CN103957351A - Three-dimensional image pick-up system and image stabilization processing method thereof - Google Patents
Three-dimensional image pick-up system and image stabilization processing method thereof Download PDFInfo
- Publication number
- CN103957351A CN103957351A CN201410161543.1A CN201410161543A CN103957351A CN 103957351 A CN103957351 A CN 103957351A CN 201410161543 A CN201410161543 A CN 201410161543A CN 103957351 A CN103957351 A CN 103957351A
- Authority
- CN
- China
- Prior art keywords
- image
- image frame
- current
- motion vector
- dimensional
- 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.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
- Studio Devices (AREA)
Abstract
The invention discloses a three-dimensional image pick-up system and an image stabilization processing method thereof. The image stabilization processing method comprises the steps that three-dimensional motion vector data of a camera are detected in real time; a current vibration vector is calculated according to the detected three-dimensional motion vector data; the pixel offset quantity corresponding to the current vibration vector is determined according to the corresponding relation table of a prestored unit vibration vector and the pixel offset quantity; a current image frame is cut or spliced according to the determined pixel offset quantity, wherein only the current image frame and the part with the same image content in a previous image frame are reserved in a cut image; the part of the current image frame and the part of the previous image frame which share the same image content overlap in the spliced image; scaling is carried out on the cut image or the spliced image. According to the technical scheme, dithering of the image output by a receiving end can be greatly weakened in visual space, and moreover the cost and price of the three-dimensional image pick-up system are greatly lowered.
Description
Technical field
The present invention relates to image processing field, relate in particular to a kind of three-dimensional camera shooting system and image debounce processing method thereof.
Background technology
In the systems such as the real-time monitoring system of video of unmanned plane real-time video capture transmission control system, bullet train, the video data that video camera is taken after treatment, send to receiving terminal by wireless mode, receiving terminal is processed received video data, then output.But in these systems, video camera is often in severe operational environment, for example, vibrates strong unmanned plane and hang storehouse.If video camera is directly installed in such environment, the image of receiving terminal output is shaken on visual space very much, and the details that observer can not resolution image, can not watch such picture for a long time.
For solving the problem of flating, generally need video camera to be arranged on the three-dimensional stability platform of damping, but accurate three-dimensional stability platform cost compare is high, and the weight of himself is also larger, especially the unmanned plane during flying device of Load Sensitive is difficult to adopt such equipment.
Summary of the invention
The technical problem to be solved in the present invention is, for the defect that above-mentioned cost is high, weight is large of prior art, provides a kind of three-dimensional high-definition camera system and image debounce processing method thereof, and cost is low and gain in weight hardly.
The technical solution adopted for the present invention to solve the technical problems is: construct a kind of image debounce processing method of three-dimensional camera shooting system, comprising:
A1. detect in real time the three-dimensional motion vector data of video camera;
B1. calculate current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
C1. according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determine the corresponding pixel-shift quantity of current vibration vector;
D1. according to determined pixel-shift quantity, current image frame is carried out to cutting process, wherein, in the image after cutting, only retain the part of same image content in current image frame and previous image frame;
E1. the image after cutting is carried out to convergent-divergent processing, make the size of convergent-divergent image after treatment equal the size of the picture frame before cutting process.
In the image debounce processing method of three-dimensional camera shooting system of the present invention, in described step C1, according to the mapping table of unit vibration vector under pre-stored different focal and pixel-shift quantity, determine that current vibration vector is when corresponding pixel-shift quantity under front focal length.
In the image debounce processing method of three-dimensional camera shooting system of the present invention, in described steps A 1, be arranged on three dimensions gyro sensor on video camera and detect in real time the three-dimensional motion vector data of video camera.
The present invention also constructs a kind of image debounce processing method of three-dimensional camera shooting system, comprising:
A2. detect in real time the three-dimensional motion vector data of video camera;
B2. calculate current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
C2. according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determine the corresponding pixel-shift quantity of current vibration vector;
D2. according to determined pixel-shift quantity, current image frame and previous image frame are spliced to processing, wherein, in current image frame and previous image frame, the part of same image content is overlapping in spliced image;
E2. spliced image is carried out to convergent-divergent processing, make the size of convergent-divergent image after treatment equal to splice the size of picture frame before treatment.
In the image debounce processing method of three-dimensional camera shooting system of the present invention, in described step C2, according to the mapping table of unit vibration vector under pre-stored different focal and pixel-shift quantity, determine that current vibration vector is when corresponding pixel-shift quantity under front focal length.
In the image debounce processing method of three-dimensional camera shooting system of the present invention, in described steps A 2, be arranged on three dimensions gyro sensor on video camera and detect in real time the three-dimensional motion vector data of video camera.
The present invention also constructs a kind of three-dimensional camera shooting system, comprises video camera, and described three-dimensional camera shooting system also comprises:
First detection module, for detecting in real time the three-dimensional motion vector data of video camera;
The first computing module, for calculating current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
The first skew determination module, for according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determines the corresponding pixel-shift quantity of current vibration vector;
Cutting process module, for current image frame being carried out to cutting process according to determined pixel-shift quantity, wherein, only retains the part of same image content in current image frame and previous image frame in the image after cutting;
The first convergent-divergent processing module, for the image after cutting is carried out to convergent-divergent processing, makes the size of convergent-divergent image after treatment equal the size of the picture frame before cutting process.
In three-dimensional camera shooting system of the present invention, described first detection module is three dimensions gyro sensor, and described three dimensions gyro sensor is arranged on video camera.
The present invention also constructs a kind of three-dimensional camera shooting system, comprises video camera, and described three-dimensional camera shooting system also comprises:
The second detection module, for detecting in real time the three-dimensional motion vector data of video camera;
The second computing module, for calculating current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
The second skew determination module, for according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determines the corresponding pixel-shift quantity of current vibration vector;
Splicing processing module, for current image frame and previous image frame being spliced to processing according to determined pixel-shift quantity, wherein, in current image frame and previous image frame, the part of same image content is overlapping in spliced image;
The second convergent-divergent processing module, for spliced image is carried out to convergent-divergent processing, makes the size of convergent-divergent image after treatment equal to splice the size of picture frame before treatment.
In three-dimensional camera shooting system of the present invention, described the second detection module is three dimensions gyro sensor, and described three dimensions gyro sensor is arranged on video camera.
Implement technical scheme of the present invention, in the time that three-dimensional camera shooting system is worked in rugged environment, first, calculate current vibration vector according to detected three-dimensional motion vector data; Then,, according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determine the corresponding pixel-shift quantity of current vibration vector; Then, according to determined pixel-shift quantity, current image frame is carried out to cutting process or splicing processing; Finally, cutting or spliced image are carried out to convergent-divergent processing.Like this, when the captured video of this three-dimensional camera shooting system is sent to after receiving terminal, the shake of the image that receiving terminal is exported on visual space weakens greatly, its effect is equivalent to video camera to be arranged on the three-dimensional stability platform of damping, and, to compare three-dimensional stability platform is installed, the cost and price of three-dimensional camera shooting system of the present invention reduces greatly, and, increase hardly the weight of three-dimensional camera shooting system.
Brief description of the drawings
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing:
Fig. 1 is the flow chart of the image debounce processing embodiment of the method one of three-dimensional camera shooting system of the present invention;
Fig. 2 is the schematic diagram that uses the method shown in Fig. 1 to process image;
Fig. 3 is the flow chart of the image debounce processing embodiment of the method two of three-dimensional camera shooting system of the present invention;
Fig. 4 is the schematic diagram that uses the method shown in Fig. 3 to process image;
Fig. 5 is the logic diagram of three-dimensional camera shooting system embodiment one of the present invention;
Fig. 6 is the logic diagram of three-dimensional camera shooting system embodiment two of the present invention.
Embodiment
Fig. 1 is the flow chart of the image debounce processing embodiment of the method one of three-dimensional camera shooting system of the present invention, and this image debounce processing method comprises:
A1. detect in real time the three-dimensional motion vector data of video camera, in this step, for example, be arranged on three dimensions gyro sensor on video camera and detect in real time the three-dimensional motion vector data of video camera;
B1. calculate current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame, in this step, it should be noted that, the detection frequency of three-dimensional motion vector data will be higher than the shutter frequency of exposure of video camera, therefore, in the time calculating current vibration vector, only need be by the shooting of the shooting moment of current image frame and previous image frame time, inscribing corresponding three-dimensional motion vector data respectively subtracts each other,
C1. according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determine the corresponding pixel-shift quantity of current vibration vector, in this step, it should be noted that, can determine mapping table by the mode of test of many times in advance, for example, control video camera and carry out unit vibration, the initial time of the unit's of obtaining vibration and two two field pictures under the finish time, in this two two field picture, some picture material is identical, and some picture material is different.Then, this two two field picture is carried out to image recognition, judge the pixel-shift quantity of same image content in two two field pictures,, for this identical part of picture material, the position of each pixel of this part in the first picture frame compared its position in the second picture frame and is offset certain quantity;
D1. according to determined pixel-shift quantity, current image frame is carried out to cutting process, wherein, in the image after cutting, only retain the part of same image content in current image frame and previous image frame;
E1. the image after cutting is carried out to convergent-divergent processing, make the size of convergent-divergent image after treatment equal the size of the picture frame before cutting process, in this step, can carry out convergent-divergent processing by scaler chip.
In conjunction with Fig. 2, the previous image frame that picture frame 1 is current image frame, picture frame 2 is current image frame.In the time that image is carried out to debounce processing, first, the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing during by the shooting of picture frame 2 and the shooting of picture frame 1 are subtracted each other, to obtain current vibration vector.Then, according to pre-stored mapping table, determine the corresponding pixel-shift quantity of current vibration vector, for example, for being upwards offset 20 row pixels.Then, picture frame 2 is carried out to cutting process, the image after cutting process is image 3, and in image 3, only retains the part (dash area) of same image content in picture frame 2 and picture frame 1.Finally, image 3 is carried out to convergent-divergent processing, convergent-divergent image after treatment is image 4, and the size of image 4 equals the size of picture frame 2.
Fig. 3 is the flow chart of the image debounce processing embodiment of the method two of three-dimensional camera shooting system of the present invention, and this image debounce processing method comprises:
A2. detect in real time the three-dimensional motion vector data of video camera, in this step, for example, be arranged on three dimensions gyro sensor on video camera and detect in real time the three-dimensional motion vector data of video camera;
B2. calculate current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame, in this step, it should be noted that, the detection frequency of three-dimensional motion vector data will be higher than the shutter frequency of exposure of video camera, therefore, in the time calculating current vibration vector, only need be by the shooting of the shooting moment of current image frame and previous image frame time, inscribing corresponding three-dimensional motion vector data respectively subtracts each other,
C2. according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determine the corresponding pixel-shift quantity of current vibration vector, in this step, it should be noted that, can determine mapping table by the mode of test of many times in advance, for example, control video camera and carry out unit vibration, the initial time of the unit's of obtaining vibration and two two field pictures under the finish time, in this two two field picture, some picture material is identical, and some picture material is different.Then, this two two field picture is carried out to image recognition, judge the pixel-shift quantity of same image content in two two field pictures,, for this identical part of picture material, the position of each pixel of this part in the first picture frame compared its position in the second picture frame and is offset certain quantity;
D2. according to determined pixel-shift quantity, current image frame and previous image frame are spliced to processing, wherein, in current image frame and previous image frame, the part of same image content is overlapping in spliced image, in this step, it should be noted that, previous image frame can be the directly picture frame of output of video camera, also can be that a predetermined number picture frame splices picture frame after treatment before current image frame, also be, it can be the splicing of adjacent two two field pictures that splicing is processed, and can be also the splicing of the picture frames such as adjacent three or four.In the time determining the quantity of the picture frame splicing, should consider the frequency of video camera output frame;
E2. spliced image is carried out to convergent-divergent processing, the size that makes the size of convergent-divergent image after treatment equal to splice picture frame before treatment, in this step, can carry out convergent-divergent processing by scaler chip.
In conjunction with Fig. 4, the previous image frame that picture frame 1 is current image frame, picture frame 2 is current image frame.In the time that image is carried out to debounce processing, first, the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing during by the shooting of picture frame 2 and the shooting of picture frame 1 are subtracted each other, to obtain current vibration vector.Then, according to pre-stored mapping table, determine the corresponding pixel-shift quantity of current vibration vector, for example, for being upwards offset 20 row pixels.Then, picture frame 2 and picture frame 1 are spliced to processing, splice image after treatment and be image 3 ', and in picture frame 2 and picture frame 1 same image content part (dash area) image 3 ' in overlapping.Finally, to image 3 ' carry out convergent-divergent processing, convergent-divergent image after treatment be image 4 ', and image 4 ' size equal the size of picture frame 2 and picture frame 1.
Preferably, for three-dimensional camera shooting system, if it has been carried out to Focussing in shooting process, in the step C1 or step C2 of above-described embodiment, according to the mapping table of unit vibration vector under pre-stored different focal and pixel-shift quantity, determine that current vibration vector is when corresponding pixel-shift quantity under front focal length.
Fig. 5 is the logic diagram of three-dimensional camera shooting system embodiment one of the present invention, and the three-dimensional camera shooting system of this embodiment comprises first detection module 11, the first computing module 12, the first skew determination module 13, cutting process module 14 and the first convergent-divergent processing module 15.Wherein, first detection module 11 is for detecting in real time the three-dimensional motion vector data of video camera, and for example, first detection module 11 is three dimensions gyro sensor, and this three dimensions gyro sensor is arranged on video camera; The first computing module 12 is for calculating current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame; The first skew determination module 13, for according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, is determined the corresponding pixel-shift quantity of current vibration vector; Cutting process module 14, for current image frame being carried out to cutting process according to determined pixel-shift quantity, wherein, only retains the part of same image content in current image frame and previous image frame in the image after cutting; The first convergent-divergent processing module 15, for the image after cutting is carried out to convergent-divergent processing, makes the size of convergent-divergent image after treatment equal the size of the picture frame before cutting process.
Fig. 6 is the logic diagram of three-dimensional camera shooting system embodiment two of the present invention, and the three-dimensional camera shooting system of this embodiment comprises the second detection module 21, the second computing module 22, the second skew determination module 23, splicing processing module 24 and the second convergent-divergent processing module 25.Wherein, the second detection module 21 is for detecting in real time the three-dimensional motion vector data of video camera, and for example, the second detection module 21 is three dimensions gyro sensor, and this three dimensions gyro sensor is arranged on video camera; The second computing module 22 is for calculating current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame; The second skew determination module 23, for according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, is determined the corresponding pixel-shift quantity of current vibration vector; Splicing processing module 24 is for splicing processing according to determined pixel-shift quantity by current image frame and previous image frame, and wherein, in current image frame and previous image frame, the part of same image content is overlapping in spliced image; The second convergent-divergent processing module 25, for spliced image is carried out to convergent-divergent processing, makes the size of convergent-divergent image after treatment equal to splice the size of picture frame before treatment.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment of doing, be equal to replacement, improvement etc., within all should being included in claim scope of the present invention.
Claims (10)
1. an image debounce processing method for three-dimensional camera shooting system, is characterized in that, comprising:
A1. detect in real time the three-dimensional motion vector data of video camera;
B1. calculate current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
C1. according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determine the corresponding pixel-shift quantity of current vibration vector;
D1. according to determined pixel-shift quantity, current image frame is carried out to cutting process, wherein, in the image after cutting, only retain the part of same image content in current image frame and previous image frame;
E1. the image after cutting is carried out to convergent-divergent processing, make the size of convergent-divergent image after treatment equal the size of the picture frame before cutting process.
2. the image debounce processing method of three-dimensional camera shooting system according to claim 1, it is characterized in that, in described step C1, according to the mapping table of unit vibration vector under pre-stored different focal and pixel-shift quantity, determine that current vibration vector is when corresponding pixel-shift quantity under front focal length.
3. the image debounce processing method of three-dimensional camera shooting system according to claim 1, is characterized in that, in described steps A 1, is arranged on three dimensions gyro sensor on video camera and detects in real time the three-dimensional motion vector data of video camera.
4. an image debounce processing method for three-dimensional camera shooting system, is characterized in that, comprising:
A2. detect in real time the three-dimensional motion vector data of video camera;
B2. calculate current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
C2. according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determine the corresponding pixel-shift quantity of current vibration vector;
D2. according to determined pixel-shift quantity, current image frame and previous image frame are spliced to processing, wherein, in current image frame and previous image frame, the part of same image content is overlapping in spliced image;
E2. spliced image is carried out to convergent-divergent processing, make the size of convergent-divergent image after treatment equal to splice the size of picture frame before treatment.
5. the image debounce processing method of three-dimensional camera shooting system according to claim 4, it is characterized in that, in described step C2, according to the mapping table of unit vibration vector under pre-stored different focal and pixel-shift quantity, determine that current vibration vector is when corresponding pixel-shift quantity under front focal length.
6. the image debounce processing method of three-dimensional camera shooting system according to claim 4, is characterized in that, in described steps A 2, is arranged on three dimensions gyro sensor on video camera and detects in real time the three-dimensional motion vector data of video camera.
7. a three-dimensional camera shooting system, comprises video camera, it is characterized in that, described three-dimensional camera shooting system also comprises:
First detection module, for detecting in real time the three-dimensional motion vector data of video camera;
The first computing module, for calculating current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
The first skew determination module, for according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determines the corresponding pixel-shift quantity of current vibration vector;
Cutting process module, for current image frame being carried out to cutting process according to determined pixel-shift quantity, wherein, only retains the part of same image content in current image frame and previous image frame in the image after cutting;
The first convergent-divergent processing module, for the image after cutting is carried out to convergent-divergent processing, makes the size of convergent-divergent image after treatment equal the size of the picture frame before cutting process.
8. three-dimensional camera shooting system according to claim 7, is characterized in that, described first detection module is three dimensions gyro sensor, and described three dimensions gyro sensor is arranged on video camera.
9. a three-dimensional camera shooting system, comprises video camera, it is characterized in that, described three-dimensional camera shooting system also comprises:
The second detection module, for detecting in real time the three-dimensional motion vector data of video camera;
The second computing module, for calculating current vibration vector according to detected three-dimensional motion vector data, the difference of the three-dimensional motion vector data of inscribing when the three-dimensional motion vector data of inscribing when shooting that described current vibration vector is current image frame and the shooting of previous image frame;
The second skew determination module, for according to pre-stored unit vibration vector and the mapping table of pixel-shift quantity, determines the corresponding pixel-shift quantity of current vibration vector;
Splicing processing module, for current image frame and previous image frame being spliced to processing according to determined pixel-shift quantity, wherein, in current image frame and previous image frame, the part of same image content is overlapping in spliced image;
The second convergent-divergent processing module, for spliced image is carried out to convergent-divergent processing, makes the size of convergent-divergent image after treatment equal to splice the size of picture frame before treatment.
10. three-dimensional camera shooting system according to claim 9, is characterized in that, described the second detection module is three dimensions gyro sensor, and described three dimensions gyro sensor is arranged on video camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410161543.1A CN103957351B (en) | 2014-04-21 | 2014-04-21 | A kind of three-dimensional camera shooting system and its image debounce processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410161543.1A CN103957351B (en) | 2014-04-21 | 2014-04-21 | A kind of three-dimensional camera shooting system and its image debounce processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103957351A true CN103957351A (en) | 2014-07-30 |
| CN103957351B CN103957351B (en) | 2017-12-15 |
Family
ID=51334569
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410161543.1A Active CN103957351B (en) | 2014-04-21 | 2014-04-21 | A kind of three-dimensional camera shooting system and its image debounce processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103957351B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104683773A (en) * | 2015-03-25 | 2015-06-03 | 成都好飞机器人科技有限公司 | Video high-speed transmission method using unmanned aerial vehicle |
| CN110529896A (en) * | 2019-08-09 | 2019-12-03 | 杭州九阳小家电有限公司 | A kind of antidote and kitchen ventilator of kitchen ventilator Smoke Detection |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7218675B1 (en) * | 1991-06-24 | 2007-05-15 | Canon Kabushiki Kaisha | Signal processing device |
| CN101426091A (en) * | 2007-11-02 | 2009-05-06 | 韩国科亚电子股份有限公司 | Apparatus for digital image stabilization using object tracking and method thereof |
| CN103685950A (en) * | 2013-12-06 | 2014-03-26 | 华为技术有限公司 | Method and device for preventing shaking of video image |
-
2014
- 2014-04-21 CN CN201410161543.1A patent/CN103957351B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7218675B1 (en) * | 1991-06-24 | 2007-05-15 | Canon Kabushiki Kaisha | Signal processing device |
| CN101426091A (en) * | 2007-11-02 | 2009-05-06 | 韩国科亚电子股份有限公司 | Apparatus for digital image stabilization using object tracking and method thereof |
| CN103685950A (en) * | 2013-12-06 | 2014-03-26 | 华为技术有限公司 | Method and device for preventing shaking of video image |
Non-Patent Citations (1)
| Title |
|---|
| 李凤山: "视频序列电子稳像技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104683773A (en) * | 2015-03-25 | 2015-06-03 | 成都好飞机器人科技有限公司 | Video high-speed transmission method using unmanned aerial vehicle |
| CN110529896A (en) * | 2019-08-09 | 2019-12-03 | 杭州九阳小家电有限公司 | A kind of antidote and kitchen ventilator of kitchen ventilator Smoke Detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103957351B (en) | 2017-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102143456B1 (en) | Depth information acquisition method and apparatus, and image collection device | |
| US11222409B2 (en) | Image/video deblurring using convolutional neural networks with applications to SFM/SLAM with blurred images/videos | |
| CN104126299B (en) | Video image stabilisation | |
| US10733705B2 (en) | Information processing device, learning processing method, learning device, and object recognition device | |
| CN109855568B (en) | Method and device for detecting automatic driving sensor, electronic equipment and storage medium | |
| US9811751B2 (en) | Image pickup apparatus with boundary identifier for areas of interest | |
| US10991340B2 (en) | Image processing apparatus and image processing method | |
| US9554046B2 (en) | Method and image processing device for image stabilization of a video stream | |
| US20210402134A1 (en) | Information processing apparatus, information processing method, and storage medium | |
| US10755081B2 (en) | Method and apparatus for identifying pupil in image | |
| WO2016091545A1 (en) | Image processing method | |
| CN107124542B (en) | Image anti-shake processing method and device | |
| US10026158B2 (en) | Imaging device | |
| JP2017022692A (en) | Method and camera for producing image stabilized video | |
| CN105635569A (en) | Photographing method and device and terminal | |
| WO2021101732A1 (en) | Joint rolling shutter correction and image deblurring | |
| KR102592745B1 (en) | Posture estimating apparatus, posture estimating method and computer program stored in recording medium | |
| JP2010124028A (en) | Monitoring apparatus | |
| KR20160098853A (en) | Method for stabilizing image and electronic device thereof | |
| CN103957351A (en) | Three-dimensional image pick-up system and image stabilization processing method thereof | |
| CN105094614B (en) | Method for displaying image and device | |
| US20170364765A1 (en) | Image processing apparatus, image processing system, vehicle, imaging apparatus and image processing method | |
| WO2020003764A1 (en) | Image processing device, moving apparatus, method, and program | |
| CN114998686A (en) | Smoke detection model construction method, device, equipment, medium and detection method | |
| CN113610779B (en) | Vibrating mirror state detection method, device, equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |