CN105959531A - Moving image detection system and method - Google Patents

Moving image detection system and method Download PDF

Info

Publication number
CN105959531A
CN105959531A CN201610267095.2A CN201610267095A CN105959531A CN 105959531 A CN105959531 A CN 105959531A CN 201610267095 A CN201610267095 A CN 201610267095A CN 105959531 A CN105959531 A CN 105959531A
Authority
CN
China
Prior art keywords
image
sobel
eigenvalue
value
consecutive images
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
CN201610267095.2A
Other languages
Chinese (zh)
Inventor
刘思翔
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.)
Lemobile Information Technology (Beijing) Co Ltd
LeTV Holding Beijing Co Ltd
LeTV Mobile Intelligent Information Technology Beijing Co Ltd
Original Assignee
LeTV Holding Beijing Co Ltd
LeTV Mobile Intelligent Information Technology Beijing Co Ltd
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 LeTV Holding Beijing Co Ltd, LeTV Mobile Intelligent Information Technology Beijing Co Ltd filed Critical LeTV Holding Beijing Co Ltd
Priority to CN201610267095.2A priority Critical patent/CN105959531A/en
Publication of CN105959531A publication Critical patent/CN105959531A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/68—Control of cameras or camera modules for stable pick-up of the scene, e.g. compensating for camera body vibrations
    • H04N23/681—Motion detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a moving image detection system and a method wherein the system comprises an eigenvalue extracting unit for extracting the Sobel eigenvalue of each block image in two consecutive frame images and a motion judging unit connected to the eigenvalue extracting unit for comparing the mean Sobel eigenvalues in the N X N adjacent blocks with the same position in two consecutive frames wherein if there is a comparison value greater than a preset threshold value, it is judged that motion exists in the continuous images. According to the scheme of the invention, through the comparison of the mean Sobel eigenvalues in the N X N adjacent blocks with the same position in two consecutive frames, it is possible to judge whether there exists a motion in the consecutive images, and then fast and accurate estimation can be made to see if the preview images of the two frames prior shooting have moved or not so that the necessary parameter adjustment can be made to take clear pictures.

Description

The detecting system of a kind of moving image and method
Technical field
The present invention relates to technical field of image detection, particularly to detecting system and the side of a kind of moving image Method.
Background technology
When there being the scene of object of which movement with camera/mobile phone shooting, such as " child plays ", " Basketball Match " Etc., owing to the main body being taken is moved quickly, therefore it is easy to take On Local Fuzzy or have residual The photo of shadow.If now, if preview image can be utilized fast and accurately to judge this kind of scene, and The shutter speed taken pictures is improved accordingly, it is possible to well avoid the occurrence of this kind of asking when user presses shutter Topic.Whether current survey image exists the algorithm of motion that frame is poor, background subtraction, light stream etc., wherein What operand was minimum is frame difference method, but, for handheld camera/mobile phone, camera/mobile phone itself when taking pictures Exist for rocking in various degree, even the object being therefore photographed does not moves, continuous print preview image Also it is to there is difference in various degree between frame, if using whether frame difference method detection image exists motion, Will consider to do image alignment, the accuracy rate that otherwise can cause detection is relatively low, and does image alignment and be again Than relatively time-consuming operation.
Summary of the invention
The present invention provides a kind of detecting system and the method for moving image, pre-to solve two continuous frames before taking pictures What whether image of looking at moved judges complex technical problem.
According to an aspect of the present invention, it is provided that the detecting system of a kind of moving image, described system includes:
Characteristics extraction unit, special for extracting the Sobel of every block image in two frame consecutive images respectively Value indicative；
Motion determination unit, is connected with described characteristics extraction unit, is used for comparing in two frame consecutive images Average Sobel eigenvalue in the adjacent piecemeal of N × N that position is identical, if there is fiducial value more than pre- If threshold value, then judge consecutive image exists motion.
Wherein, in above-mentioned invention, described motion determination unit includes:
Computing module, for the average Sobel calculated in two frame consecutive images in the adjacent piecemeal of N × N respectively Eigenvalue；
Comparison module, is connected with described computing module, for the adjacent piecemeal of N × N that comparison position is identical Interior average Sobel eigenvalue, generates multiple fiducial value；
Judge module, is connected with described comparison module, for when minimum fiducial value is less than predetermined threshold value Judge consecutive image exists motion.
Wherein, in above-mentioned invention, described characteristics extraction unit includes:
Gray scale module, for becoming gray level image I by two frame consecutive images1(x) and I2(x)；
Piecemeal divides module, for by image I1(x) and I2X () is averagely divided into the block diagram that quantity is equal As B (x)；
Blocking characteristic value computing module, is used for calculating the Sobel eigenvalue F of every piece of image B (x).
Wherein, in above-mentioned invention, described blocking characteristic value computing module uses Sobel operator to do flat Face convolution algorithm obtains Sobel eigenvalue F；
This feature value F meets: F=S-preset value；
WhereinPreset value is test gained empirical value,
S x = - 1 0 1 - 2 0 2 - 1 0 1 * B ( x ) , S y = - 1 - 2 - 1 0 0 0 - 1 - 2 - 1 * B ( x ) .
Wherein, in above-mentioned invention, described gray scale module is to utilize yuv data to obtain gray level image, And gray level image is zoomed to the size of 640x480.
According to another aspect of the present invention, it is provided that the detection method of a kind of moving image, described method bag Include:
Extract the Sobel eigenvalue of every block image in two frame consecutive images respectively；
Average Sobel eigenvalue in the adjacent piecemeal of N × N that in relatively two frame consecutive images, position is identical, If there is fiducial value more than predetermined threshold value, then judge consecutive image exists motion.
Wherein, in above-mentioned invention, described compare N × N phase that in two frame consecutive images, position is identical Average Sobel eigenvalue in adjacent piecemeal, if there is fiducial value more than predetermined threshold value, then judges continuously Image exists motion, including:
Calculate the average Sobel eigenvalue in the adjacent piecemeal of N × N in two frame consecutive images respectively；
Average Sobel eigenvalue in the adjacent piecemeal of N × N that comparison position is identical, generates multiple comparison Value；
When minimum fiducial value is less than predetermined threshold value, it is judged that consecutive image exists motion.
Wherein, in above-mentioned invention, described extract every block image in two frame consecutive images respectively Sobel eigenvalue includes:
Two frame consecutive images are become gray level image I1(x) and I2(x)；
By image I1(x) and I2X () is averagely divided into block image B (x) that quantity is equal；
Calculate the Sobel eigenvalue F of every piece of image B (x).
Wherein, in above-mentioned invention, the Sobel eigenvalue F of described every piece of image B (x) of calculating adopts Do planar convolution computing with Sobel operator and obtain Sobel eigenvalue F；
This feature value F meets: F=S-preset value；
WhereinPreset value is test gained empirical value,
S x = - 1 0 1 - 2 0 2 - 1 0 1 * B ( x ) , S y = - 1 - 2 - 1 0 0 0 - 1 - 2 - 1 * B ( x ) .
Wherein, in above-mentioned invention, described two frame consecutive images are become gray level image I1(x) and I2(x) In utilize yuv data to obtain gray level image, and gray level image is zoomed to the size of 640x480.
The detecting system of the moving image of the present invention and method, by comparing position phase in two frame consecutive images The same average Sobel eigenvalue in the adjacent piecemeal of N × N, judges whether there is fortune in consecutive image Dynamic, and then realize estimating whether front cross frame preview image of taking pictures moves rapidly and accurately, in order to when taking pictures Do the parameter adjustment of necessity to shoot clear photograph.
Accompanying drawing explanation
Fig. 1 shows the detection fundamental diagram of moving image of the present invention；
Fig. 2 shows that the structure of the detecting system of the moving image provided in the specific embodiment of the invention is shown It is intended to；
Fig. 3 shows the flow process of the detection method of the moving image provided in the specific embodiment of the invention Figure；
Fig. 4 shows the method flow diagram of step S1 in Fig. 3；
Fig. 5 shows the method flow diagram of step S3 in Fig. 3.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention of greater clarity, below in conjunction with being embodied as Mode referring to the drawings, the present invention is described in more detail.It should be understood that these describe simply example Property, and it is not intended to limit the scope of the present invention.Additionally, in the following description, eliminate known knot Structure and the description of technology, to avoid unnecessarily obscuring idea of the invention.
The present invention utilizes the two continuous frames preview image data before camera/mobile phone photograph to calculate every two field picture Sobel characteristic point is distributed, the difference of comparative feature point distribution, if difference arrives empirical value, then judges There is motion in image, in order to does the parameter adjustment of necessity when taking pictures to shoot clear photograph.Concrete next Say, as it is shown in figure 1, the present invention travels through the adjacent piecemeal of all N × N such as 2 × 2 in two two field pictures respectively, Calculate their difference, find out 2 × 2 adjacent piecemeals that difference is minimum, and remember that their difference is d, will D and predetermined threshold value i.e. give experience threshold values t and compare, and judge whether picture material has motion, as d >=t Time show that picture material has a motion, otherwise when d < shows during t that picture material is without motion.
Fig. 2 shows that the structure of the detecting system of the moving image provided in the specific embodiment of the invention is shown It is intended to.
Below in conjunction with the structural representation described in Fig. 2, illustrate the detection of the moving image of the present embodiment System.
The detecting system of the moving image of the present embodiment at least includes: characteristics extraction unit 10 and motion are sentenced Disconnected unit 20.
Characteristics extraction unit 10 is special for the Sobel extracting every block image in two frame consecutive images respectively Value indicative.Specifically, characteristics extraction unit 10 includes: gray scale module 11, piecemeal divide module 12 With blocking characteristic value computing module 13.Wherein gray scale module 11 is for becoming gray scale by two frame consecutive images Image I1(x) and I2(x), this gray scale module 11 is to utilize yuv data to obtain gray level image, and by ash Degree image (i.e. the Y passage of YUV image) zooms to the size of 640 × 480 to improve performance.Piecemeal Divide module 12 for by image I1(x) and I2X () is averagely divided into block image B (x) that quantity is equal, Such as it is divided into 5 × 5 and has 25 pieces altogether.Blocking characteristic value computing module 13 is used for calculating every piece of image B (x) Sobel eigenvalue F.
In this enforcement, blocking characteristic value computing module 13 uses Sobel operator to do planar convolution computing and obtains Sobel eigenvalue F, eigenvalue F meet: F=S-preset value；
WhereinPreset value is test gained empirical value, typically takes 200.
S x = - 1 0 1 - 2 0 2 - 1 0 1 * B ( x ) , S y = - 1 - 2 - 1 0 0 0 - 1 - 2 - 1 * B ( x ) .
I.e. in one image block of note, S value is more than the number of the pixel of 220 (empirical values that experiment obtains) Amount is the Sobel eigenvalue of this image block.
Motion determination unit 20 is connected with characteristics extraction unit 10, this motion determination unit 20 for than Average Sobel feature N × N preferably 2 × 2 adjacent piecemeal in identical compared with position in two frame consecutive images Value, if there is fiducial value more than predetermined threshold value, then judges to exist in consecutive image motion.Specifically, Motion determination unit 20 includes: computing module 21, comparison module 22 and judge module 23.Wherein calculate Module 21 is used for the average Sobel eigenvalue calculated in two frame consecutive images in the adjacent piecemeal of N × N respectively, I.e. calculate the average Sobel eigenvalue in 2 × 2 adjacent piecemeals in two frame consecutive images.Comparison module 22 Being connected with computing module 21, this comparison module 22 is in the adjacent piecemeal of N × N that comparison position is identical Average Sobel eigenvalue, generate multiple fiducial value.Judge module 23 is connected with comparison module 22, When minimum fiducial value is more than predetermined threshold value, it is judged that module 23 judges to exist in consecutive image motion, When minimum fiducial value is less than or equal to predetermined threshold value, it is judged that module 23 judges not exist in consecutive image fortune Dynamic.
In the present embodiment, predetermined threshold value t is 20, and characteristics extraction unit 10 travels through in two two field pictures respectively All 2 × 2 adjacent piecemeals, calculate their difference, find out 2 × 2 adjacent piecemeals that difference is minimum, and The difference remembering them is d, d and predetermined threshold value is i.e. given experience threshold values t and compares, come in process decision chart picture Hold and whether have motion, show that as d>=t picture material has motion, otherwise when d<shows image during t Content is without motion.
As it has been described above, the detecting system of the moving image of the present invention, by comparing position in two frame consecutive images Put the average Sobel eigenvalue in the adjacent piecemeal of identical N × N, judge whether consecutive image is deposited In motion, and then realize estimating whether front cross frame preview image of taking pictures moves rapidly and accurately, in order to clap According to time do necessity parameter adjustment to shoot clear photograph.
Fig. 3 shows the flow process of the detection method of the moving image provided in the specific embodiment of the invention Figure.
Below in conjunction with the flow chart shown in Fig. 3, illustrate the detection method of the moving image of the present embodiment.
The detection method of the moving image of the present embodiment includes:
Step S1, extracts the Sobel eigenvalue of every block image in two frame consecutive images respectively.
Specifically, method flow diagram shown in Figure 4, this step S1 includes:
Two frame consecutive images are become gray level image I by step S111(x) and I2(x).In step S11, profit Obtain gray level image with yuv data, and gray level image (i.e. the Y passage of YUV image) is zoomed to The size of 640 × 480 is to improve performance.
Step S12, by image I1(x) and I2X () is averagely divided into block image B (x) that quantity is equal, Such as it is divided into 5 × 5 and has 25 pieces altogether.
Step S13, calculates the Sobel eigenvalue F of every piece of image B (x).In step S13, use Sobel Operator do planar convolution computing obtain Sobel eigenvalue F, eigenvalue F meet: F=S-preset value；
WhereinPreset value is test gained empirical value, typically takes 200.
S x = - 1 0 1 - 2 0 2 - 1 0 1 * B ( x ) , S y = - 1 - 2 - 1 0 0 0 - 1 - 2 - 1 * B ( x ) .
I.e. in one image block of note, S value is more than the number of the pixel of 220 (empirical values that experiment obtains) Amount is the Sobel eigenvalue of this image block.
Step S2, compares in the adjacent piecemeal of N × N preferably 2 × 2 that in two frame consecutive images, position is identical Average Sobel eigenvalue, if there is fiducial value more than predetermined threshold value, then judge consecutive image is deposited In motion.
Specifically, method flow diagram shown in Figure 5, this step S2 includes:
Step S21, calculates the average Sobel feature in the adjacent piecemeal of N × N in two frame consecutive images respectively Value.I.e. calculate the average Sobel eigenvalue in 2 × 2 adjacent piecemeals in two frame consecutive images.
Step S22, the average Sobel eigenvalue in the adjacent piecemeal of N × N that comparison position is identical, generate Multiple fiducial values.Average Sobel eigenvalue in 2 × 2 adjacent piecemeals that i.e. comparison position is identical, generates Multiple fiducial values.
Step S23, when minimum fiducial value is more than predetermined threshold value, it is judged that there is motion in consecutive image； When minimum fiducial value is less than or equal to predetermined threshold value, it is judged that consecutive image does not exist motion.
In sum, by above-mentioned flow process, first gray level image (i.e. the Y passage of YUV image) is contracted Being put into the size of 640 × 480 to improve performance, the image after note scaling is I (x)；Then gray-scale map is calculated The Sobel feature of picture: gray level image I (x) is equally divided into 5 × 5 and has 25 pieces altogether, each piece is counted respectively Calculating its Sobel feature, computational methods are as follows:
Remember that each image block is B (x), to its each pixel x, use Sobel operator to do plane volume Long-pending computing obtains:
Wherein, horizontal direction:
Wherein, vertical direction:
Here, in one image block of note, S value is more than the pixel of 220 (empirical values that experiment obtains) Quantity be the Sobel eigenvalue that F, F are this image block；Note I1(x) and I2The Sobel of (x) It is characterized as F1(x) and F2(x)。
To F1(x) and F2X () calculates the average Sobel eigenvalue difference in all 2 × 2 adjacent piecemeals Absolute value, is designated as A.
Such as F1X the data of () are as follows:
1 2 3 4 5
6 7 8 9 10
11 12 13 14 15
16 17 18 19 20
21 22 23 24 25
Such as F2X the data of () are as follows:
92 94 3 4 5
96 98 8 9 10
11 12 13 14 15
16 17 18 19 20
21 22 23 24 25
Then 1,2,6,7 piecemeals are had (abs is for taking absolute value):
A=abs ((1+2+6+7)/4 (92+94+96+98)/4)=91
In like manner, 19,20,24,25 piecemeals are had:
A=abs ((19+20+24+25)/4 (19+20+24+25)/4)=0
The minima calculating A successively is designated as Amin, if Amin is more than or equal to predetermined threshold value t, then schemes Having motion in Xiang, if Amin is less than or equal to predetermined threshold value t, then picture material is not moved.This In embodiment, t is 20.
It should be appreciated that the above-mentioned detailed description of the invention of the present invention is used only for exemplary illustration or explanation The principle of the present invention, and not being construed as limiting the invention, such as can also use as canny etc. other Do the operator of Image Edge-Detection to replace sobel operator.Therefore, without departing from the present invention spirit and Any modification, equivalent substitution and improvement etc. done in the case of scope, should be included in the guarantor of the present invention Within the scope of protecting.Additionally, claims of the present invention be intended to fall into scope and Whole in the equivalents on border or this scope and border change and modifications example.

Claims (10)

1. a detecting system for moving image, wherein, described system includes:
Characteristics extraction unit (10), for extracting every block image in two frame consecutive images respectively Sobel eigenvalue；
Motion determination unit (20), is connected with described characteristics extraction unit (10), is used for comparing Average Sobel eigenvalue in the adjacent piecemeal of N × N that in two frame consecutive images, position is identical, if There is fiducial value and be more than predetermined threshold value, then judge consecutive image exists motion.
2. the system as claimed in claim 1, wherein, described motion determination unit (20) including:
Computing module (21), for calculating in two frame consecutive images in the adjacent piecemeal of N × N respectively Average Sobel eigenvalue；
Comparison module (22), is connected with described computing module (21), for the N that comparison position is identical Average Sobel eigenvalue in the adjacent piecemeal of × N, generates multiple fiducial value；
Judge module (23), is connected with described comparison module (22), at minimum fiducial value Judge consecutive image exists motion during less than predetermined threshold value.
3. the system as claimed in claim 1, wherein, described characteristics extraction unit (10) is wrapped Include:
Gray scale module (11), for becoming gray level image I by two frame consecutive images1(x) and I2(x)；
Piecemeal divides module (12), for by image I1(x) and I2X () is averagely divided into quantity equal Block image B (x)；
Blocking characteristic value computing module (13), is used for calculating the Sobel eigenvalue F of every piece of image B (x).
4. system as claimed in claim 3, wherein, described blocking characteristic value computing module (13) Use Sobel operator to do planar convolution computing and obtain Sobel eigenvalue F；
This feature value F meets: F=S-preset value；
WhereinPreset value is test gained empirical value,
S x = - 1 0 1 - 2 0 2 - 1 0 1 * B ( x ) , S y = - 1 - 2 - 1 0 0 0 - 1 - 2 - 1 * B ( x ) .
5. system as claimed in claim 3, wherein, described gray scale module (11) is to utilize YUV Data obtain gray level image, and gray level image zooms to the size of 640x480.
6. a detection method for moving image, wherein, described method includes:
Extract the Sobel eigenvalue of every block image in two frame consecutive images respectively；
Average Sobel feature in the adjacent piecemeal of N × N that in relatively two frame consecutive images, position is identical Value, if there is fiducial value more than predetermined threshold value, then judges to exist in consecutive image motion.
7. method as claimed in claim 6, wherein, described compares position in two frame consecutive images Average Sobel eigenvalue in the adjacent piecemeal of identical N × N, if there is fiducial value more than presetting Threshold value, then judge to exist in consecutive image motion, including:
Calculate the average Sobel eigenvalue in the adjacent piecemeal of N × N in two frame consecutive images respectively；
Average Sobel eigenvalue in the adjacent piecemeal of N × N that comparison position is identical, generates multiple ratio Relatively it is worth；
When minimum fiducial value is less than predetermined threshold value, it is judged that consecutive image exists motion.
8. method as claimed in claim 6, wherein, described extracts in two frame consecutive images respectively The Sobel eigenvalue of every block image includes:
Two frame consecutive images are become gray level image I1(x) and I2(x)；
By image I1(x) and I2X () is averagely divided into block image B (x) that quantity is equal；
Calculate the Sobel eigenvalue F of every piece of image B (x).
9. method as claimed in claim 8, wherein, the Sobel of described every piece of image B (x) of calculating Eigenvalue F uses Sobel operator do planar convolution computing and obtain Sobel eigenvalue F；
This feature value F meets: F=S-preset value；
WhereinPreset value is test gained empirical value,
S x = - 1 0 1 - 2 0 2 - 1 0 1 * B ( x ) , S y = - 1 - 2 - 1 0 0 0 - 1 - 2 - 1 * B ( x ) .
10. two frame consecutive images wherein, described are become gray scale by method as claimed in claim 8 Image I1(x) and I2X () utilizes yuv data obtain gray level image, and gray level image is zoomed to The size of 640x480.
CN201610267095.2A 2016-04-26 2016-04-26 Moving image detection system and method Pending CN105959531A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610267095.2A CN105959531A (en) 2016-04-26 2016-04-26 Moving image detection system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610267095.2A CN105959531A (en) 2016-04-26 2016-04-26 Moving image detection system and method

Publications (1)

Publication Number Publication Date
CN105959531A true CN105959531A (en) 2016-09-21

Family

ID=56916289

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610267095.2A Pending CN105959531A (en) 2016-04-26 2016-04-26 Moving image detection system and method

Country Status (1)

Country Link
CN (1) CN105959531A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107231530A (en) * 2017-06-22 2017-10-03 维沃移动通信有限公司 A kind of photographic method and mobile terminal
CN107241504A (en) * 2017-06-08 2017-10-10 努比亚技术有限公司 A kind of image processing method, mobile terminal and computer-readable recording medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1688157A (en) * 2005-06-13 2005-10-26 北京中星微电子有限公司 Sports detecting method
CN102496276A (en) * 2011-12-01 2012-06-13 青岛海信网络科技股份有限公司 High efficiency vehicle detection method
CN103177454A (en) * 2011-12-24 2013-06-26 南京理工大学常熟研究院有限公司 Dynamic image moving object detection method
CN104835182A (en) * 2015-06-03 2015-08-12 上海建炜信息技术有限公司 Method for realizing dynamic object real-time tracking by using camera

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1688157A (en) * 2005-06-13 2005-10-26 北京中星微电子有限公司 Sports detecting method
CN102496276A (en) * 2011-12-01 2012-06-13 青岛海信网络科技股份有限公司 High efficiency vehicle detection method
CN103177454A (en) * 2011-12-24 2013-06-26 南京理工大学常熟研究院有限公司 Dynamic image moving object detection method
CN104835182A (en) * 2015-06-03 2015-08-12 上海建炜信息技术有限公司 Method for realizing dynamic object real-time tracking by using camera

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107241504A (en) * 2017-06-08 2017-10-10 努比亚技术有限公司 A kind of image processing method, mobile terminal and computer-readable recording medium
CN107241504B (en) * 2017-06-08 2020-03-27 努比亚技术有限公司 Image processing method, mobile terminal and computer readable storage medium
CN107231530A (en) * 2017-06-22 2017-10-03 维沃移动通信有限公司 A kind of photographic method and mobile terminal
CN107231530B (en) * 2017-06-22 2019-11-26 维沃移动通信有限公司 A kind of photographic method and mobile terminal

Similar Documents

Publication Publication Date Title
CN110692083B (en) Block-matched optical flow and stereoscopic vision for dynamic vision sensor
US8229172B2 (en) Algorithms for estimating precise and relative object distances in a scene
US9179071B2 (en) Electronic device and image selection method thereof
US8588546B2 (en) Apparatus and program for producing a panoramic image
US9202263B2 (en) System and method for spatio video image enhancement
US8243807B2 (en) Image processing method, a program of an image processing method, a recording medium on which a program of an image processing method is recorded and an image processing circuit
CN105635559B (en) Camera control method and device for terminal
US8111877B2 (en) Image processing device and storage medium storing image processing program
CN106952274B (en) Pedestrian detection and distance measuring method based on stereoscopic vision
US9361704B2 (en) Image processing device, image processing method, image device, electronic equipment, and program
CN110661977B (en) Subject detection method and apparatus, electronic device, and computer-readable storage medium
CN105931213B (en) The method that high dynamic range video based on edge detection and frame difference method removes ghost
WO2021017809A1 (en) Video denoising method and apparatus, and computer-readable storage medium
US9332185B2 (en) Method and apparatus for reducing jitters of video frames
US20090207260A1 (en) Image pickup apparatus and image pickup method
CN102317974A (en) The quality evaluation of image sequence
CN107784281B (en) Method for detecting human face, device, equipment and computer-readable medium
CN106469435B (en) Image processing method, device and equipment
KR102003460B1 (en) Device and Method for dewobbling
CN105959531A (en) Moving image detection system and method
CN106385546A (en) Method and system for improving image-pickup effect of mobile electronic device through image processing
CN105913395B (en) A kind of observation of moving target and smear restoration method
JP6579816B2 (en) Image processing apparatus, image processing method, and program
CN104616320A (en) Method for detecting vehicle in low-altitude aerial video based on gradient inhibition and epipolar constraint
JP2007087049A (en) Dynamic image processor

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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20160921