CN106384090A - AER sensor image display and target data interception method - Google Patents

AER sensor image display and target data interception method Download PDF

Info

Publication number
CN106384090A
CN106384090A CN201610813829.2A CN201610813829A CN106384090A CN 106384090 A CN106384090 A CN 106384090A CN 201610813829 A CN201610813829 A CN 201610813829A CN 106384090 A CN106384090 A CN 106384090A
Authority
CN
China
Prior art keywords
frame
event
aer
image
data
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
Application number
CN201610813829.2A
Other languages
Chinese (zh)
Inventor
姚素英
王含宇
高志远
徐江涛
高静
史再峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin University
Original Assignee
Tianjin University
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 Tianjin University filed Critical Tianjin University
Priority to CN201610813829.2A priority Critical patent/CN106384090A/en
Publication of CN106384090A publication Critical patent/CN106384090A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51—Motion estimation or motion compensation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51—Motion estimation or motion compensation
    • H04N19/513—Processing of motion vectors
    • H04N19/517—Processing of motion vectors by encoding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/14—Picture signal circuitry for video frequency region
    • H04N5/144—Movement detection
    • H04N5/145—Movement estimation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30232—Surveillance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/44—Event detection

Abstract

The invention relates to a technical field of image processing, for solving a problem that an image acquisition and processing method in the prior art and an AER sensor are incompatible, the invention aims at providing an image display and target data interception method for the AER sensor, displaying directly performed on a display device can be realized, and a target event can be quickly extracted. Therefore, a technical scheme applied by the invention provides the image display and target data interception method for the AER sensor, a data source is provided by the AER image sensor having an N*N pixel array, wherein the N is a natural number, when the data is generated, the data comprises event positive polarity ON and negative polarity OFF properties and also comprises time labels, the time labels only comprise time information and position coordinate information, and the time labels inputted in sequence are processed based on an event drive-based feature extraction algorithm. The method is mainly applied to image processing.

Description

The image of AER sensor shows and target data intercept method
Technical field
The present invention relates to technical field of image processing, more particularly, to a kind of back-end image number based on AER imageing sensor According to display packing and target data intercept method.
Background technology
Because the development of the information processing technology and the range of application of sensor constantly expand, novel sensor and its collection Data class gets more and more, and represents the motion of (address-time representation, AER) based on event address-time Object features extract researching value also just day by day show especially, it traffic monitoring, perimeter protection, virtual reality, man-machine interaction and The fields such as Mobile Robotics Navigation suffer from being widely applied.In intelligent monitor system, moving target is identified, thus Realize nobody and detect, monitor to be the hot issue of current research, be related to the multi-field core such as Image Acquisition, view synthesis Heart technology.
Traditional imaging and target data intercepting are that the data reading by frame is carried out the comparison on room and time and pressure Contracting, finds out the violent frame of moving target change, is subsequently identified work after being extracted.Data due to AER sensor Asynchronous continuously transmit, there is not the concept of frame it is impossible to directly be shown on the display device, and object event be not Can rapidly be extracted.
Content of the invention
For overcoming the deficiencies in the prior art, cannot be compatible with AER sensor for traditional images collection and processing method Problem, it is contemplated that propose a kind of image for AER sensor showing and target data intercept method, realizes directly aobvious Show and shown on equipment, and object event can rapidly be extracted.For this reason, the technical solution used in the present invention is, The image of AER sensor shows and target data intercept method, and data source is by the AER imageing sensor with N × N pel array There is provided, wherein N is natural number, except comprising event positive polarity ON, negative polarity OFF attribute when data produces, also subsidiary time tag, Time tag only comprises temporal information and location coordinate information, based on event driven feature extraction algorithm to sequentially input when Between label processed；
Comprise the concrete steps that, the data flow completing AER sensor first carries out event to the conversion of image, sets and uses S frame Speed per second carries out image and shows, array of display size is identical with AER sensor, is N × N, each second S frame, then the 1/S second it Interior event can merge into a two field picture；So according to the time tag in AER output data, flow of event is split, often Event within the 1/S second is divided into a frame in；Again that the data of this frame in is grey on the basis of a width N × N according to position coordinateses Angle value be 128 gray image on be overlapped；One positive event just increases P on the gray value of correspondence positionon, a negative thing Part just reduces P on the gray value of correspondence positionoff, PonAnd PoffBy the unit gray value being increased or decreased, positive and negative event increasing The gray value numerical value adding deduct few is identical, and an intraframe data is all superimposed after gray level image, using this piece image as a frame Image plays out, and starts simultaneously at next frame conversion work；According to above method in real time the AER sensor of N × N is carried out with S frame frame per second per second shows.
Additionally, completing the extraction step of effective exercise target data while being additionally included in image display：Conducting frame figure After dividing, count event number T in this two field picture, and record M event number T of M frame in0～T(M-1), wherein M is Odd number, when the frame event number of and if only if (M+1)/2 is the maximum in M event number, by the frame number of this event number maximum According to extracting；This frame event number maximum means that motion change is the most violent, then just this frame number is presently considered to be target data Frame, this frame data is used for follow-up data processing and target recognition.
The feature of the present invention and beneficial effect are：
Image proposed by the present invention shows and target intercept method, due to using frame forming tech it is thus possible to right in real time The data flow of AER sensor is shown, and intercepts effective target data.
Brief description：
Fig. 1 image shows and target data intercepts flow chart.
Specific embodiment
The data source of the present invention is provided by the AER imageing sensor with N × N pel array, and wherein N is natural number, number According to when producing except comprising event positive-negative polarity (ON/OFF) attribute, also subsidiary time tag, time tag only comprises temporal information And location coordinate information, based on event driven feature extraction algorithm, the time tag sequentially inputting is processed.
Carry out event firstly the need of the data flow completing AER sensor to the conversion of image.Set the speed per second using S frame Degree carries out image and shows, array of display size is identical with AER sensor, is N × N.Each second S frame, then the event within the 1/S second A two field picture can be merged into.So according to the time tag in AER output data, flow of event is split, every 1/S second it Interior event is divided into a frame in.Again the data of this frame according to position coordinateses in the base grey value of a width N × N is It is overlapped on 128 gray image.One positive event just increases P on the gray value of correspondence positionon, a negative event just exists P is reduced on the gray value of correspondence positionoff, PonAnd PoffBy increasedd unit gray value, generally positive and negative event increase The gray value numerical value of (or minimizing) is identical.One intraframe data is all superimposed after gray level image, using this piece image as a frame Image plays out, and starts simultaneously at next frame conversion work.In real time the AER sensor of N × N can be entered according to above method Row is shown with S frame frame per second per second.
The extraction of effective exercise target data can also be completed while image shows.Conducting frame image division it Afterwards, count event number T in this two field picture, and record M event number T of M frame in0～T(M-1), wherein M is odd number.When And if only if when the frame event number of (M+1)/2 is the maximum in M event number, the maximum frame data of this event number are extracted Come.This frame event number maximum means that motion change is the most violent, then just this frame number can be presently considered to be target data frame.This Frame data can be used for follow-up data processing and target recognition.
Display data intercept method proposed by the invention captures the picture of light intensity change first with AER imageing sensor Vegetarian refreshments, pel array selects 40 × 40 resolution, and the present invention is shown according to 60 frames speed per second, then by 16.66ms Interior event data is divided in a frame, and carries out event number statistics to the data within 9 frames, when the 5th frame event number When maximum, using the data output of the 5th frame as effective target data, for post processing.

Claims (3)

1. a kind of image of AER sensor shows and target data intercept method, it is characterized in that, data source is by having N × N pixel The AER imageing sensor of array provides, and wherein N is natural number, removes and comprise event positive polarity ON, negative polarity OFF when data produces Attribute, also subsidiary time tag, time tag is only comprised temporal information and location coordinate information, is carried based on event driven feature Take algorithm that the time tag sequentially inputting is processed.
2. the image of AER sensor as claimed in claim 1 shows and target data intercept method, it is characterized in that, concrete step Suddenly it is that the data flow completing AER sensor first carries out event to the conversion of image, sets and is carried out using S frame speed per second Image shows, array of display size is identical with AER sensor, is N × N, each second S frame, then the event within the 1/S second can close And be a two field picture；So according to the time tag in AER output data, flow of event is split, the thing within every 1/S second Part is divided into a frame in；Again by the data of this frame according to position coordinateses a width N × N base grey value be 128 ash It is overlapped on color image；One positive event just increases P on the gray value of correspondence positionon, a negative event is just in corresponding position P is reduced on the gray value putoff, PonAnd PoffBy the unit gray value being increased or decreased, the ash that positive and negative event increases or decreases Angle value numerical value is identical, and an intraframe data is all superimposed after gray level image, and this piece image is played out as a two field picture, Start simultaneously at next frame conversion work；In real time the AER sensor of N × N is carried out with S frame frame per second per second according to above method Display.
3. the image of AER sensor as claimed in claim 1 shows and target data intercept method, it is characterized in that, also includes The extraction step of effective exercise target data is completed while image shows：After conducting frame image division, count this frame Event number T in image, and record M event number T of M frame in0～T(M-1), wherein M is odd number, and if only if (M+1)/ When 2 frame event number is the maximum in M event number, the maximum frame data of this event number are extracted；This frame event Number maximum means that motion change is the most violent, then just this frame number is presently considered to be target data frame, and this frame data is used for follow-up Data processing and target recognition.
CN201610813829.2A 2016-09-10 2016-09-10 AER sensor image display and target data interception method Pending CN106384090A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610813829.2A CN106384090A (en) 2016-09-10 2016-09-10 AER sensor image display and target data interception method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610813829.2A CN106384090A (en) 2016-09-10 2016-09-10 AER sensor image display and target data interception method

Publications (1)

Publication Number Publication Date
CN106384090A true CN106384090A (en) 2017-02-08

Family

ID=57935716

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610813829.2A Pending CN106384090A (en) 2016-09-10 2016-09-10 AER sensor image display and target data interception method

Country Status (1)

Country Link
CN (1) CN106384090A (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101340521A (en) * 2007-07-03 2009-01-07 佳能株式会社 Image display control apparatus and method
CN103139495A (en) * 2013-02-27 2013-06-05 天津大学 Asynchronous pixel array with arbitration time error correction function
CN103595931A (en) * 2013-11-05 2014-02-19 天津大学 CMOS asynchronous time domain image sensor capable of achieving real-time time stamp
CN104143096A (en) * 2014-07-22 2014-11-12 天津大学 Binary connected domain labeling method based on address event representation
CN104284158A (en) * 2014-10-23 2015-01-14 南京信必达智能技术有限公司 Event-oriented intelligent camera monitoring method
CN104808057A (en) * 2015-05-04 2015-07-29 中国电子科技集团公司第二十六研究所 Acousto-optic real-time signal analyzer based on asynchronous detection
CN105469039A (en) * 2015-11-19 2016-04-06 天津大学 Target identification system based on AER image sensor
CN105721772A (en) * 2016-01-20 2016-06-29 天津师范大学 Asynchronous time domain visual information imaging method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101340521A (en) * 2007-07-03 2009-01-07 佳能株式会社 Image display control apparatus and method
CN103139495A (en) * 2013-02-27 2013-06-05 天津大学 Asynchronous pixel array with arbitration time error correction function
CN103595931A (en) * 2013-11-05 2014-02-19 天津大学 CMOS asynchronous time domain image sensor capable of achieving real-time time stamp
CN104143096A (en) * 2014-07-22 2014-11-12 天津大学 Binary connected domain labeling method based on address event representation
CN104284158A (en) * 2014-10-23 2015-01-14 南京信必达智能技术有限公司 Event-oriented intelligent camera monitoring method
CN104808057A (en) * 2015-05-04 2015-07-29 中国电子科技集团公司第二十六研究所 Acousto-optic real-time signal analyzer based on asynchronous detection
CN105469039A (en) * 2015-11-19 2016-04-06 天津大学 Target identification system based on AER image sensor
CN105721772A (en) * 2016-01-20 2016-06-29 天津师范大学 Asynchronous time domain visual information imaging method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
BRANDLI C等: ""A 240×180 130dB 3μs Latency Global Shutter Spatiotemporal Vision Sensor"", 《IEEE JOURNAL OF SOLID-STATE CIRCUITS》 *
LENERO-BARDALLO J A等: ""A 3.6μs Latency Asynchronous Frame-free Event-driven Dynamic-vision-sensor"", 《IEEE JOURNAL OF SOLID-STATE CIRCUITS》 *
闫石等: ""基于地址-事件表示的高速二值连通域标记方法"", 《传感技术学报》 *

Similar Documents

Publication Publication Date Title
Tedaldi et al. Feature detection and tracking with the dynamic and active-pixel vision sensor (DAVIS)
US9948911B2 (en) Method and apparatus for efficient depth image transformation
CN103942843A (en) Fairway and ship three-dimensional model dynamic presenting method based on video
CN108765333B (en) Depth map perfecting method based on depth convolution neural network
CN111291674B (en) Method, system, device and medium for extracting expression actions of virtual figures
CN104821010A (en) Binocular-vision-based real-time extraction method and system for three-dimensional hand information
JP2014116716A (en) Tracking device
CN107239760B (en) Video data processing method and system
CN102509255B (en) High-speed image acquiring and processing method and device
CN108370454A (en) System and method for video processing
CN104539890A (en) Target tracking method and system
CN110096144B (en) Interactive holographic projection method and system based on three-dimensional reconstruction
CN103336578A (en) Novel motion induction interactive advertising device
CN105335451A (en) Processing method and apparatus for display data in finder frame, shooting method and terminal
CN112330753B (en) Target detection method of augmented reality system
JP2017162141A (en) Image identification system, control device therefor, and image identification method
CN106384090A (en) AER sensor image display and target data interception method
US20140055644A1 (en) Apparatus and method for extracting object
CN204131656U (en) Be applied to the assistant images processing unit of augmented reality system
Das Biswas et al. HALSIE: Hybrid Approach to Learning Segmentation by Simultaneously Exploiting Image and Event Modalities
CN105511602A (en) 3d virtual roaming system
CN202110488U (en) Gesture control system based on computer vision
CN111754543B (en) Image processing method, device and system
CN108058170A (en) A kind of vision robot's data acquisition processing system
KR101350068B1 (en) Electronic device for outputting region of interest image

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170208

WD01 Invention patent application deemed withdrawn after publication