CN103379258A - Video compression method assisted by means of rotation operation and image acquisition device thereof - Google Patents
Video compression method assisted by means of rotation operation and image acquisition device thereof Download PDFInfo
- Publication number
- CN103379258A CN103379258A CN2012101188767A CN201210118876A CN103379258A CN 103379258 A CN103379258 A CN 103379258A CN 2012101188767 A CN2012101188767 A CN 2012101188767A CN 201210118876 A CN201210118876 A CN 201210118876A CN 103379258 A CN103379258 A CN 103379258A
- Authority
- CN
- China
- Prior art keywords
- reference picture
- present
- angle information
- rotation
- inventory
- 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
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention provides a video compression method assisted by means of rotation operation and an image acquisition device thereof. The method includes the rotation operation for image acquisition device detection and calculation of a rotation angle formed by the rotation operation. An image obtaining device is rotated by the rotation angle is utilized to shoot a current picture. In addition, whether a reference picture containing information of the rotation angle exists in a reference picture list stored in the image acquisition device or not is judged. If the reference picture containing the information of the rotation angle exists in the reference picture list stored in the image acquisition device, rotation shifting vectors are respectively used for searching matching blocks corresponding to current blocks in a first reference picture in the reference picture list specific to various blocks, and picture data of the current picture are compressed according to the rotation shifting vectors.
Description
Technical field
The invention relates to a kind of video-frequency compression method and image acquiring device, and particularly relevant for a kind of method and image acquiring device thereof that utilizes the compression of rotary manipulation auxiliary video.
Background technology
Be indebted to the lifting of network transmission technology and multimedia technology, the propagation of information is transformed into literal with a large amount of pictures by pure words and deposits, further develop into the multimedia video information of combining image and sound, add the in recent years prosperity of social network sites (social network), many people all can be with various multimedia video File Uploads to social network sites, in order to share with friend or discuss.
Yet, with regard to the data volume of the required transmission of various multimedia video file, the data volume of video image is quite large, and the increase along with various video camera resolution, the data volume of the video pictures that it obtains significantly promotes especially, if uncompressed is processed, will cause very large congested and time delay at network, so the technology that most usage data compresses when storing or transmit the multimedia video file.
Variation between the compression of above-mentioned video pictures and each picture has absolute relation, general amount of movement estimation (motion estimation) is crucial technology, block matching then is the most frequently used algorithm of realizing amount of movement estimation, and squeezed data according to this.Generally speaking, the amount of movement estimation often needs the sizable computational resource of cost and time, and the content of video pictures often can affect the accuracy of estimation.For instance, if the scene of video pictures changes or the shooting angle of video camera or camera site change, will cause the violent change between each video pictures, so that the usefulness of amount of movement estimation greatly descends.
Summary of the invention
In view of this, the invention provides a kind of method and image acquiring device thereof that utilizes the compression of rotary manipulation auxiliary video, can correspondingly according to user's rotary manipulation control the captured picture of image acquiring device, and utilize efficient and the accuracy of this rotary manipulation lifting mobile amount estimation.
The invention provides a kind of method of utilizing the compression of rotary manipulation auxiliary video, be applicable to image acquiring device, the method comprises the following steps.At first, the rotary manipulation that the detected image deriving means receives, and calculate the formed anglec of rotation of this rotary manipulation.Then, thereby utilize this anglec of rotation image rotating deriving means to take a present picture (current frame), and this anglec of rotation is saved as present angle information.In addition, judge in the stored reference picture inventory of image acquiring device (reference frame list) whether have the reference picture with rotation angle information.If not, the vector take this anglec of rotation as the function calculation rotary moving, and to distinguish present picture be a plurality of present blocks, utilize respectively the first reference picture search of this rotary moving vector in the reference picture inventory corresponding to coupling block (the Matching Block of each present block for each present block, MB), and according to this compress the picture data of each present block and corresponding coupling block.
In one embodiment of this invention, the above-mentioned method of utilizing the compression of rotary manipulation auxiliary video also comprises renewal reference picture inventory.Wherein, present picture is saved as the second reference picture in the reference picture inventory, and present angle information is stored in rotation angle information corresponding to the second reference picture.
In one embodiment of this invention, if the storage area of above-mentioned judgement reference picture inventory is full, then seek three reference picture the most similar to present angle information in the rotation angle information of self-reference picture inventory.And the rotation angle information of deleting the 3rd reference picture and having, and present picture and present angle information be stored in the reference picture inventory.
In one embodiment of this invention, if above-mentioned judging in the reference picture inventory exists the reference picture with rotation angle information, then also comprise the following steps.Seek the 4th reference picture in the self-reference picture inventory, wherein the rotation angle information that has of the 4th reference picture and the differential seat angle of present angle information are less than a preset value.Then, distinguishing the 4th reference picture is a plurality of reference block.And calculate respectively the difference summation of the brightness value of a plurality of pixels in each present block and the corresponding reference block, and respectively with a threshold value relatively.Wherein, if there is the difference summation of N present block to be not more than threshold value, then carry out mobile estimating (motion estimation) take the relative position of present block as starting point, thereby at the coupling block of the 4th reference picture search corresponding to each present block, and compress according to this picture data of present block and corresponding coupling block, wherein N is the positive integer greater than 1.
In one embodiment of this invention, also comprise this reference picture inventory of renewal after the above-mentioned step carrying out mobile estimating.Wherein, if the rotation angle information that the 4th reference picture has is same as present angle information, the rotation angle information of then deleting the 4th reference picture and having, and present picture and present angle information be stored in the reference picture inventory.
In one embodiment of this invention, the rotary manipulation that above-mentioned detected image deriving means receives, and the step of calculating the formed anglec of rotation of this rotary manipulation comprises the touching rotary manipulation that the touch-screen on the detected image deriving means receives, and calculates the formed anglec of rotation of this touching rotary manipulation.
In one embodiment of this invention, the rotary manipulation that above-mentioned detected image deriving means receives, and the step of calculating the formed anglec of rotation of this rotary manipulation comprises that a three-dimensional rotation that receives by displacement detection module detected image deriving means operates, and operates the formed anglec of rotation thereby calculate this three-dimensional rotation.
The present invention provides again a kind of image acquiring device, and it comprises camera lens module, detection module, memory cell and control unit.Wherein, detection module is in order to detect rotary manipulation.Memory cell is in order to store the reference picture inventory.Control unit is coupled to camera lens module, detection module and memory cell.Control unit is taken a present picture thereby utilize this anglec of rotation control camera lens module to be rotated, and this anglec of rotation is saved as present angle information in order to calculate the formed anglec of rotation of rotary manipulation.Wherein, control unit and judge in the stored reference picture inventory of memory cell whether exist a reference picture with rotation angle information.If not, the one rotary moving vector take this anglec of rotation as function calculation then.Moreover, distinguishing present picture is a plurality of present blocks, utilize this rotary moving vector in the first reference picture of reference picture inventory, to search for coupling block corresponding to each present block for each present block, and compress according to this picture data of present block and corresponding coupling block.
In one embodiment of this invention, above-mentioned image acquiring device also comprises the amount of movement estimation unit that is coupled to control unit.If seek the 4th reference picture in the control unit self-reference picture inventory, wherein the rotation angle information that has of the 4th reference picture and the differential seat angle of present angle information are less than a preset value, and then distinguishing the 4th reference picture is a plurality of reference block.And calculate respectively the difference summation of the brightness value of a plurality of pixels in each present block and the corresponding reference block, and respectively with a threshold value relatively.Wherein, if there is the difference summation of N present block to be not more than threshold value, then mobile estimating is carried out take the relative position of present block as starting point in amount of movement estimation unit, and provide carry out the mobile estimating gained corresponding to the coupling block of each present block to control unit.Compress according to this picture data of present block and corresponding coupling block by control unit, wherein N is the positive integer greater than 1.
In one embodiment of this invention, above-mentioned image acquiring device also comprises the touch-screen that couples detection module.Detection module detects the touching rotary manipulation that touch-screen receives, and this touching rotary manipulation is offered control unit, calculates according to this formed anglec of rotation of this touching rotary manipulation.
In one embodiment of this invention, above-mentioned image acquiring device also comprises the displacement detection module that couples control unit.The three-dimensional rotation operation that displacement detection module detected image deriving means receives operates the formed anglec of rotation thereby provide control unit to calculate this three-dimensional rotation.
In sum, method and the image acquiring device thereof of rotary manipulation auxiliary video compression provided by the present invention, the picture that can want to obtain according to the camera lens of the corresponding control of user's rotary manipulation image acquiring device, the user does not need direct moving images deriving means.In addition, accelerate the search of coupling block thereby utilize above-mentioned rotary manipulation can further calculate the rotary moving vector, accordingly, but efficient and the accuracy of the estimation of lifting mobile amount, and assist according to this compressed video picture.
For above-mentioned feature and advantage of the present invention can be become apparent, embodiment cited below particularly, and cooperate accompanying drawing to be described in detail below.
Description of drawings
Fig. 1 is the calcspar of the image acquiring device that illustrates according to one embodiment of the invention;
Fig. 2 is a kind of method flow diagram that utilizes the compression of rotary manipulation auxiliary video that illustrates according to one embodiment of the invention;
Fig. 3 A is a kind of schematic diagram of touching rotary manipulation at touch-screen that illustrates according to one embodiment of the invention;
Fig. 3 B is the schematic diagram of touching rotary manipulation according to the another kind that one embodiment of the invention illustrates at touch-screen;
Fig. 4 is a kind of schematic diagram that utilizes anglec of rotation control image acquiring device to be rotated rear captured present picture that illustrates according to one embodiment of the invention;
Fig. 5 is the calcspar of the image acquiring device that illustrates according to another embodiment of the present invention;
Fig. 6 is a kind of method flow diagram that utilizes the compression of rotary manipulation auxiliary video that illustrates according to another embodiment of the present invention;
Fig. 7 is the schematic diagram of a kind of reference picture inventory of illustrating according to another embodiment of the present invention.
Description of reference numerals:
100,500: image acquiring device;
110: camera lens module;
120: detection module;
130: memory cell;
140: control unit;
142: amount of movement estimation unit;
150: touch-screen;
560: displacement detection module;
700: the reference picture inventory;
A, B, A ', B ', C, D, D ': touch points;
CB: present block;
W: horizontal width;
L: vertical length;
D1～d3: the direction of arrow;
F1: previous picture;
F2: present picture;
MV1, MV2: motion-vector;
Ref1～refn: reference picture;
Tri1～tri4: triangular form zone;
θ, θ 1～θ k, α: the anglec of rotation;
S210～S240: a kind of each step of utilizing the method for rotary manipulation auxiliary video compression;
S610～S670: another kind utilizes each step of the method for rotary manipulation auxiliary video compression.
Embodiment
Fig. 1 is the calcspar of the image acquiring device that illustrates according to one embodiment of the invention.The image acquiring device of present embodiment for example is that digital camera, monitoring camera, roll film camera or other can in order to the image acquiring device of video recording, be not limited to above-mentioned.Please refer to Fig. 1, image acquiring device 100 comprises camera lens module 110, detection module 120, memory cell 130 and control unit 140, and in addition, image acquiring device 100 also can comprise touch-screen 150.Its function is described below:
The triggering signal that detection module 120 produces when carrying out trigger action in order to detect the user.So-called trigger action comprises following several different execution mode: in the present embodiment, image acquiring device 100 comprises the touch-screen 150 that is coupled to detection module 120, so the detection module 120 touching rotary manipulation that can carry out on touch-screen 150 in order to detect the user; If image acquiring device 100 comprises speech identifying function, then the user can utilize voice to carry out trigger action, and 120 of detection modules can be in order to detect voice content.Therefore, detection module 120 is that the function that has of collocation image acquiring device 100 is designed, is not limited to above-mentioned.
Touch-screen 150 for example is by liquid crystal display (Liquid Crystal Display, LCD) or light-emitting diode (Light-Emitting Diode, LED) display such as display, and the contact panels such as resistance-type or condenser type combine, and it can provide simultaneously and show and the touching operating function.
Fig. 2 is a kind of method flow diagram that utilizes the compression of rotary manipulation auxiliary video that illustrates according to one embodiment of the invention.The method of present embodiment is applicable to the image acquiring device 100 of Fig. 1, below namely the arrange in pairs or groups detailed step of each member explanation present embodiment auxiliary video compression method in the image acquiring device 100:
Please be simultaneously with reference to Fig. 1 and Fig. 2, shown in step S210, the rotary manipulation that detection module 120 detected image deriving means 100 receive, and offer control unit 140 and calculate the formed anglec of rotation of these rotary manipulations.In the present embodiment, image acquiring device 100 has a touch-screen 150, and therefore, detection module 120 is to detect the touching rotary manipulation that touch-screen 150 receives.It for example is the touching rotary manipulation that utilizes hand on touch-screen 150, to produce by a user.
In more detail, Fig. 3 A is a kind of schematic diagram of touching rotary manipulation at touch-screen that illustrates according to one embodiment of the invention.Please refer to Fig. 3 A, suppose that the user uses two fingering row touching rotary manipulations simultaneously, be two touch points A and the touch points B that refer to be positioned at first respectively on the touch-screen 150 of user, and respectively according to direction of arrow d1 and the d2 towing slides into touch points A ' and touch points B ' just leaves touch-screen 150 with two fingers afterwards.Detection module 120 for example offers control unit 140 with the position of touch points A, B, A ', B '; Perhaps detection module 120 directly provides line between control unit 140 touch points A, the B and the link information between touch points A ', the B ', in order to allow control unit 140 calculate anglec of rotation θ.
Fig. 3 B is the schematic diagram of touching rotary manipulation according to the another kind that one embodiment of the invention illustrates at touch-screen.Please refer to Fig. 3 B, suppose that the user uses two fingering row touching rotary manipulations simultaneously, be two touch points C and the touch points D that refer to be positioned at first respectively on the touch-screen 150 of user, and incite somebody to action wherein finger towing according to direction of arrow d3 and slide into touch points D ' (another refers to be fixed in touch points C simultaneously), just two fingers are left touch-screen 150 afterwards.Detection module 120 for example offers control unit 140 with the position of touch points C, D, D '; Perhaps detection module 120 directly provides about the link information between the line between touch points C, the D and touch points C, the D ' to control unit 140, in order to allow control unit 140 calculate anglec of rotation α.
Therefore, the present invention does not limit the form of touching rotary manipulation, everyly can in order to produce the touching drag operation of an anglec of rotation, all can be used as alternate embodiment of the present invention.Next, as described in step S220, control unit 140 utilizes the camera lens module 110 of above-mentioned anglec of rotation control image acquiring device 100 to be rotated, thereby takes a present picture (current frame), and this anglec of rotation is saved as present angle information.In detail, Fig. 4 is a kind of schematic diagram that utilizes anglec of rotation control image acquiring device to be rotated rear captured present picture that illustrates according to one embodiment of the invention.Please refer to Fig. 4, camera lens module 110 image of clapping before rotation is previous picture f1, supposes that present embodiment receives the touching rotary manipulation such as Fig. 3 A, so camera lens module 110 is taken after foundation anglec of rotation θ is rotated and obtained a present picture f2.And anglec of rotation θ namely saves as present angle information.
In step S230, control unit 140 judges whether reference picture inventory stored in the memory cell 130 of image acquiring device 100 (reference frame list) exists the reference picture with rotation angle information.Wherein, the reference picture inventory be for image acquiring device 100 in the captured video pictures of obtaining of previous different time points, generally speaking, the reference picture inventory can be and dynamically updates, that is to say and work as image acquiring device 100 along with the time is constantly obtained new video pictures, and many stored reference pictures also can upgrade in time in the reference picture inventory.Moreover the reference picture inventory for example can store simultaneously 16 reference pictures or store simultaneously 32 reference pictures, can be set according to actual conditions by the user.
If the stored reference picture of reference picture inventory does not have rotation angle information, subsequent steps S240 then, the vector take this anglec of rotation as the function calculation rotary moving, and to distinguish present picture be a plurality of present blocks, utilize respectively the first reference picture search of this rotary moving vector in the reference picture inventory corresponding to coupling block (the Matching Block of each present block for each present block, MB), and according to this compress the picture data of each present block and corresponding coupling block.
Specifically, suppose that one among Fig. 4 is (Xpos in the absolute position at block CB place at present, Ypos), and suppose present block CB formerly among the picture f1 the known motion-vector of correspondence position be that MV1 and its coordinate are (x, y), the present invention rotates front known motion-vector MV1 as searching for the coupling block with reference to value, but one new rotary moving vector M V2 and its coordinate are (x ', y '), above-mentioned Xpos take above-mentioned anglec of rotation θ as function calculation, Ypos, x, y, x ', y ' is all integer, and its conversion formula is as follows:
After trying to achieve rotary moving vector M V2, just can distinguish present picture is a plurality of present blocks, utilize respectively first reference picture (as shown in Figure 4 the previous picture f1) search of rotary moving vector M V2 in the reference picture inventory corresponding to the coupling block of each present block for each present block, and compress according to this picture data of each present block and corresponding coupling block.Because the present invention considers postrotational video pictures for the impact of motion-vector, therefore utilize rotary moving vector M V2 can significantly promote the speed of search coupling block algorithm.
Yet what must illustrate is that delta-shaped region tri1～tri4 as shown in Figure 4 is owing to accessed video pictures after the rotation, and can't adopt said method to utilize rotary moving vector M V2 formerly to find the coupling block among the picture f1.Therefore, for the delta-shaped region tri1～tri4 that avoids accurately estimating, can be via the scope of following formula restricting rotation motion-vector MV2, its formula is as follows:
x”＝Clip(Xpos+x’，0，W)
y”＝Clip(Ypos+y’，0，L) (2)
Wherein, W is the horizontal width of previous picture f1, L is the vertical length of previous picture f1, coordinate (x "; y ") be the rotary moving vector behind the limited field, function Clip allows coordinate (x ", y ") satisfy 0≤condition of (x "=Xpos+x ')≤W and 0≤(y "=Ypos+y ')≤L.
Except the scope of above-mentioned restricting rotation motion-vector, estimation also provides another kind of solution for the motion-vector of delta-shaped region tri1～tri4 in the present invention, below comes with another embodiment namely that the present invention will be described.
Fig. 5 is the calcspar of the image acquiring device that illustrates according to another embodiment of the present invention.In the present embodiment, image acquiring device 500 also comprises amount of movement estimation unit 142 and displacement detection module 560 except comprising camera lens module 110, detection module 120, memory cell 130, control unit 140 and touch-screen 150.Wherein, displacement detection module 560 is coupled to control unit 140, and it for example is acceleration transducer (G-sensor), inclination sensor (tilt sensor) or gyro sensor (Gyro sensor).
Fig. 6 is a kind of method flow diagram that utilizes the compression of rotary manipulation auxiliary video that illustrates according to another embodiment of the present invention.Below the function mode of image acquiring device 500 will be described with Fig. 6.Please be simultaneously with reference to Fig. 5 and Fig. 6.
At first, the rotary manipulation that detected image deriving means 500 receives, and offer the control unit 140 calculating formed anglecs of rotation of this rotary manipulation (step S610).It should be noted that, suppose that touch-screen 150 in the present embodiment is touch-screen turning or rotation, then displacement detection module 560 can be in order to the tilt variation that detects touch-screen 150, various speed or acceleration information etc., therefore thereby present embodiment calculates the anglec of rotation on a plane except receive a touching drag operation by touch-screen 150, also can calculate by the information that displacement detection module 560 provides a three-dimensional rotation angle of touch-screen 150.
Then, control unit 140 utilizes the camera lens module 110 of above-mentioned anglec of rotation control image acquiring device 100 to be rotated, thereby takes a present picture, and this anglec of rotation is saved as present angle information (step S620).Wherein, camera lens module 110 can carry out two-dimentional spin or three-dimensional rotation by corresponding touch-screen 150.
Afterwards, control unit 140 judges whether reference picture inventory stored in the memory cell 130 of image acquiring device 100 (reference frame list) exists the reference picture (step S630) with rotation angle information.If not, subsequent steps S640 then.Above-mentioned steps S610～S640 and previous embodiment step S210～S240 is identical or similar, so its detailed content does not repeat them here.
Different from previous embodiment is, if there is the reference picture with rotation angle information really in the reference picture inventory of present embodiment, the rotation angle information that then searching one reference picture in the self-reference picture inventory, and this reference picture has and the differential seat angle of present angle information are less than a preset value (step S650).
Specifically, Fig. 7 is the schematic diagram of a kind of reference picture inventory of illustrating according to another embodiment of the present invention.Please refer to Fig. 7, reference picture inventory 700 has n and opens reference picture, and n is a positive integer.Wherein, reference picture ref1 is set as previous picture (previous frame), namely classifies according to acquisition time, and previous picture is acquisition time point near a reference picture of present picture.In other words, new present picture of every shooting, previous picture will constantly be done renewal.In addition, reference picture ref2 then has a rotation angle information (that is, anglec of rotation θ 1), and it has represented reference picture ref2 is the captured video pictures of obtaining after the rotation of camera lens module 110 according to anglec of rotation θ 1; Reference picture ref3 also has a rotation angle information (that is, anglec of rotation θ 2), and it has represented reference picture ref3 is the captured video pictures of obtaining after the rotation of camera lens module 110 according to anglec of rotation θ 2; The rest may be inferred for all the other.
According to the above-mentioned method that illustrates step S650 searching reference picture.Suppose that the shown anglec of rotation of present angle information is 30 degree, in the present embodiment, if preset value is set as 5 degree, then expression must be sought rotation angle information and be presented at 25 degree to the reference picture between 35 degree from reference picture inventory 700, if there are simultaneously many reference pictures all to meet above-mentioned condition, then choose a reference picture the most similar to present angle information (that is, the 4th reference picture), for example can select the anglec of rotation to be similarly the reference picture of 30 degree.Wherein, preset value can be done setting according to actual conditions by those skilled in the art.
Next in step S660, just the 4th reference picture is divided into a plurality of reference block.And calculate respectively the difference summation of the brightness value of a plurality of pixels in each present block and the corresponding reference block, and respectively with a threshold value relatively.In one embodiment, suppose that most reference block is neither greater than threshold value with the difference summation of present block, then utilize amount of movement estimating module 142 to carry out mobile estimatings (motion estimation) take the relative position of present block as starting point, thereby search is corresponding to the coupling block of each present block in reference picture, and compresses according to this picture data of present block and corresponding coupling block.
Yet, what must illustrate is, (for example: half) the difference summation of above reference block and present block is all greater than threshold value if suppose that some is arranged, although then representing reference picture is by the captured video pictures of the identical anglec of rotation with present picture, but scene being shot may change (scene change), just can cause the difference sum variation violent, so will change the estimating and measuring method that adopts the described rotary moving vector of previous embodiment.Accordingly, the present invention does not need to know in advance scene content being shot, can use the method for present embodiment to judge.
At last, after the step of carrying out mobile estimating, must upgrade reference picture inventory (step S670).If subsequent steps S640, owing to wouldn't have any reference picture with rotation angle information in the reference picture inventory of step S640, therefore update method only need deposit present picture in reference picture ref1 (namely, previous picture) and with present picture deposit in addition reference picture ref2～refn one of them, present angle information corresponding to record simultaneously.If the storage area of reference picture ref2～refn is full, then seek three reference picture the most similar to present angle information in the rotation angle information of self-reference picture inventory, in other words namely seek the anglec of rotation and differ minimum reference picture.The rotation angle information of deleting the 3rd reference picture and having, and present picture and present angle information deposited in the reference picture inventory.
On the other hand, if subsequent steps S660, then represent to have existed in the reference picture inventory reference picture with rotation angle information, therefore in one embodiment, if the rotation angle information that the 4th reference picture in the reference picture inventory has is same as present angle information, the rotation angle information of then directly deleting the 4th reference picture and having, and present picture and present angle information deposited in the reference picture inventory.In another embodiment, if the storage area is full, then delete employed reference picture in step S660, and correspondence deposits present picture and present angle information in.All the other situations directly deposit present picture and present angle information in then that storable position gets final product in the reference picture inventory.
In sum, the present invention can be according to the user by the corresponding camera lens of controlling image acquiring device of rotary manipulation to touch-screen, therefore the user does not need direct moving images deriving means, has simplified the degree of difficulty of operation for the video recording equipment that is applied to large-scale difficult movement.The shooting angle of camera lens between video pictures is not considered in the search of existing coupling block, therefore needs the more computational resource of cost and time for mobile estimation.Yet the present invention utilizes above-mentioned rotary manipulation can further calculate the rotary moving vector and uses the search of accelerating the coupling block, but so efficient and the accuracy of the estimation of lifting mobile amount.At last, be recorded in the reference picture inventory with reference to the picture rotation angle information corresponding with it simultaneously, the accuracy that has also improved the amount of movement estimation also can be in order to auxiliary compressed video picture.
It should be noted that at last: above each embodiment is not intended to limit only in order to technical scheme of the present invention to be described; Although with reference to aforementioned each embodiment the present invention is had been described in detail, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment puts down in writing, and perhaps some or all of technical characterictic wherein is equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution break away from the scope of various embodiments of the present invention technical scheme.
Claims (14)
1. a method of utilizing the compression of rotary manipulation auxiliary video is characterized in that be applicable to an image acquiring device, the method comprises the following steps:
Detect the rotary manipulation that this image acquiring device receives, and calculate the formed anglec of rotation of this rotary manipulation;
Utilize this anglec of rotation to rotate this image acquiring device taking a present picture, and this anglec of rotation is saved as a present angle information;
Judge in the stored reference picture inventory of this image acquiring device and whether have the reference picture with a rotation angle information; And
If not, the one rotary moving vector take this anglec of rotation as function calculation, and this present picture of differentiation is a plurality of present blocks, utilize the one first reference picture search of this rotary moving vector in this reference picture inventory to mate block corresponding to one of this present block respectively for this present block respectively, and compress according to this picture data of this present block and corresponding this coupling block.
2. the method for utilizing rotary manipulation auxiliary video compression according to claim 1 also comprises:
Upgrade this reference picture inventory, wherein:
This present picture is saved as one second reference picture in this reference picture inventory, and this present angle information is stored in this rotation angle information corresponding to this second reference picture.
3. the method for utilizing rotary manipulation auxiliary video compression according to claim 2, wherein:
If judge that the storage area of this reference picture inventory is full, certainly seek one three reference picture the most similar to this present angle information in this rotation angle information of this reference picture inventory; And
This rotation angle information of deleting the 3rd reference picture and having, and should present picture and this present angle information be stored in this reference picture inventory.
4. the method for utilizing rotary manipulation auxiliary video compression according to claim 1 if wherein judge and exist this reference picture with this rotation angle information in this reference picture inventory, also comprises:
Seek one the 4th reference picture in this reference picture inventory, wherein this rotation angle information of having of the 4th reference picture and the differential seat angle of this present angle information are less than a preset value;
Distinguishing the 4th reference picture is a plurality of reference block;
Calculate respectively a difference summation of the brightness value of a plurality of pixels in this present block respectively and corresponding this reference block, and respectively with threshold value comparison; And
If there is this difference summation of N present block to be not more than this threshold value, carry out a mobile estimating take the relative position of this present block as starting point, thereby mate block in the 4th reference picture search corresponding to this of this present block respectively, and compress according to this picture data of this present block and corresponding this coupling block, wherein N is the positive integer greater than 1.
5. the method for utilizing rotary manipulation auxiliary video compression according to claim 4 wherein also comprises after carrying out this mobile estimating:
Upgrade this reference picture inventory, wherein:
If this rotation angle information that the 4th reference picture has is same as this present angle information, this rotation angle information of directly deleting the 4th reference picture and having, and should present picture and this present angle information be stored in this reference picture inventory.
6. the method for utilizing rotary manipulation auxiliary video compression according to claim 1 wherein detects this rotary manipulation that this image acquiring device receives, and the step of calculating formed this anglec of rotation of this rotary manipulation comprises:
Detect the touching rotary manipulation that the touch-screen on this image acquiring device receives, and calculate formed this anglec of rotation of this touching rotary manipulation.
7. the method for utilizing rotary manipulation auxiliary video compression according to claim 1 wherein detects this rotary manipulation that this image acquiring device receives, and the step of calculating formed this anglec of rotation of this rotary manipulation comprises:
Detect the three-dimensional rotation operation that this image acquiring device receives by a displacement detection module, thereby calculate formed this anglec of rotation of this three-dimensional rotation operation.
8. an image acquiring device is characterized in that, comprising:
One camera lens module;
One detection module detects a rotary manipulation;
One memory cell stores a reference picture inventory; And
One control unit, couple this camera lens module, this detection module and this memory cell, calculate the formed anglec of rotation of this rotary manipulation, be rotated and take a present picture thereby utilize this anglec of rotation to control this camera lens module, and this anglec of rotation saved as a present angle information, wherein:
This control unit judges in this stored reference picture inventory of this memory cell whether have the reference picture with a rotation angle information, if not, the one rotary moving vector take this anglec of rotation as function calculation, and this present picture of differentiation is a plurality of present blocks, utilize this rotary moving vector in one first reference picture of this reference picture inventory, to search for corresponding to a coupling block of this present block respectively for this present block respectively, and compress according to this picture data of this present block and corresponding this coupling block.
9. image acquiring device according to claim 8, wherein:
This control unit upgrades this stored reference picture inventory of this memory cell, this present picture is stored in this memory cell with as one second reference picture in this reference picture inventory, and this present angle information is stored in this rotation angle information corresponding to this second reference picture.
10. image acquiring device according to claim 9, wherein:
If this control unit judges that in this memory cell the storage area in order to store this reference picture inventory is full, this control unit is sought one three reference picture the most similar to this present angle information in this rotation angle information of this reference picture inventory, delete this rotation angle information that the 3rd reference picture and the 3rd reference picture have, and should present picture and this present angle information be stored in this reference picture inventory.
11. image acquiring device according to claim 8, wherein this control unit comprises:
One amount of movement estimation unit, if this control unit is sought one the 4th reference picture in this reference picture inventory, wherein this rotation angle information of having of the 4th reference picture and the differential seat angle of this present angle information are less than a preset value, distinguishing the 4th reference picture is a plurality of reference block, calculate respectively a difference summation of the brightness value of a plurality of pixels in this present block respectively and corresponding this reference block, and compare with a threshold value respectively
Wherein, if there is this difference summation of N present block to be not more than this threshold value, a mobile estimating is carried out take the relative position of this present block as starting point in this amount of movement estimation unit, and provide carry out this mobile estimating gained corresponding to this coupling block of this present block respectively to this control unit, and compress according to this picture data of this present block and corresponding this coupling block by this control unit, wherein N is the positive integer greater than 1.
12. image acquiring device according to claim 11, wherein:
This control unit upgrades this stored reference picture inventory of this memory cell, if this rotation angle information that the 4th stored reference picture of this memory cell has is same as this present angle information, this rotation angle information that this control unit is directly deleted the 4th reference picture and had, and should present picture and this present angle information be stored in this reference picture inventory.
13. image acquiring device according to claim 8 also comprises:
One touch-screen couples this detection module, and this detection module detects the touching rotary manipulation that this touch-screen receives, and should touch rotary manipulation and offer this control unit, calculates according to this formed this anglec of rotation of this touching rotary manipulation.
14. image acquiring device according to claim 8 also comprises:
One displacement detection module couples this control unit, detects the three-dimensional rotation operation that this image acquiring device receives, thereby provides this control unit to calculate formed this anglec of rotation of this three-dimensional rotation operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210118876.7A CN103379258B (en) | 2012-04-20 | 2012-04-20 | Utilize method and the image acquiring device thereof of rotation process auxiliary video compress |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210118876.7A CN103379258B (en) | 2012-04-20 | 2012-04-20 | Utilize method and the image acquiring device thereof of rotation process auxiliary video compress |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103379258A true CN103379258A (en) | 2013-10-30 |
| CN103379258B CN103379258B (en) | 2016-08-03 |
Family
ID=49463783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210118876.7A Active CN103379258B (en) | 2012-04-20 | 2012-04-20 | Utilize method and the image acquiring device thereof of rotation process auxiliary video compress |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103379258B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1906948A (en) * | 2004-03-15 | 2007-01-31 | 三星电子株式会社 | Image coding apparatus and method for predicting motion using rotation matching |
| US20080219574A1 (en) * | 2007-03-06 | 2008-09-11 | Canon Kabushiki Kaisha | Image processing apparatus and image processing method |
| CN101727452A (en) * | 2008-10-22 | 2010-06-09 | 富士通株式会社 | Image processing method and device |
| WO2011150109A1 (en) * | 2010-05-26 | 2011-12-01 | Qualcomm Incorporated | Camera parameter- assisted video frame rate up conversion |
-
2012
- 2012-04-20 CN CN201210118876.7A patent/CN103379258B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1906948A (en) * | 2004-03-15 | 2007-01-31 | 三星电子株式会社 | Image coding apparatus and method for predicting motion using rotation matching |
| US20080219574A1 (en) * | 2007-03-06 | 2008-09-11 | Canon Kabushiki Kaisha | Image processing apparatus and image processing method |
| CN101727452A (en) * | 2008-10-22 | 2010-06-09 | 富士通株式会社 | Image processing method and device |
| WO2011150109A1 (en) * | 2010-05-26 | 2011-12-01 | Qualcomm Incorporated | Camera parameter- assisted video frame rate up conversion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103379258B (en) | 2016-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108596976B (en) | Method, device and equipment for relocating camera attitude tracking process and storage medium | |
| TWI511547B (en) | Method for assisting in video compression using rotation operation and image capturing device thereof | |
| CN110555883B (en) | Repositioning method and device for camera attitude tracking process and storage medium | |
| TWI769635B (en) | Network training pedestrian re-identification method and storage medium | |
| Wu et al. | Robust global motion estimation for video security based on improved k-means clustering | |
| CN100389601C (en) | Video electronic flutter-proof device | |
| CN100421455C (en) | Video electronic flutter-proof method | |
| TW202034270A (en) | Vehicle accident identification method and apparatus, and electronic device | |
| CN110782433B (en) | Dynamic information violent parabolic detection method and device based on time sequence and storage medium | |
| CN103763483A (en) | Method and device for shaking resistance in photo taking process of mobile terminal and mobile terminal | |
| WO2018063608A1 (en) | Place recognition algorithm | |
| WO2019183784A1 (en) | Method and electronic device for video recording | |
| CN111147787B (en) | Method for processing interpolation frame and related equipment | |
| CN110691259A (en) | Video playing method, system, device, electronic equipment and storage medium | |
| WO2021031790A1 (en) | Information processing method, apparatus, electronic device, storage medium, and program | |
| CN107087121A (en) | A kind of Autoconducting method and device based on motion detection | |
| CN112033351A (en) | Monocular camera-based distance measuring method and electronic equipment | |
| CN113691758A (en) | Frame insertion method and device, equipment and medium | |
| US7961907B2 (en) | Portable electronic device | |
| CN113438409B (en) | Delay calibration method, delay calibration device, computer equipment and storage medium | |
| WO2022016909A1 (en) | Method and apparatus for acquiring spatial distribution of wi-fi fingerprints, and electronic device | |
| CN111986229A (en) | Video target detection method, device and computer system | |
| CN114390188A (en) | Image processing method and electronic equipment | |
| CN103379258A (en) | Video compression method assisted by means of rotation operation and image acquisition device thereof | |
| EP3550817A1 (en) | Apparatus and method for associating images from two image streams |
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 |