CN1264338C - Method of image stability improvement for video camera and assistor thereof - Google Patents

Method of image stability improvement for video camera and assistor thereof Download PDF

Info

Publication number
CN1264338C
CN1264338C CN 200410014862 CN200410014862A CN1264338C CN 1264338 C CN1264338 C CN 1264338C CN 200410014862 CN200410014862 CN 200410014862 CN 200410014862 A CN200410014862 A CN 200410014862A CN 1264338 C CN1264338 C CN 1264338C
Authority
CN
China
Prior art keywords
circuit
video camera
angular speed
image
camera
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.)
Expired - Fee Related
Application number
CN 200410014862
Other languages
Chinese (zh)
Other versions
CN1571475A (en
Inventor
李宏生
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.)
Southeast University
Original Assignee
Southeast 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 Southeast University filed Critical Southeast University
Priority to CN 200410014862 priority Critical patent/CN1264338C/en
Publication of CN1571475A publication Critical patent/CN1571475A/en
Application granted granted Critical
Publication of CN1264338C publication Critical patent/CN1264338C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Studio Devices (AREA)
  • Television Signal Processing For Recording (AREA)
  • Gyroscopes (AREA)

Abstract

The present invention relates to a method for improving the stability of a video camera and an auxiliary device thereof. The present invention uses an angular velocity sensor to detect the angular speed of a video camera in real time when the video camera moves around three special axes of the video camera and synchronously record the angular speed and image signals. In post processing software, the variable quantity of the attitude angle of the video camera between two frames of images is obtained through a calculus. The variable quantity of the attitude angle corresponding to a slow-varying low-frequency component in the movement angular speed is reserved. The variable quantity of the attitude angle corresponding to a dithering angular speed component is eliminated. Then image dithering is eliminated through parallelly moving and rotating the image. The device is composed of three identical circuits, and each processing circuit is composed of an angular speed detection circuit in a signal acquisition circuit part, a level shift circuit in a signal processing circuit, a voltage / frequency conversion circuit, a frequency dividing circuit, a waveform transform circuit, and a signal synthesis and amplitude adjustment circuit which are serially connected in sequence. On output signal is sent to a sound recording interface of the video camera.

Description

Camera review increases steady method and servicing unit thereof
Technical field
The present invention when to be a kind of minimizing use shooting under hand-held and motion state the fuselage shake method and apparatus of image quality influence is belonged to the technical field of the design and the manufacturing of video camera servicing unit.
Background technology
Along with scientific and technical high development and living standards of the people improve constantly, video camera is increasingly extensive in the application in fields such as family, industry, military affairs.Yet when using under hand-held or motion state, the shake of video camera fuselage causes flating or fuzzy, has had a strong impact on visual effect and to identification, processing and the utilization of image.
From kinematic angle, the jitter motion of video camera can be decomposed into along the translation shake of spatial triaxial with around the corner of spatial triaxial shakes, wherein since translation shake to the influence of image and video camera range-independence to subject, the flating that the translation shake of trace causes is very slight, and corner shake is depended on shake angle and the product of video camera to object distance to the influence of image, thus corner to shake be the principal element that causes flating.
For a long time, people also studied the method that some remove flating.Mainly contain two kinds: the one, follow traditional way of on movable body, stablizing single component, promptly measure the corner shake of video camera,, make the camera lens of video camera make reciprocal equal amount of jitter with actuator by dynamo-electric feedback loop, thereby offset the shake of video camera, keep image stabilization and clear; The 2nd, along with the raising of computing power and the continuous development of image processing theory and technology, people attempt by the computing to image, remove flating with the method for pure software, promptly seek the corresponding points in the image, by translation and image rotating, reach the purpose that makes clear picture.
The driving actuator of dynamo-electric feedback loop directly and camera lens link, imaging is also moved with camera lens with focus adjusting mechanism, increase the structural complexity of video camera greatly and caused cost to rise and the volume increase, simultaneously dynamo-electric mechanism, stablizes the camera lens removal of images by electromechanical circuit and shakes the application of this method in video camera thereby these shortcomings have directly influenced all more than not going up field of electronic components at aspects such as reliability, life-span, anti-environment capacity and even maintainings.
The method that software goes to shake depends on the searching of the corresponding reference point in the adjacent two auspicious images and this reference point the asking for of due displacement in image.When subject and video camera itself all move (especially having the corner motion), because moment of momentum the unknown of video camera itself, verifiedly in theory can only obtain indefinite separating (being that infinite organize separated more) by the anti-moment of momentum of video camera and the kinematic parameter of subject asked of image reference point.By seek a plurality of reference points, the search best approximation method of separating is in theoretical method and realize that technical difficulty is all very big, effect is also very limited simultaneously.Just because this, increase steady the poster processing soft though there is the people to attempt to develop this class image, make slow progress, effect is also undesirable, does not still have application in video camera.
Yet if the moment of momentum of video camera self is known between adjacent two two field pictures, the method that so above-mentioned software goes to shake does not in theory just have problems, and the realization technology also can be simplified greatly.
Summary of the invention
Technical problem: at the problem of the above existence, the invention provides a kind of camera review and increase steady method and servicing unit thereof, this method use servicing unit measure and with the angle wobble information of image synchronization record video camera, increase the reprocessing of steady software by image, the flating that removal causes because of DE Camera Shake and fuzzy improves image definition.
Technical scheme: the angular movement speed when video camera is normally made a video recording is slowly and stably, and therefore, the alternation composition in its actual motion angular speed is shake angular speed.The present invention utilize angular-rate sensor detect in real time its spatial triaxial of camera intrinsic motion angular speed and with the picture signal synchronous recording, in the poster processing soft, obtain the variable quantity of video camera self attitude angle between two two field pictures through integration, keep the pairing attitude angle variable quantity of gradual low-frequency component in the motion angular speed, reject the pairing attitude angle variable quantity of shake angular speed composition, promptly ask for moment of momentum according to the pairing angular speed of gradual low frequency component in the camera motion angular speed, ask due translational movement of image and corner according to this moment of momentum of asking for, and then come translation and image rotating according to above-mentioned translational movement of asking for and corner, the flating that removal causes because of DE Camera Shake or vibration, realize image stabilization, improve image definition.Estimate this difficult problem of video camera attitude angle variable quantity thereby solved in the poster processing soft, greatly reduced difficulty and complexity that image increases steady software, make to handle and improve this technological approaches of image definition and become a reality by the poster processing soft.
The operating procedure of this method the poster processing soft is:
1), read adjacent two frame image datas,
2), seek on two two field pictures with reference to corresponding points, calculate translational movement and corner between two two field pictures,
3), read the motion angular speed of video camera between this two two field picture from the recording channel, get its low frequency component as analysis of spectrum,
4), ask for the video camera moment of momentum corresponding with low frequency component,
5), ask image that translational movement and corner should be arranged, revise second two field picture according to the video camera moment of momentum,
6), with second two field picture as first frame, read a follow-up two field picture as second frame, and then return step 2.
Camera review of the present invention increases the servicing unit of steady method, comprise signal acquisition circuit and signal processing circuit, this device is made up of the three tunnel identical circuit of measuring respectively with handling around spatial triaxial motion angular speed, each road circuit by angular velocity detection circuit, the level shift circuit of signal processing circuit part, voltage/frequency translation circuit, frequency dividing circuit, the waveform transformation circuit of signal acquisition circuit part, signal is synthetic is composed in series with the amplitude adjusting circuit order, output signal is delivered to the video camera recording access.
In first via treatment circuit, the reverse input end of the operational amplifier in the output termination level shift circuit of the angular-rate sensor in the angular velocity detection circuit, the input of the voltage/frequency converter in the output termination voltage/frequency translation circuit of operational amplifier, the input end of clock of voltage/frequency converter connects the clock end of the d type flip flop in the frequency dividing circuit, the reverse input end of the operational amplifier in the output welding wave translation circuit of frequency dividing circuit, the input of the two second-order low-pass filters of the output termination of operational amplifier, the output termination signal of two second-order low-pass filters synthetic with amplitude adjusting circuit in resistance, three road signals are synthetic after capacitance is connected to the video camera recording access.
It is proportional with the motion angular speed of camera intrinsic spatial triaxial respectively that the angular velocity detection circuit is exported three road voltages; Level shift circuit is transferred to different level ranges respectively with this three road voltage, guarantees non-overlapping copies; Thereby the voltage/frequency translation circuit can become be in the pulse signal in the frequency range of three non-overlapping copies with this three tunnel voltage transitions; Frequency dividing circuit with three road pulse signals respectively two divided-frequency become square wave, to reduce harmonic components wherein, be convenient to obtain good sine wave; The waveform transformation circuit is transformed to three road sine waves with three road square waves, and frequency separately is constant; Signal is synthetic to synthesize one road signal with amplitude adjusting circuit with three road sine waves, and exports the video camera recording access to after amplitude is adjusted into the video camera recording level.
Beneficial effect: the present invention has overcome and uses pure hardware means to remove the problem that complex structure, volume are big, cost is high and reliability is low that DE Camera Shake brings, and has overcome to use the pure software method to estimate camera motion angular speed result uncertain and realize technical complexity at software in theory.Combine by hardware and software, the motion angular speed of video camera measured and writes down in real time by use hardware, by the reprocessing of embedded software, eliminates because of fuselage and shake cause image blurring when video camera is shut down, and improves image definition, realizes image stabilization.
By different bias voltages is provided, the output voltage of No. three angular-rate sensors is transferred to different level ranges, and then by the voltage/frequency conversion, make them be in different frequency ranges, remove odd harmonic composition in each road pulse through two divided-frequency, finally obtain three tunnel frequency non-overlapping copies by filtering, but all be in the sine wave of audiorange, through synthetic boil down to one road signal, thereby can only use a recording sound channel of video camera just can write down the angular movement speed of camera intrinsic spatial triaxial with image synchronization ground.
Along with the high development of integrated circuit technique and the birth of micro-mechanical gyroscope, the sensor cluster of miniaturization and treatment circuit can be built in video camera, significantly do not increase the complexity of its structure.Digital camera has become the main product of video camera simultaneously, and image processing can be undertaken by the processor that is embedded in video camera, does not need to increase in addition microprocessor.
Description of drawings
Fig. 1 increases the composition frame chart of steady servicing unit for camera review.Wherein there are angular velocity detection circuit 1, level shift circuit 2, voltage/frequency translation circuit 3, frequency dividing circuit 4, waveform transformation circuit 5, signal to synthesize and amplitude adjusting circuit 6, video camera recording access 7.
Fig. 2 increases the electric schematic circuit of steady servicing unit for the present invention's camera review.
Fig. 3 is the FB(flow block) of the poster processing soft.
Embodiment
Fig. 2 is the schematic circuit that the present invention's camera review increases steady servicing unit, and each element takes off the row model and can realize:
Angular-rate sensor: U11, U21, U31:GYROCHIPII-QRS14;
Operational amplifier: U2:A, U2:B, U2:C:LF444;
Voltage → frequency changer: U13, U23, U33:AD650;
D type flip flop: U14, U24, U34:CD4013;
Assembly in operational amplifier: U15, U25, U35:U16, U26, the U36;
Two second-order low-pass filters; U16, U26, U36:MAX273;
Voltage stabilizing didoe: Z11, Z21, Z31:2DW232;
Below describe the operation principle of this device in detail.
At first define the reference axis and the angular movement of video camera.The OX axle is along the shooting direction, thereby the OX axle is perpendicular to the imaging surface of video camera; OY axle level in imaging surface is pointed to the video camera left side; The OZ axle points to zenith perpendicular to the OY axle in imaging surface.The rotation of camera intrinsic OX axle is called inclination, and Right deviation is for just; Rotation around the OY axle is called pitching, and nutation is for just; Be called around the rotation of OZ axle and sweep away, sweep to just on a left side.
When reality was made a video recording, the motion of video camera can be too inviolent, thinks to be no more than ± 25 °/S in pitching and the angular speed that sweeps away direction, is no more than ± 15 °/S at the angular speed of inclination direction.
The model of angular-rate sensor U11, U21 and U31 is GYROCHIPII-QRS14.Wherein angular-rate sensor U11, U21 measure video camera respectively at the angular speed that sweeps away with pitch orientation, and this two angular speed is in ± 25 °/S scope the time, and angular-rate sensor U11, U21 output voltage are all in 2.5V ± 1.25V scope; Angular-rate sensor U13 measures the angular speed of video camera in the inclination direction, and when this angular speed was in ± 15 °/S scope, angular-rate sensor U31 output voltage was in 2.5V ± 0.75V scope.
Operational amplifier U2 is the integrated circuit block that includes four operational amplifiers, and model is LF444, uses wherein three operational amplifier U2A, U2B and U2C here.Z11, Z21, Z31 are the 6.2V voltage stabilizing didoe, provide operating current by the 12V power supply by resistance R 105, R205, R305 respectively, through the dividing potential drop of resistance R 103 with R104, R203 and R204, R303 and R304, "+" input of operational amplifier U2A, U2B and U2C obtains the bias voltage of 2.125V, 3.625V and 4.875V respectively, resistance R 101 equates with the R302 value with R202, R301 with R102, R201, thereby the output voltage range of operational amplifier U2A, U2B and U2C is respectively 1.75V ± 1.25V, 4.75V ± 1.25V and 7.25V ± 0.75V.
The model of voltage → frequency changer U13, U23, U33 is AD650, and three's peripheral cell parameter is all identical, input voltage range 0-10V, output pulse frequency scope 0-40KHz.Under the effect of operational amplifier U2A, U2B and U2C output voltage, the frequency range of voltage → frequency changer U13, U23, U33 output pulse is respectively 2KHz-12KHz, 14KHz-24KHz and 26KHz-32KHz.
D type flip flop U14, U24, U34 are the CMOS d type flip flop, and model is CD4013, are square wave with input pulse two divided-frequency separately respectively, so that reject odd harmonic component wherein.Therefore the frequency range of d type flip flop U14, U24, U34 output square wave is respectively 1KHz-6KHz, 7KHz-12KHz and 13KHz-16KHz.
Operational amplifier U15, U25, U35 are the operational amplifier that is integrated in two second-order low-pass filter U16, U26, U35 inside, with peripheral Resistor-Capacitor Unit separately, constitute low-pass first order filter when reducing square wave level amplitude.The Resistor-Capacitor Unit that is positioned at operational amplifier U15, U25, U35 output is that resistance R 113 and capacitor C 14, resistance R 213 and capacitor C 24, resistance R 313 and capacitor C 34 are formed three passive low ventilating filters.
All there are two second-order low-pass filters two second-order low-pass filter U16, U26, U36 inside, and model is MAX270, and series connection is the quadravalence low pass filter separately.By the data wire pin D0-D6 of MAX270, the cut-off frequency of second-order low-pass filter U16, U26, U36 is set to 6KHz, 12KHz and 16KHz respectively.
Through six rank filtering, three road signals become frequency all in audiorange but non-overlapping copies, sine wave that waveform is good.Resistance R 117, R217, R317 and resistance R are formed summing circuit, and three road signals are synthesized one road signal, select the resistance of resistance R to make the amplitude of output signal and the level of video camera recording access require to adapt simultaneously.Capacitor C is isolated the DC component in the signal of front, only allows AC signal to export recording access to.
Fig. 3 is the main flow process of the poster processing soft.
At first read the data of continuous two auspicious images, seek on the two auspicious images with reference to corresponding points, calculate relative translation amount and corner between the two auspicious images according to the relative position relation of reference corresponding points; Read the motion angular speed of gamma camera between this two auspicious image from the recording channel, obtain its low frequency component through analysis of spectrum, integration is tried to achieve therewith the corner of low frequency component correspondence and is between this two auspicious image gamma camera around the due corner of its spatial triaxial, try to achieve due relative translation amount and corner between the two auspicious images in view of the above, and then revise the second auspicious image.
Preserve the above-mentioned first auspicious image, the above-mentioned second auspicious image as the first auspicious image, is read back one auspicious image as the second auspicious image, carry out above-mentioned processing again.The circulation said process disposes until all images that is write down.
1 in the technology that has proposed, and micro-mechanical gyroscope is used for camera lens more and stablizes closed loop in video camera.The present invention has founded a kind of new application process for micro-mechanical gyroscope in video camera, promptly be used for measuring with image synchronization ground and the angular movement information of record video camera, the information of this using method and surveying record can be used for Science and Technology fields such as the identification of image de-jittering, scene being shot and special object object and analysis and research, image processing, computer graphics study.
2 abandon by stablizing the pure hardware approach that camera lens reaches clear picture, thereby have avoided consequent complex structure, volume is big, cost is high, reliability is low shortcoming; Abandon estimating video camera angular movement parameter and then reaching the pure software method of clear picture, thereby avoided that consequent technical difficulty is big, amount of calculation is big, the complicated especially result of software has shortcomings such as uncertainty by image processing by software.Combine by hardware and software, use hardware measure in real time video camera angular move-ment signals and with the image synchronization record, between the video camera down period,, remove flating by the reprocessing of embedded software, realize image stabilization and clear.
The angular move-ment signals of the 3 camera intrinsic spatial triaxials that micro-mechanical gyroscope is recorded, transfer to different level ranges respectively, again by the voltage/frequency conversion, frequency division, filtering and synthetic, boil down to one tunnel audio signal, thereby can only use a recording sound channel of video camera, with the angular movement information of image synchronization ground record camera intrinsic spatial triaxial, other recording sound channels of video camera still can be used for writing down the sound of shooting scene.
The flating that 4 technology of the present invention and method not only can be used for removing hand-held camera, and can be used for being in the image de-jittering of the video camera in vehicle-mounted, Review for Helicopter, the multiple vibration environment such as unmanned airborne, realize image stabilization and clear, be beneficial to identification, research and processing special object.

Claims (4)

1, a kind of camera review increases steady method, the motion angular speed that it is characterized in that utilizing angular-rate sensor to detect its spatial triaxial of camera intrinsic in real time and with the picture signal synchronous recording, in the poster processing soft, obtain the variable quantity of video camera self attitude angle between two two field pictures through integration, keep the pairing attitude angle variable quantity of gradual low-frequency component in the motion angular speed, reject the pairing attitude angle variable quantity of shake angular speed composition, promptly ask for moment of momentum according to the pairing angular speed of gradual low frequency component in the camera motion angular speed, ask due translational movement of image and corner according to this moment of momentum of asking for, and then come translation and image rotating according to above-mentioned translational movement of asking for and corner, the flating that removal causes because of DE Camera Shake or vibration, realize image stabilization, improve image definition.
2, camera review according to claim 1 increases steady method, it is characterized in that the operating procedure of this method the poster processing soft is:
1), read adjacent two frame image datas,
2), seek on two two field pictures with reference to corresponding points, calculate translational movement and corner between two two field pictures,
3), read the motion angular speed of video camera between this two two field picture from the recording channel, get its low frequency component as analysis of spectrum,
4), ask for the video camera moment of momentum corresponding with low frequency component,
5), ask image that translational movement and corner should be arranged, revise second two field picture according to the video camera moment of momentum,
6), with second two field picture as first frame, read a follow-up two field picture as second frame, and then return step 2).
3, a kind of claim 1 or 2 described camera reviews of being applicable to increase the servicing unit of steady method, comprise signal acquisition circuit and signal processing circuit, it is characterized in that this device is made up of the three tunnel identical circuit of measuring respectively with handling around spatial triaxial motion angular speed, each road circuit is by the angular velocity detection circuit (1) of signal acquisition circuit part, the level shift circuit (2) of signal processing circuit part, voltage/frequency translation circuit (3), frequency dividing circuit (4), waveform transformation circuit (5), signal is synthetic to be composed in series with amplitude adjusting circuit (6) order, and output signal is delivered to video camera recording access (7).
4, servicing unit according to claim 3, it is characterized in that in first via treatment circuit, the output (OUT) of the angular-rate sensor (U11) in the angular velocity detection circuit (1) connects the reverse input end of the operational amplifier (U2A) in the level shift circuit (2), the input of the voltage/frequency converter (U13) in the output termination voltage/frequency translation circuit (3) of operational amplifier (U2A) (IN), the input end of clock (F0) of voltage/frequency converter (U13) connects the clock end (CLK1) of the d type flip flop in the frequency dividing circuit (4), the reverse input end of the operational amplifier (U15) in output (Q1) the welding wave translation circuit (5) of frequency dividing circuit (4), the input (INA) of the two second-order low-pass filters (U16) of the output termination of operational amplifier (U15), the output (OUTB) of two second-order low-pass filters (U16) connect signal synthetic with amplitude adjusting circuit (6) in resistance (R117), three road signals are synthetic after capacitance (C) is connected to video camera recording access (7).
CN 200410014862 2004-05-10 2004-05-10 Method of image stability improvement for video camera and assistor thereof Expired - Fee Related CN1264338C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200410014862 CN1264338C (en) 2004-05-10 2004-05-10 Method of image stability improvement for video camera and assistor thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200410014862 CN1264338C (en) 2004-05-10 2004-05-10 Method of image stability improvement for video camera and assistor thereof

Publications (2)

Publication Number Publication Date
CN1571475A CN1571475A (en) 2005-01-26
CN1264338C true CN1264338C (en) 2006-07-12

Family

ID=34478646

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200410014862 Expired - Fee Related CN1264338C (en) 2004-05-10 2004-05-10 Method of image stability improvement for video camera and assistor thereof

Country Status (1)

Country Link
CN (1) CN1264338C (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4505740B2 (en) 2005-05-16 2010-07-21 ソニー株式会社 Imaging apparatus and method for starting the same
JP4580838B2 (en) * 2005-07-29 2010-11-17 オリンパスイメージング株式会社 Electronic image stabilization device
US8134603B2 (en) 2005-08-12 2012-03-13 Nxp B.V. Method and system for digital image stabilization
CN101340513B (en) * 2007-07-02 2010-10-13 华硕电脑股份有限公司 Image processing method and electronic apparatus applying the same
EP2387229B1 (en) * 2010-05-14 2016-04-06 Casio Computer Co., Ltd. Image capturing apparatus and camera shake correction method, and computer-readable medium
CN103295213B (en) * 2013-06-07 2016-08-31 广州大学 Image stability augmentation method based on object tracing
CN104796596B (en) * 2014-01-20 2018-07-06 联想(北京)有限公司 A kind of information processing method and electronic equipment
TWI568186B (en) * 2014-06-09 2017-01-21 宏達國際電子股份有限公司 Portable device, manipulation method and non-transitory computer readable storage medium
CN104869310A (en) * 2015-05-18 2015-08-26 成都平行视野科技有限公司 Video shooting anti-shaking method based on mobile apparatus GPU and angular velocity sensor
CN105173102B (en) * 2015-09-18 2016-11-30 西北农林科技大学 A kind of quadrotor stability augmentation system based on many images and method
CN107241544B (en) * 2016-03-28 2019-11-26 展讯通信(天津)有限公司 Video image stabilization method, device and camera shooting terminal
DE112018002967T5 (en) 2017-06-12 2020-02-20 Fujifilm Corporation FURNITURE DETECTING DEVICE, IMAGING DEVICE, LENS ORB. LENS DEVICE, IMAGE DEVICE MAIN BODY, BLURRING DETECTION METHOD AND BLURRING DETECTION PROGRAM
CN107295255B (en) * 2017-06-21 2020-08-04 深圳市万普拉斯科技有限公司 Shooting mode determining method and device and terminal
CN110290323B (en) * 2019-06-28 2021-09-07 Oppo广东移动通信有限公司 Image processing method, image processing device, electronic equipment and computer readable storage medium
KR20230129058A (en) * 2021-01-04 2023-09-05 퀄컴 인코포레이티드 Image stabilization for integrated visual-inertial odometry and image processing

Also Published As

Publication number Publication date
CN1571475A (en) 2005-01-26

Similar Documents

Publication Publication Date Title
CN1264338C (en) Method of image stability improvement for video camera and assistor thereof
US10558057B2 (en) MEMS-based optical image stabilization
Zou et al. Learning to reconstruct high speed and high dynamic range videos from events
CN106550174B (en) A kind of real time video image stabilization based on homography matrix
US5629988A (en) System and method for electronic image stabilization
de Agapito et al. Self-Calibration of a Rotating Camera with Varying Intrinsic Parameters.
CN110595466B (en) Lightweight inertial-assisted visual odometer implementation method based on deep learning
CN1932841A (en) Petoscope based on bionic oculus and method thereof
CN101383899A (en) Video image stabilizing method for space based platform hovering
CN109079799A (en) It is a kind of based on bionical robot perception control system and control method
CN107145167B (en) Video target tracking method based on digital image processing technology
CN109922258A (en) Electronic image stabilization method, device and the readable storage medium storing program for executing of in-vehicle camera
CN104796596A (en) Information processing method and electronic equipment
CN1304878C (en) Compound eye stereoscopic vision device
CN114979489A (en) Gyroscope-based heavy equipment production scene video monitoring and image stabilizing method and system
CN111595332B (en) Full-environment positioning method integrating inertial technology and visual modeling
CN109238304A (en) Space camera ultrahigh speed becomes line frequency test device
CN108449528A (en) Image shift compensation circuit, aviation friction speed picture move recovery device and method
Suszynski et al. 2D image processing for auto-guiding system
WO2003107652A2 (en) A system for improving image resolution via sensor rotation
CN110493516A (en) A kind of friction speed image motion compensation device and system
Cai et al. Robust video stabilisation algorithm using feature point selection and delta optical flow
Gu et al. Real-time image mosaicing system using a high-frame-rate video sequence
Ahrns et al. Real-time monocular fixation control using the log-polar transformation and a confidence-based similarity measure
JPH09186951A (en) Image recording and reproducing device

Legal Events

Date Code Title Description
C06 Publication
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
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20060712