CN107506773A - A kind of feature extracting method, apparatus and system - Google Patents
A kind of feature extracting method, apparatus and system Download PDFInfo
- Publication number
- CN107506773A CN107506773A CN201710875722.5A CN201710875722A CN107506773A CN 107506773 A CN107506773 A CN 107506773A CN 201710875722 A CN201710875722 A CN 201710875722A CN 107506773 A CN107506773 A CN 107506773A
- Authority
- CN
- China
- Prior art keywords
- result
- feature
- pending
- lbp
- feature extraction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/467—Encoded features or binary features, e.g. local binary patterns [LBP]
Abstract
The invention discloses a kind of method of feature extraction, it should will be preserved by the pending view data that processor is handled into the preset buffer memory at FPGA ends, and the LBP algorithms for starting FPGA ends are controlled to handle pending view data by processor end, processing completes preprocessor end and obtains result, obtains the result of final feature extraction.As can be seen here, processor resource is consumed when largely transferring to FPGA development boards to complete the processing to view data in this programme, therefore feature extraction can be reduced, lifting LBP algorithms carry out the efficiency of feature extraction.The embodiment of the present invention also provides a kind of device of feature extraction, a kind of system of feature extraction, can equally realize above-mentioned technique effect.
Description
Technical field
The present invention relates to data processing field, more specifically to a kind of feature extracting method, apparatus and system.
Background technology
LBP is a kind of operator for being used for describing image texture characteristic, and the operator is by T.Ojala of Oulu, Finland university et al.
Itd is proposed in 1996.LBP core concept is using the gray value of center pixel as threshold value, and phase is obtained compared with his field
Corresponding binary code represents Local textural feature.
Due to exploding for information age data volume, when having had a strong impact on the response of user's request to the speed of data processing
Between.Especially for the application under cloud computing, the quick response of mass data processing needs to take a large amount of computing resources, has a strong impact on
The performance and power consumption of cloud computing data processing centre.At present, realize and apply in a large amount of occasions in recognition of face, based on LBP
The advantages of recognition of face of method is notable is to illumination-insensitive, but still not solving the problems, such as posture and expression.But
Compared to other method, LBP discrimination has been greatly improved.Existing many researchs and technology are to LBP recognitions of face
Algorithm be modified to further improving performance.However, the acceleration to the technology is improved in CPU/GPU hardware environment
Carry out, the parallel processing of data can not be realized, therefore performance boost space is limited.
Therefore, processor resource is consumed when how to reduce feature extraction, is that those skilled in the art need what is solved to ask
Topic.
The content of the invention
It is an object of the invention to provide a kind of feature extracting method, apparatus and system, to place during reducing feature extraction
Manage device resource consumption.
To achieve the above object, the embodiments of the invention provide following technical scheme:
A kind of method of feature extraction, including:
The pending view data sent by processor end is received, and the pending view data is deposited into DDR
Preset buffer memory;
To the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and handled
As a result;The feature extraction result of pending view data is obtained so that the processor end obtains the result.
Wherein, it is described to the pending image data extraction feature and to carry out LBP characteristic matchings using circular LBP operators
Result is obtained, including:
The pending view data is read from the DDR to caching on the piece at the FPGA ends;
To the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and handled
As a result;
The result is write back into the DDR.
Wherein, it is described to the pending image data extraction feature and to carry out LBP characteristic matchings using circular LBP operators
Result is obtained, including:
Using the hardware circuit in the FPGA obtained after LBP algorithms are compiled beforehand through instrument to described pending
Image data extraction feature simultaneously carries out LBP characteristic matchings and obtains result;Wherein described LBP algorithms are described using OpenCL
LBP algorithms.
A kind of device of feature extraction, including:
View data receiving module, for receiving the pending view data sent by processor end, and wait to locate by described
Reason view data deposits the preset buffer memory into DDR;
First processing module, for the pending image data extraction feature and carrying out LBP using circular LBP operators
Characteristic matching obtains result;The feature of pending view data is obtained so that the processor end obtains the result
Extract result.
Wherein, the first processing module, including:
Reading unit, for reading the pending view data from the DDR to slow on the piece at the FPGA ends
Deposit;
Processing unit, for the pending image data extraction feature and carrying out LBP features using circular LBP operators
Matching obtains result;
Writeback unit, for the result to be write back into the DDR.
A kind of device of feature extraction, including:
First memory, for storing computer program;
First processor, a kind of spy as described in any one of claims 1 to 3 is realized during for performing the computer program
The step of levying the method for extraction.
A kind of method of feature extraction, including:
The preset buffer memory pending view data being transferred in the DDR at FPGA ends;So that the FPGA ends utilize circle
LBP operators are to the pending image data extraction feature and carry out LBP characteristic matchings and obtain result;
The result at the FPGA ends is read, feature extraction result is obtained using the result.
A kind of device of feature extraction, including:
Second memory, for storing computer program;
Second processor, a kind of feature extraction as claimed in claim 7 is realized during for performing the computer program
The step of method.
A kind of device of feature extraction, including:
Transport module, for the preset buffer memory being transferred to pending view data in the DDR at FPGA ends;It is so that described
FPGA ends using circular LBP operators to the pending image data extraction feature and carry out LBP characteristic matchings obtain processing knot
Fruit;
Second processing module, for reading the result at the FPGA ends, obtain feature using the result and carry
Take result.
A kind of system of feature extraction, including:
Processor and the FPGA development boards being connected with the processor;
Wherein, the processor is used for the preset buffer memory being transferred to pending view data in the DDR of FPGA ends, and obtains
The result at FPGA ends is to obtain the feature extraction result of the pending data;
The FPGA development boards are used to utilize circular LBP after receiving the pending view data sent by the processor
Operator is to the pending image data extraction feature and carries out LBP characteristic matchings and obtains result.
By above scheme, a kind of method of feature extraction provided in an embodiment of the present invention, receive by processor end
The pending view data sent, and the pending view data is deposited into the preset buffer memory into DDR;Calculated using circular LBP
Son is to the pending image data extraction feature and carries out LBP characteristic matchings and obtains result;So as to the processor end
Obtain the result and obtain the feature extraction result of pending view data.
As can be seen here, the method for a kind of feature extraction provided in an embodiment of the present invention, should will be treated by what processor was handled
Processing view data is preserved into the preset buffer memory at FPGA ends, and controls the LBP algorithms for starting FPGA ends to treat by processor end
The view data of processing is handled, and processing completes preprocessor end and obtains result, obtains the knot of final feature extraction
Fruit.As can be seen here, largely transfer to FPGA development boards to complete the processing to view data in this programme, therefore can drop
Processor resource is consumed during low feature extraction, lifting LBP algorithms carry out the efficiency of feature extraction.The embodiment of the present invention also provides
A kind of device of feature extraction, a kind of system of feature extraction, can equally realize above-mentioned technique effect.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is a kind of method flow diagram of feature extraction disclosed in the embodiment of the present invention;
Fig. 2 is a kind of apparatus structure schematic diagram of feature extraction disclosed in the embodiment of the present invention;
Fig. 3 is a kind of apparatus structure schematic diagram of feature extraction disclosed in the embodiment of the present invention;
Fig. 4 is a kind of method flow diagram of feature extraction disclosed in the embodiment of the present invention;
Fig. 5 is a kind of apparatus structure schematic diagram of feature extraction disclosed in the embodiment of the present invention;
Fig. 6 is a kind of apparatus structure schematic diagram of feature extraction disclosed in the embodiment of the present invention;
Fig. 7 is a kind of system structure diagram of feature extraction disclosed in the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
The embodiment of the invention discloses a kind of method, apparatus of feature extraction and system, to place during reducing feature extraction
Manage device resource consumption.
Referring to Fig. 1, a kind of method of feature extraction provided in an embodiment of the present invention, specifically include:
S101, the pending view data sent by processor end is received, and the pending view data is deposited to DDR
In preset buffer memory;
Specifically, FPGA ends receive the pending view data sent by processor, wherein pending view data stores
Preset buffer memory in the DDR internal memories of FPGA development boards.
It should be noted that preset buffer memory can be created by processor according to the size for just handling view data at FPGA ends
Build.
S102, to the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and obtain
Result;The feature extraction result of pending view data is obtained so that the processor end obtains the result.
Specifically, at FPGA ends, the extraction of feature, Ran Houjin are carried out to pending view data using circular LBP operators
Row LBP feature matching operations, finally give result.It can send what processing was completed to processor end after obtaining result
Signal, so that processor end timely gets result.
It should be noted that the circular LBP operators at FPGA ends and the algorithm of characteristic matching, are to pre-deposit FPGA ends
's.
As can be seen here, the method for a kind of feature extraction provided in an embodiment of the present invention, should will be treated by what processor was handled
Processing view data is preserved into the preset buffer memory at FPGA ends, and controls the LBP algorithms for starting FPGA ends to treat by processor end
The view data of processing is handled, and processing completes preprocessor end and obtains result, obtains the knot of final feature extraction
Fruit.As can be seen here, largely transfer to FPGA development boards to complete the processing to view data in this programme, therefore can drop
Processor resource is consumed during low feature extraction, lifting LBP algorithms carry out the efficiency of feature extraction.
The embodiment of the present invention provides a kind of feature extracting method specifically based on LBP algorithms, based on above-described embodiment, sheet
Inventive embodiments have done further restriction and explanation to the S102 of above-described embodiment, and other step contents and above-described embodiment are big
Cause identical, specifically may be referred to part corresponding to above-described embodiment, here is omitted.Specifically, S102 includes:
The pending view data is read from the DDR to caching on the piece at the FPGA ends;
To the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and handled
As a result;
The result is write back into the DDR.
In particular it is required that when handling pending view data, FPGA development boards first are by the pending image in DDR internal memories
Data are read cached on piece in FPGA in, it is necessary to which explanation, the operation of reading can be that batch is carried out, i.e., by pending figure
As data are read to caching on the piece of FPGA development boards in bulk from the DDR of FPGA development boards;In being cached on piece, pass through number
Handled according to independent mode parallelization, the extraction of feature carried out using circular LBP operators, then using Dynamic data exchange mode simultaneously
Rowization processing, carry out LBP characteristic matchings and obtain result, result is written back in FPGA DDR.
The embodiment of the present invention provides a kind of feature extracting method specifically based on LBP algorithms, based on above-described embodiment, sheet
Inventive embodiments to the pending image data extraction feature and are carried out to the described of above-described embodiment using circular LBP operators
LBP characteristic matchings obtain result, have done further restriction and explanation, other step contents and above-described embodiment substantially phase
Together, part corresponding to above-described embodiment is specifically may be referred to, here is omitted.Specifically include:
Using the hardware circuit in the FPGA obtained after LBP algorithms are compiled beforehand through instrument to described pending
Image data extraction feature simultaneously carries out LBP characteristic matchings and obtains result;Wherein described LBP algorithms are described using OpenCL
LBP algorithms.
Specifically, LBP algorithms can be described using OpenCL language in advance, generates run on a processor respectively
Host side program and the kernel programs run on FPGA development boards.Then GCC compilers are respectively adopted to host side
Program is compiled, and the executable file that can be performed on a processor is generated, using Altera SDK for OpenCL
(AOC) High Level Synthesis instrument is compiled synthesis to Kernel program files, generates the AOCX files that can be run on FPGA,
The final hardware circuit for being mapped as FPGA development boards.Then processor and FPGA ends cooperation is made to realize to pending figure
As the feature extraction of data.
It should be noted that OpenCL language carries out the generation of the hardware bit stream in FPGA development boards, can effectively improve
The efficiency that algorithm is realized, reduce the construction cycle that algorithm is realized.
A kind of device of feature extraction provided in an embodiment of the present invention is introduced below, a kind of feature described below
The device of extraction, can be with cross-referenced with a kind of above-described method of feature extraction.
Reference picture 2, a kind of device of feature extraction provided in an embodiment of the present invention, including:
View data receiving module 201, for receiving the pending view data sent by processor end, and treated described
Processing view data deposits the preset buffer memory into DDR;
Specifically, view data receiving module 201 receives the pending view data sent by processor, wherein pending
View data is stored in the preset buffer memory in the DDR internal memories of FPGA development boards.
It should be noted that preset buffer memory can be created by processor according to the size for just handling view data at FPGA ends
Build.
First processing module 202, for the pending image data extraction feature and being carried out using circular LBP operators
LBP characteristic matchings obtain result;Pending view data is obtained so that the processor end obtains the result
Feature extraction result.
Specifically, at FPGA ends, first processing module 201 is carried out special using circular LBP operators to pending view data
The extraction of sign, LBP feature matching operations are then carried out, finally give result.Obtaining can be to processor after result
End sends the signal that processing is completed, so that processor end timely gets result.
It should be noted that the circular LBP operators at FPGA ends and the algorithm of characteristic matching, are to pre-deposit FPGA ends
's.
As can be seen here, the method for a kind of feature extraction provided in an embodiment of the present invention, view data receiving module 201 incite somebody to action this
It should be preserved by the pending view data that processor is handled into the preset buffer memory at FPGA ends, first processing module 202 utilizes LBP
Algorithm is handled pending view data, and after the completion of processing, processor obtains result to FPGA ends again, obtains most
The result of whole feature extraction.As can be seen here, FPGA is largely transferred to complete the processing to view data in this programme, therefore
Processor resource is consumed when can reduce feature extraction, lifting LBP algorithms carry out the efficiency of feature extraction.
The embodiment of the present invention provides a kind of feature deriving means specifically based on LBP algorithms, based on above-described embodiment, sheet
Inventive embodiments have done further restriction and explanation to the first processing module 202 of above-described embodiment, other module contents with
Above-described embodiment is roughly the same, specifically may be referred to part corresponding to above-described embodiment, here is omitted.Specifically, first
Processing module 202 includes:
Reading unit, for reading the pending view data from the DDR to slow on the piece at the FPGA ends
Deposit;
Processing unit, for the pending image data extraction feature and carrying out LBP features using circular LBP operators
Matching obtains result;
Writeback unit, for the result to be write back into the DDR.
In particular it is required that when handling pending view data, reading unit FPGA development boards first will be treated in DDR internal memories
Processing view data is read cached on piece in FPGA in, it is necessary to which explanation, the operation of reading can be that batch is carried out, will
Pending view data is read to caching on the piece of FPGA development boards in bulk from the DDR of FPGA development boards;Cached on piece
In, processing unit parallelization by way of Dynamic data exchange is handled, and the extraction of feature, Ran Houli are carried out using circular LBP operators
Handled with the mode parallelization of Dynamic data exchange, carry out LBP characteristic matchings and obtain result, writeback unit writes result again
Return in FPGA DDR.
A kind of device of feature extraction provided in an embodiment of the present invention is introduced below, a kind of feature described below
The device of extraction, can mutually it join with a kind of above-described method of feature extraction.
A kind of reference picture 3, device of feature extraction provided in an embodiment of the present invention, is specifically included:
First memory 301, for storing computer program;
First processor 302, realize that a kind of feature as described in above-mentioned embodiment carries during for performing the computer program
The step of method taken.
The embodiment of the present invention provides a kind of method of feature extraction, and a kind of method described below is entered based on processor end
OK, the method, apparatus with a kind of above-mentioned feature extraction based on FPGA ends can be with cross-referenced.
A kind of reference picture 4, method of feature extraction provided in an embodiment of the present invention, is specifically included:
S401, the preset buffer memory pending view data being transferred in the DDR at FPGA ends;So that the FPGA ends utilize
Circular LBP operators are to the pending image data extraction feature and carry out LBP characteristic matchings and obtain result;
Specifically, pending view data is transferred to preset buffer memory in FPGA ends on DDR internal memories by processor, by FPGA
End carries out feature extraction processing to pending view data.
It should be noted that preset buffer memory can be created by processor according to the size of view data at FPGA ends.
S402, reads the result at the FPGA ends, and feature extraction result is obtained using the result.
Specifically, after the completion of being handled at FPGA ends, to FPGA ends reading process result, processor obtains final feature and carried
The result taken.
The embodiment of the present invention provides a kind of device of feature extraction, a kind of device of feature extraction described below with above
A kind of method of feature extraction of description can be with cross-referenced.
With reference to figure 5, a kind of device of feature extraction provided in an embodiment of the present invention, specifically include:
Transport module 501, for the preset buffer memory being transferred to pending view data in the DDR at FPGA ends;So that institute
State FPGA ends and to the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and handled
As a result;
Specifically, pending view data is transferred to preset buffer memory in FPGA ends on DDR internal memories by transport module 501,
Feature extraction processing is carried out to pending view data by FPGA ends.
It should be noted that preset buffer memory can be created by processor according to the size of view data at FPGA ends.
Second processing module 502, for reading the result at the FPGA ends, feature is obtained using the result
Extract result.
Specifically, after the completion of being handled at FPGA ends, Second processing module 502 is to FPGA ends reading process result, processor
Obtain the result of final feature extraction.
The embodiment of the present invention provides a kind of device of feature extraction, a kind of device of feature extraction described below with above
A kind of method of feature extraction of description can be with cross-referenced.
With reference to figure 6, a kind of device of feature extraction provided in an embodiment of the present invention, including:
Second memory 601, for storing computer program;
Second processor 602, realize that a kind of feature as described in above-mentioned embodiment carries during for performing the computer program
The step of method taken.
With reference to figure 7, the embodiment of the present invention provides a kind of system of feature extraction, specifically includes:
Processor 701 and the FPGA development boards 702 being connected with the processor;
Wherein, the processor processor 701 is default slow in the DDR of FPGA ends for pending view data to be transferred to
Deposit, and obtain the result at FPGA ends to obtain the feature extraction result of the pending data;
The FPGA development boards 702 are used to after receiving the pending view data sent by the processor 701 utilize
Circular LBP operators are to the pending image data extraction feature and carry out LBP characteristic matchings and obtain result.
Specifically, processor 701 sends pending view data to FPGA ends, by FPGA development features extraction processs,
Processing obtains result after terminating, and obtains final feature extraction result.
Specifically, FPGA development boards 702 carry out feature extraction, characteristic matching after pending view data is received to it
Processing.
It should be noted that the processor 701 can be connected with the FPGA development boards 702 by PCI-E, wherein handling
Device can be CPU, the method for prestored in FPGA circular LBP operators and characteristic matching, co-operated and realized by CPU and FPGA
The acceleration of feature extraction operation.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be and other
The difference of embodiment, between each embodiment identical similar portion mutually referring to.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (10)
- A kind of 1. method of feature extraction, it is characterised in that including:The pending view data sent by processor end is received, and the pending view data is deposited into presetting into DDR Caching;To the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and obtain result; The feature extraction result of pending view data is obtained so that the processor end obtains the result.
- 2. according to the method for claim 1, it is characterised in that described to utilize circular LBP operators to the pending image Data, which extract feature and carry out LBP characteristic matchings, obtains result, including:The pending view data is read from the DDR to caching on the piece at the FPGA ends;To the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and obtain result;The result is write back into the DDR.
- 3. method according to claim 1 or 2, it is characterised in that described to utilize circular LBP operators to the pending figure Extract feature as data and carry out LBP characteristic matchings and obtain result, including:Using the hardware circuit in the FPGA obtained after LBP algorithms are compiled beforehand through instrument to the pending image Data, which extract feature and carry out LBP characteristic matchings, obtains result;Wherein described LBP algorithms are described using OpenCL LBP algorithms.
- A kind of 4. device of feature extraction, it is characterised in that including:View data receiving module, for receiving the pending view data that is sent by processor end, and by the pending figure The preset buffer memory into DDR is deposited as data;First processing module, for the pending image data extraction feature and carrying out LBP features using circular LBP operators Matching obtains result;The feature extraction of pending view data is obtained so that the processor end obtains the result As a result.
- 5. device according to claim 4, it is characterised in that the first processing module, including:Reading unit, for reading the pending view data from the DDR to caching on the piece at the FPGA ends;Processing unit, for the pending image data extraction feature and carrying out LBP characteristic matchings using circular LBP operators Obtain result;Writeback unit, for the result to be write back into the DDR.
- A kind of 6. device of feature extraction, it is characterised in that including:First memory, for storing computer program;First processor, realize that a kind of feature as described in any one of claims 1 to 3 carries during for performing the computer program The step of method taken.
- A kind of 7. method of feature extraction, it is characterised in that including:The preset buffer memory pending view data being transferred in the DDR at FPGA ends;So that the FPGA ends are calculated using circular LBP Son is to the pending image data extraction feature and carries out LBP characteristic matchings and obtains result;The result at the FPGA ends is read, feature extraction result is obtained using the result.
- A kind of 8. device of feature extraction, it is characterised in that including:Second memory, for storing computer program;Second processor, a kind of method of feature extraction as claimed in claim 7 is realized during for performing the computer program The step of.
- A kind of 9. device of feature extraction, it is characterised in that including:Transport module, for the preset buffer memory being transferred to pending view data in the DDR at FPGA ends;So that the FPGA ends To the pending image data extraction feature and carry out LBP characteristic matchings using circular LBP operators and obtain result;Second processing module, for reading the result at the FPGA ends, feature extraction knot is obtained using the result Fruit.
- A kind of 10. system of feature extraction, it is characterised in that including:Processor and the FPGA development boards being connected with the processor;Wherein, the processor is used for the preset buffer memory being transferred to pending view data in the DDR of FPGA ends, and obtains FPGA The result at end is to obtain the feature extraction result of the pending data;The FPGA development boards are used to utilize circular LBP operators after receiving the pending view data sent by the processor To the pending image data extraction feature and carry out LBP characteristic matchings and obtain result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710875722.5A CN107506773A (en) | 2017-09-25 | 2017-09-25 | A kind of feature extracting method, apparatus and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710875722.5A CN107506773A (en) | 2017-09-25 | 2017-09-25 | A kind of feature extracting method, apparatus and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107506773A true CN107506773A (en) | 2017-12-22 |
Family
ID=60698790
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710875722.5A Pending CN107506773A (en) | 2017-09-25 | 2017-09-25 | A kind of feature extracting method, apparatus and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107506773A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108492243A (en) * | 2018-04-13 | 2018-09-04 | 福州新迪微电子有限公司 | It is a kind of based on block processing picture orbiting facility, system and method |
CN110689475A (en) * | 2019-09-10 | 2020-01-14 | 浪潮电子信息产业股份有限公司 | Image data processing method, system, electronic equipment and storage medium |
CN111110273A (en) * | 2019-12-31 | 2020-05-08 | 无锡祥生医疗科技股份有限公司 | Ultrasonic image processing method and device and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080148227A1 (en) * | 2002-05-17 | 2008-06-19 | Mccubbrey David L | Method of partitioning an algorithm between hardware and software |
CN101727314A (en) * | 2009-11-24 | 2010-06-09 | 成都市华为赛门铁克科技有限公司 | Data processing method and processor |
CN203338372U (en) * | 2013-04-09 | 2013-12-11 | 杭州富光科技有限公司 | FPGA-based system for real-time classification of plant quarantine larvae |
CN104320615A (en) * | 2014-10-17 | 2015-01-28 | 智擎信息系统(上海)有限公司 | Intelligent video security and protection system and signal processing method thereof |
CN105205476A (en) * | 2015-10-21 | 2015-12-30 | 上海红神信息技术有限公司 | Face recognition hardware framework based on LBP characteristics |
CN106127672A (en) * | 2016-06-21 | 2016-11-16 | 南京信息工程大学 | Image texture characteristic extraction algorithm based on FPGA |
-
2017
- 2017-09-25 CN CN201710875722.5A patent/CN107506773A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080148227A1 (en) * | 2002-05-17 | 2008-06-19 | Mccubbrey David L | Method of partitioning an algorithm between hardware and software |
CN101727314A (en) * | 2009-11-24 | 2010-06-09 | 成都市华为赛门铁克科技有限公司 | Data processing method and processor |
CN203338372U (en) * | 2013-04-09 | 2013-12-11 | 杭州富光科技有限公司 | FPGA-based system for real-time classification of plant quarantine larvae |
CN104320615A (en) * | 2014-10-17 | 2015-01-28 | 智擎信息系统(上海)有限公司 | Intelligent video security and protection system and signal processing method thereof |
CN105205476A (en) * | 2015-10-21 | 2015-12-30 | 上海红神信息技术有限公司 | Face recognition hardware framework based on LBP characteristics |
CN106127672A (en) * | 2016-06-21 | 2016-11-16 | 南京信息工程大学 | Image texture characteristic extraction algorithm based on FPGA |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108492243A (en) * | 2018-04-13 | 2018-09-04 | 福州新迪微电子有限公司 | It is a kind of based on block processing picture orbiting facility, system and method |
CN110689475A (en) * | 2019-09-10 | 2020-01-14 | 浪潮电子信息产业股份有限公司 | Image data processing method, system, electronic equipment and storage medium |
CN111110273A (en) * | 2019-12-31 | 2020-05-08 | 无锡祥生医疗科技股份有限公司 | Ultrasonic image processing method and device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110582785B (en) | Power efficient deep neural network module configured for executing layer descriptor lists | |
US20230334294A1 (en) | Multi-memory on-chip computational network | |
US9235769B2 (en) | Parallel object detection method for heterogeneous multithreaded microarchitectures | |
JP2019036298A (en) | Intelligent high bandwidth memory system and logic dies therefor | |
CN104915322A (en) | Method for accelerating convolution neutral network hardware and AXI bus IP core thereof | |
CN102667752A (en) | Memory having internal processors and data communication methods in memory | |
US11016773B2 (en) | Processor trace extensions to facilitate real-time security monitoring | |
CN104115093A (en) | Method, apparatus, and system for energy efficiency and energy conservation including power and performance balancing between multiple processing elements | |
CN107506773A (en) | A kind of feature extracting method, apparatus and system | |
CN107766079A (en) | Processor and the method for execute instruction on a processor | |
CN114003392B (en) | Data accelerated computing method and related device | |
CN111465943A (en) | On-chip computing network | |
EP3836031A2 (en) | Neural network processor, chip and electronic device | |
CN205608814U (en) | Augmented reality system based on zynq software and hardware concurrent processing | |
WO2021115208A1 (en) | Neural network processor, chip and electronic device | |
CN110100274A (en) | With the mixed reality system for reducing power rendering | |
Mandal et al. | An Embedded Vision Engine (EVE) for automotive vision processing | |
CN114661353A (en) | Data handling device and processor supporting multithreading | |
CN112306951B (en) | CNN-SVM resource efficient acceleration architecture based on FPGA | |
WO2021115149A1 (en) | Neural network processor, chip and electronic device | |
CN115994115B (en) | Chip control method, chip set and electronic equipment | |
CN103838694A (en) | FPGA high-speed USB interface data reading method | |
CN110648356A (en) | Multi-target tracking processing optimization method based on visual digital signal processing | |
CN113806250B (en) | Method for coordinating general processor core and vector component, interface and processor | |
Momeni et al. | Exploring the efficiency of the opencl pipe semantic on an FPGA |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171222 |
|
RJ01 | Rejection of invention patent application after publication |