in the features such as edge, COLOR COMPOSITION THROUGH DISTRIBUTION, material, Corner;

In step 204, cut apart module 103 for the scope that gathers thing, do dividing processing, be specially: by B prime processing module 102, process the data of coming, at this by simple comparison mechanism as eigenwert comparison, the cutting of front background, dynamically cutting is divided into the image that may gather thing and non-collection thing by image data as SVM, Adaboost etc. as MHI etc. or use sorter, and may gather thing image data and cut and be stored in local side base module 106, at this, ignore the image of non-collection thing.

In step 205, local side base module 104 is done preliminary similarity contrast conting by cutting apart the collection thing template that may gather thing image data and storage that module 103 cutting obtains, by comparing with collection thing template, collection thing is done to preliminary classification, local side base module 104 stores and gathers thing template, and described collection thing template is upgraded and added by high in the clouds base module 106;

In step 206, the original ROI image data that the relative position of cutting apart collection thing place block is obtained by image acquiring module 101;

In step 207, the collection thing image data that the original ROI image data obtaining is filtered out is more afterwards uploaded high in the clouds base module 106 and is done meticulous comparison.

By this kind of mode, can not must the whole Zhang Shangchuan of image high in the clouds base module 106 is made comparison, only need upload gathering thing image data block.

Referring to Fig. 3 is described in detail high in the clouds system processing operation.

In step 208, the collection thing image data receiving is carried out to content analysis, do more complicated eigenwert sum operation.As SIFT, HOG, AAM, LBP etc., draw the eigenwert after conversion.Utilize the different eigenwert after transforming to do different descriptions, the foundation using different description contents as identification by gathering thing image data.

In step 209, by as sorters such as neural network, SVM by the collection thing image data of required differentiation by image acquiring module 101, prime processing module 102, cut apart strong eigenwert resolution information training gained judgement information storage that collection thing image data that module 103, expression and describing module 105 and identification and parsing module 107 get transformed out in local side base module 104, as the template comparison that gathers thing, use.In high in the clouds base module 106, gather the data of thing template comparison, can upgrade and add according to required identification objects feature.

In step 210, by by representing and describing module 105 is carried out the collection thing image data after Eigenvalues analysis, carry out with the stored collection thing template of high in the clouds base module 106 that similarity is meticulous to be compared, collection thing image data is distinguished.

In step 211, the collection thing image data result after passback identification is explained.

Although specifically show and introduced this patent in conjunction with preferred embodiment; but those skilled in the art should be understood that; within not departing from the spirit and scope of the present invention that appended claims limits; can make a variety of changes the present invention in the form and details, be protection scope of the present invention.

Claims (10)

1. the article identification based on image in conjunction with cloud computing, applies to carry out, on the electronic equipment of image recognition coupling, it is characterized in that: this system comprises,
Image acquiring module, for capturing video data;
Prime processing module, for receiving the image data that described image acquiring module catches the feature of extracting image data;
Cut apart module, described in cut apart module and prime processing module is extracted to image data after feature divide into and gather thing image data and non-collection thing image data, and the cutting of collections thing image data is stored in to local side base module;
Local side base module, for access, gather thing image data and be added with several and gather thing template, the collection thing template of the collection thing image data of storage and interpolation is done to preliminary similarity contrast conting, and the collection thing image data filtering out is more afterwards uploaded to high in the clouds base module;
Represent and describing module, extract collection thing image data further Eigenvalues analysis in local side base module;
High in the clouds base module, gathers thing template for access, and upgrades and add collection thing template to local side base module;
Identification and parsing module, the collection thing image data that described expression and describing module are done after further Eigenvalues analysis is done similarity contrast conting with the collection thing template in the base module of described high in the clouds, identifies collection thing image data.
2. the article identification of combination cloud computing according to claim 1 based on image, is characterized in that: after described image acquiring module first leaves in local side knowledge base by the image data of catching, then image data is sent to prime processing module.
3. the article identification of combination cloud computing according to claim 1 based on image, is characterized in that: the image data feature that described prime processing module is extracted comprises: the combination of one or more in edge, COLOR COMPOSITION THROUGH DISTRIBUTION, material or angle point.
4. the article identification based on image according to the combination cloud computing described in claim 1,2 or 3, it is characterized in that: described prime processing module is before extracting image data feature, and the image data of first described image acquiring module being caught carries out size and dwindles processing.
5. the article identification of combination cloud computing according to claim 1 based on image, is characterized in that: the collection thing template in described local side base module is upgraded or added by high in the clouds base module.
6. the article identification of combination cloud computing according to claim 1 based on image, is characterized in that: described Indigenous knowledge library module is uploaded to high in the clouds base module by the described collection thing image data of cutting apart module cutting by the original ROI image in image acquiring module.
7. the article identification of combination cloud computing according to claim 1 based on image, it is characterized in that: described identification and parsing module are done eigenwert sum operation to the content gathering in thing image data, utilize different eigenwerts to do different descriptions by gathering thing image data.
8. the article identification of combination cloud computing according to claim 1 based on image, is characterized in that: the collection thing template in described high in the clouds base module contrasts data used, will do corresponding renewal or interpolation according to different image data features.
9. the article identification based on image according to the combination cloud computing described in claim 1 or 7, is characterized in that: the collection thing image data identifying in described expression and describing module is stored in the knowledge base of high in the clouds and uses as gathering the contrast of thing template.
10. the article identification of combination cloud computing according to claim 1 based on image, is characterized in that: described identification and parsing module make an explanation the collection thing image data identifying, and return to the result of explanation.
CN201310651359.0A 2013-12-05 2013-12-05 Cloud computing combined object identifying system on basis of images CN103632143B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310651359.0A CN103632143B (en) 2013-12-05 2013-12-05 Cloud computing combined object identifying system on basis of images

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310651359.0A CN103632143B (en) 2013-12-05 2013-12-05 Cloud computing combined object identifying system on basis of images

Publications (2)

Publication Number Publication Date
CN103632143A true CN103632143A (en) 2014-03-12
CN103632143B CN103632143B (en) 2017-02-08

Family

ID=50213172

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310651359.0A CN103632143B (en) 2013-12-05 2013-12-05 Cloud computing combined object identifying system on basis of images

Country Status (1)

Country Link
CN (1) CN103632143B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105205436A (en) * 2014-06-03 2015-12-30 北京创思博德科技有限公司 Gesture identification system based on multiple forearm bioelectric sensors
CN105224066A (en) * 2014-06-03 2016-01-06 北京创思博德科技有限公司 A kind of gesture identification method based on high in the clouds process
CN106844569A (en) * 2017-01-03 2017-06-13 陕西甄宝网络有限公司 A kind of frame selects identifying system and method
WO2017177903A1 (en) * 2016-04-13 2017-10-19 芋头科技(杭州)有限公司 Online verification method and system for real-time gesture detection

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101216884A (en) * 2007-12-29 2008-07-09 北京中星微电子有限公司 A method and system for face authentication
US20130195363A1 (en) * 2009-09-14 2013-08-01 Trimble Navigation Limited Image-based georeferencing
CN103310466A (en) * 2013-06-28 2013-09-18 安科智慧城市技术(中国)有限公司 Single target tracking method and achievement device thereof
CN103324955A (en) * 2013-06-14 2013-09-25 浙江智尔信息技术有限公司 Pedestrian detection method based on video processing

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101216884A (en) * 2007-12-29 2008-07-09 北京中星微电子有限公司 A method and system for face authentication
US20130195363A1 (en) * 2009-09-14 2013-08-01 Trimble Navigation Limited Image-based georeferencing
CN103324955A (en) * 2013-06-14 2013-09-25 浙江智尔信息技术有限公司 Pedestrian detection method based on video processing
CN103310466A (en) * 2013-06-28 2013-09-18 安科智慧城市技术(中国)有限公司 Single target tracking method and achievement device thereof

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
孙艳: "基于内容图像检索与敏感图像过滤的若干算法研究", 《中国博士学位论文全文数据库 信息科技辑》 *
曹阳: "医学图像处理与检索技术的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *
赵艳丽: "图像特征提取与匹配算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105205436A (en) * 2014-06-03 2015-12-30 北京创思博德科技有限公司 Gesture identification system based on multiple forearm bioelectric sensors
CN105224066A (en) * 2014-06-03 2016-01-06 北京创思博德科技有限公司 A kind of gesture identification method based on high in the clouds process
CN105205436B (en) * 2014-06-03 2019-09-06 北京创思博德科技有限公司 A kind of gesture recognition system based on forearm bioelectricity multisensor
WO2017177903A1 (en) * 2016-04-13 2017-10-19 芋头科技(杭州)有限公司 Online verification method and system for real-time gesture detection
CN106844569A (en) * 2017-01-03 2017-06-13 陕西甄宝网络有限公司 A kind of frame selects identifying system and method

Also Published As

Publication number Publication date
CN103632143B (en) 2017-02-08

Similar Documents

Publication Publication Date Title
Zhang et al. Informed haar-like features improve pedestrian detection
Mäkinen et al. An experimental comparison of gender classification methods
Pham et al. Count forest: Co-voting uncertain number of targets using random forest for crowd density estimation
Long et al. Accurate object localization in remote sensing images based on convolutional neural networks
Pflug et al. Ear biometrics: a survey of detection, feature extraction and recognition methods
Zhou et al. Principal visual word discovery for automatic license plate detection
US8792682B2 (en) Method and system for identifying a license plate
Lempitsky et al. Learning to count objects in images
Li et al. Saliency detection via dense and sparse reconstruction
Kim et al. Spatiotemporal saliency detection and its applications in static and dynamic scenes
US8792722B2 (en) Hand gesture detection
Yun et al. An automatic hand gesture recognition system based on Viola-Jones method and SVMs
Ikizler et al. Recognizing actions from still images
Diao et al. Efficient saliency-based object detection in remote sensing images using deep belief networks
US20120027263A1 (en) Hand gesture detection
Levine et al. Saliency detection based on frequency and spatial domain analysis
Zhao et al. Sobel-lbp
Bai et al. A graph-based classification method for hyperspectral images
Gao et al. Multi-view discriminative and structured dictionary learning with group sparsity for human action recognition
Li et al. Fast and effective text detection
Yuan et al. Mid-level features and spatio-temporal context for activity recognition
Khurana et al. Techniques for object recognition in images and multi-object detection
Angelova et al. Image segmentation for large-scale subcategory flower recognition
US20160196350A1 (en) Image searching method and apparatus
CN103927387B (en) Image indexing system and its correlation technique and device

Legal Events

Date Code Title Description
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant