CN107154022B - A kind of dynamic panorama mosaic method suitable for trailer - Google Patents
A kind of dynamic panorama mosaic method suitable for trailer Download PDFInfo
- Publication number
- CN107154022B CN107154022B CN201710325495.9A CN201710325495A CN107154022B CN 107154022 B CN107154022 B CN 107154022B CN 201710325495 A CN201710325495 A CN 201710325495A CN 107154022 B CN107154022 B CN 107154022B
- Authority
- CN
- China
- Prior art keywords
- image
- trailer
- pixel
- original image
- obtains
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000013441 quality evaluation Methods 0.000 claims abstract description 20
- 238000001914 filtration Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000012216 screening Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 5
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 239000004744 fabric Substances 0.000 claims 1
- 238000011156 evaluation Methods 0.000 abstract description 4
- 238000005457 optimization Methods 0.000 abstract description 3
- 238000012545 processing Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004438 eyesight Effects 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of dynamic panorama mosaic methods suitable for trailer, fully consider the profile and color information of stitching image, propose a kind of new image mosaic quality evaluating method, more meet human eye actual evaluation standard;The vibration of interference in view of to(for) image mosaic simultaneously carries out Kalman filtering to the quantized value of quality evaluation result, improves the robustness of this method;Using the quantized value of joining quality evaluation result as feedback, when camera relative position changes can real-time optimization splice parameter, finally realize the vehicle-mounted dynamic panoramic mosaic suitable for trailer based on multi-cam;Software-based image mosaic system can merge with the vehicle-mounted splicing system of looking around of tradition, and then make up its disadvantage too strong to hardware-dependence, on-line proving splicing system and can improve joining quality.
Description
Technical field
The invention belongs to computer visions and field of image processing, are related to a kind of dynamic panoramic mosaic side suitable for trailer
Method.
Background technique
With advances in technology with development, people are more and more for the demand of information, wherein most information come from
In vision;But due to the limited viewing angle of human eye, people start with computer research Panorama Mosaic technology, that is, use image
The mode of splicing is by several image mosaics with overlapping region collected from Same Scene using multiple cameras at one
Width large-size images;And panoramic video splicing is then an extension of Panorama Mosaic, it can be by the result of image mosaic
Real-time display comes out, and possesses extensively in video monitoring, rail traffic, video conference, virtual reality, the vehicle-mounted numerous areas such as look around
Application prospect.
For current vehicle-mounted splicing system of looking around primarily directed to integrated vehicle, i.e. headstock vehicle body is relatively fixed.System acquires simultaneously
The image information of vehicle periphery is spliced into a width panorama sketch, intuitively shows the location of vehicle after once demarcating
With periphery situation.Driver has greatly been expanded to the sensing capability of surrounding and environment, the hair of traffic accident can be effectively reduced
It is raw.And for the Trailer equipment with chain structure, since camera is typically distributed across on headstock and vehicle body, and work as Trailer equipment
When headstock and vehicle body relative motion, the positional relationship between camera is also changed correspondingly, and real-time update is needed to splice parameter, tradition
Vehicle-mounted splicing system of looking around cannot then be applicable in.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of dynamic panorama mosaic methods suitable for chain type trailer, it is intended to
Realize that the panoramic video occurred under dynamic changes in camera relative position splices using existing image processing techniques.
A kind of dynamic panorama mosaic method suitable for trailer arranges camera around the headstock and vehicle body of the trailer,
The image sequence of camera acquisition is obtained in real time, wherein is included the following steps: to the joining method of image
Step 1: the original sequence that acquisition multi-cam is shot, and initialize homography matrix;
Step 2: to original sequence carry out distortion correction and resolution decreasing pretreatment, finally to image sequence into
The processing of row cylindrical surface projecting, obtains the projection model image suitable for panorama sketch;
Step 3: adjacent two width projection model image alignment into the same coordinate system and is aligned image according to homography matrix
Image co-registration is carried out, stitching image is obtained;Then splicing effect is quantified, specifically:
S31, the image obtained for step 1 carry out contours extract, obtain original image Io, spelling that step 3 is obtained
Map interlinking picture carries out contours extract, and finds original image I whereinoCorresponding part Ip;
S32, it is directed to original image IoIn each pixel O, in corresponding part IpCorresponding position pixel X is found, is judged
It whether there is point identical with the pixel value of pixel O in the setting neighborhood of pixel X, if it is present thinking to find and picture
The matched point of vegetarian refreshments O;If it does not exist, then thinking not find and the matched point of pixel O;Profile Duplication are as follows:
Wherein sum (Id) indicate original image IoNumber of pixels and corresponding part IpThe smaller value of middle number of pixels, numel
(Id) indicate in image IpIn find and original image IoThe middle matched number of pixel;
S33, the channel image zooming-out H that step 1 is obtained information, obtained image definition be original image Ho;To institute
After stating the information that stitching image extracts the channel H, and found and original image H in stitching imageoCorresponding part, is defined as Hp;
Calculate original image HoWith corresponding part HpSimilarity:
Wherein Ho(i, j) indicates original image HoMiddle coordinate is the pixel value at (i, j), Hp(i, j) indicates corresponding part Hp
Middle coordinate is the pixel value at (i, j), and M and N are HoHeight and the width on number of pixels;
S34, the profile Duplication of S32 is multiplied with the similarity of S33, obtains image mosaic quality evaluation index H'=α
β;Then Kalman filtering is done to joining quality evaluation index H', finally obtains joining quality evaluation index H;
Step 4: the quality evaluation index H that step 3 obtains is compared with the threshold value of setting, spelled if comparing display
It connects quality and reaches expected, image after output splicing, and return step one;Otherwise next step is carried out;
Step 5: carrying out characteristic point detection to the projection model image that step 2 obtains, and adjacent image is carried out special
Sign point matching, recalculates homography matrix, and return step three executes step 3 to step 4 according to the homography matrix of update.
Preferably, the projection model in the step 1 is cylinder model, Sphere Measurement Model or cube model.
Preferably, contours extract is carried out in S31 by the way of convolution, wherein the convolution kernel of use are as follows:
Preferably, in step 2 Image Feature Point Matching process are as follows: splicing regions characteristic point detection, Feature Points Matching with
And match point screening.
Preferably, the feature point detecting method is SIFT, SURF or ORB.
Preferably, the characteristic point screening technique is RANSAC, consistent symmetry is screened, arest neighbors or secondary neighbour's ratio sieve
Choosing.
Preferably, the size for setting neighborhood is 3 × 3.
The invention has the following beneficial effects:
1. the present invention fully considers the profile and color information of stitching image, proposes the new image mosaic quality of one kind and comment
Valence method more meets human eye actual evaluation standard;The vibration of interference in view of to(for) image mosaic simultaneously, to quality evaluation result
Quantized value carry out Kalman filtering, improve the robustness of this method.
2. the present invention is using the quantized value of joining quality evaluation result as feedback, when camera relative position changes
Can real-time optimization splice parameter, finally realize the vehicle-mounted dynamic panoramic mosaic suitable for trailer based on multi-cam.
3. the system is software-based image mosaic system, can be merged with the vehicle-mounted splicing system of looking around of tradition, in turn
Its disadvantage too strong to hardware-dependence is made up, on-line proving splicing system and joining quality can be improved.
Detailed description of the invention
Fig. 1, dynamic camera head splicing system flow chart;
Fig. 2, joining quality evaluation module flow chart;
Fig. 3, Kalman filtering flow chart;
Fig. 4, the trailer schematic diagram with chain structure;
Fig. 5, camera group installation site and observation area schematic diagram;
Evaluation result under Fig. 6, different degree of misalignment;
Image mosaic quality evaluation result comparison under Fig. 7, camera shock conditions.
Wherein, 1- camera, 2- headstock, 3- vehicle body.
Specific embodiment
The present invention will now be described in detail with reference to the accompanying drawings and examples.
As shown in figure 4, the present invention is based on chain type trailer, configured with power supply, industrial personal computer, camera 1 and dynamic in trailer
State video-splicing host computer etc..It (1) is input equipment, that is, USB camera 1 of whole device, by the way that camera 1 is fixed on vehicle
The different parts of body 3 and headstock 2 can observe the environmental information around car body on a large scale in real time;It (2) is headstock 2, control
It makes the direction of motion of vehicle, equipped with 4 cameras 1 on headstock 2, the environmental information of vehicle front can be obtained in real time, install
As shown in abcd in Fig. 5, grey area is the observation area of camera 1 for position;Industrial personal computer is installed in the car, is taken the photograph with USB
Picture head 1 connects, and after power supply power supply, can read the image sequence of the acquisition of camera 1 in real time.Host computer can be acquired with real-time display
Obtained image sequence and spliced panoramic picture, while the work such as the setting of parameter, video storage can be completed.(3) it is
Vehicle body 3, since trailer vehicle body 3 is mostly long, and the observation scope of each camera 1 is limited, in order to reduce the dead of observation
Angle needs the quantity for determining to carry camera 1 according to the length of vehicle body 3, and vehicle body 3 is equipped with 6 cameras 1, installation position in figure
It sets as shown in efghij in Fig. 5;The headstock 2 of trailer and vehicle body 3 are to be connected together through the hinges, d camera and e in Fig. 3
Dynamic change can occur for camera relative positional relationship during trailer turns to, and therefore, the present invention provides one kind to be applicable in
In the dynamic panorama mosaic method of trailer, panorama sketch can be obtained in real time.
Method process of the invention mainly include 1 Image Acquisition of multi-cam, image preprocessing, Image Feature Point Matching,
Homography matrix calculates, image co-registration and joining quality are evaluated, as shown in Figure 1, specifically comprising the following steps:
Step 1: system initialization, starting and initialization, the connection of camera 1 and industrial personal computer, dynamic including industrial personal computer
The initialization of video-splicing parameter.
Step 2: host computer reads the image that camera 1 acquires.If read successfully, enter next step, otherwise after
It is continuous to carry out the step.
Step 3: image preprocessing.It distorts since camera 1 exists, is unfavorable for splicing, distortion school is carried out to image first
Just;Simultaneously for speed up processing, gray level image is converted by color image, resolution decreasing processing is carried out to image;Finally
Image sequence is handled, the projection model suitable for panorama sketch is obtained;Wherein projection model can be but not limited to cylinder
Model, Sphere Measurement Model, cube model, the purpose of the step reduce the splicing time while being to improve joining quality, improve
Stitching algorithm real-time;
Step 4: detecting the characteristic point of image to be spliced, and Feature Points Matching is carried out to adjacent image, calculates two width figures
Join outside the matching of picture, i.e. homography matrix, is snapped to adjacent two images in the same coordinate system according to homography matrix.Wherein, scheme
As Feature Points Matching includes the detection of splicing regions characteristic point, Feature Points Matching, match point screening etc., wherein feature point detecting method
It can be but not limited to SIFT, SURF, ORB, screening technique has RANSAC, consistent symmetry screening, arest neighbors and time neighbour's ratio
Rate screening;
Step 5: the alignment image that the 4th step is obtained carries out image co-registration according to specific blending algorithm, spliced
Then image quantifies splicing effect, specifically:
Image after the splicing that the original image and the 5th step obtained for third step obtains, first extraction profile information
And its Duplication is detected, if the profile information Duplication of image is higher after original image and splicing, illustrate image mosaic matter
It measures higher.The present invention carries out edge extracting by the convolution kernel of 5 × 5 sizes of design, and convolution kernel can be, but not limited to
Each width original image I obtained for third stepo, original image I is found in image after splicingoCorresponding part,
By original image IoWith the profile information image I of corresponding partpDifference information is sought, difference image I is obtainedd, specifically: it is directed to
Original image IoEach pixel O finds corresponding position pixel X in the corresponding part of stitching image, judges the n* of pixel X
It whether there is point identical with the pixel value of pixel O in the region n, if it is present thinking to find matched with pixel O
Point, is denoted as 1;If it does not exist, then think not find with the matched point of pixel O, be denoted as 0, specific formula is expressed as follows:
Then Duplication calculation method is
Wherein sum (Id) indicate original image IoWith the profile information image I of corresponding partpMiddle number of pixels smaller value, table
Show in image IpIn find and original image IoThe middle matched number of pixel.
For image, HSV space can more intuitively express the light and shade of color, tone and fresh than rgb space
Gorgeous degree, the variation in face of illumination are more stable;The wherein form and aspect information of H channel table diagram picture, therefore can reaction color well
Essence.Image is transformed into HSV space, the stitching image that the original image and the 5th step obtain to third step obtains mentions respectively
The information for taking the channel H obtains original image HoWith part H corresponding with original image in stitching imagep, calculate original image Ho
With corresponding part HpSimilarity, similarity is higher, then illustrates that image joining quality is higher.The calculation method of similarity are as follows:
Wherein Ho(i, j) indicates original image HoMiddle coordinate is the pixel value at (i, j), Hp(i, j) indicates corresponding part Hp
Middle coordinate is the pixel value at (i, j), and M and N are HoHeight and the width.
In order to more accurately evaluate image mosaic quality, this method has fully considered the profile and form and aspect information pair of image
Profile Duplication is multiplied by influence as a result with similarity, obtains image mosaic quality evaluation index H'=α β.In practical application
In the process, Trailer equipment can occur acutely to shake in bumpy sections, and 1 relative position of camera is caused to occur continuously irregularly to change
Become, but the case where this 1 position acute variation of the camera as caused by road bump is equivalent to one for panorama system
External disturbance, it will cause image quality evaluation index and is acutely irregularly widely varied, and then lead to the repetition for splicing parameter
It calculates, that is, increases calculating cost, and splicing effect cannot be improved.To avoid bumpy sections to the shadow of image mosaic quality evaluation
It rings, the present invention does Kalman filtering in response to this, to joining quality evaluation index, reduces its degree of jitter.Kalman's filter
The schematic diagram of wave is as shown in figure 3, the effect picture before and after Kalman filtering is added is as shown in Figure 7.Finally obtain joining quality evaluation
Index H.
Step 6: the quality evaluation index H that step 5 obtains is compared with the threshold value of setting, spelled if comparing display
It connects quality and reaches expected, image after output splicing, and return to third step;Otherwise next step is carried out;
Step 7: recalculating homography matrix.Since the quality of stitching image is lower than expection, i.e., joining quality, which is not met, wants
It asks, illustrates that the relative position of truck camera 1 is changed, it is therefore desirable to recalculate homography matrix.First to image into
The detection of row feature and extraction, then carry out characteristic matching, calculate homography matrix, and return to the 5th step, re-start melting for image
It closes.
Present invention employs anastomosing and splicing image outlines and the image mosaic quality evaluating method of color information to spliced map
As effect real-time perfoming quantitatively evaluating, evaluation effect when joining quality is less than the threshold value of setting as shown in fig. 6, need again
Optimization splicing parameter, splicing when this method is to picture runing have certain robustness;Splicing parameter is mainly the outer of camera
Ginseng, related with the relative position of camera 1, outer ginseng can use the calculated homography matrix of Image Feature Point Matching and obtain.
In conclusion the above is merely preferred embodiments of the present invention, being not intended to limit the scope of the present invention.
All within the spirits and principles of the present invention, any modification, equivalent replacement, improvement and so on should be included in of the invention
Within protection scope.
Claims (7)
1. a kind of dynamic panorama mosaic method suitable for trailer, which is characterized in that cloth around the headstock and vehicle body of the trailer
Camera is set, obtains the image sequence of camera acquisition in real time, wherein include the following steps: to the joining method of image
Step 1: the original sequence that acquisition multi-cam is shot, and initialize homography matrix;
Step 2: carrying out the pretreatment of distortion correction and resolution decreasing to original sequence, column finally is carried out to image sequence
Face projection process obtains the projection model image suitable for panorama sketch;
Step 3: according to homography matrix by adjacent two width projection model image alignment into the same coordinate system and to alignment image into
Row image co-registration, obtains stitching image;Then splicing effect is quantified, specifically:
S31, the image obtained for step 1 carry out contours extract, obtain original image Io, stitching image that step 3 is obtained
Contours extract is carried out, and finds original image I whereinoCorresponding part Ip;
S32, it is directed to original image IoIn each pixel O, in corresponding part IpCorresponding position pixel X is found, judges pixel
It whether there is point identical with the pixel value of pixel O in the setting neighborhood of point X, if it is present thinking to find and pixel O
Matched point;If it does not exist, then thinking not find and the matched point of pixel O;Profile Duplication are as follows:
Wherein sum (Id) indicate original image IoNumber of pixels and corresponding part IpThe smaller value of middle number of pixels, numel (Id) table
Show in image IpIn find and original image IoThe middle matched number of pixel;
S33, the channel image zooming-out H that step 1 is obtained information, obtained image definition be original image Ho;To the spelling
After the information for connecing the channel image zooming-out H, and found and original image H in stitching imageoCorresponding part, is defined as Hp;It calculates
Original image HoWith corresponding part HpSimilarity:
Wherein Ho(i, j) indicates original image HoMiddle coordinate is the pixel value at (i, j), Hp(i, j) indicates corresponding part HpMiddle seat
The pixel value being designated as at (i, j), M and N are HoHeight and the width on number of pixels;
S34, the profile Duplication of S32 is multiplied with the similarity of S33, obtains image mosaic quality evaluation index H'=α β;So
Kalman filtering is done to joining quality evaluation index H' afterwards, finally obtains joining quality evaluation index H;
Step 4: the quality evaluation index H that step 3 obtains is compared with the threshold value of setting, if comparing display splicing matter
Amount reaches expected, image after output splicing, and return step one;Otherwise next step is carried out;
Step 5: carrying out characteristic point detection to the projection model image that step 2 obtains, and characteristic point is carried out to adjacent image
Matching, recalculates homography matrix, and return step three executes step 3 to step 4 according to the homography matrix of update.
2. a kind of dynamic panorama mosaic method suitable for trailer as described in claim 1, which is characterized in that the step 2
In projection model be cylinder model, Sphere Measurement Model or cube model.
3. a kind of dynamic panorama mosaic method suitable for trailer as described in claim 1, which is characterized in that used in S31
The mode of convolution carries out contours extract, wherein the convolution kernel of use are as follows:
4. a kind of dynamic panorama mosaic method suitable for trailer as described in claim 1, which is characterized in that scheme in step 5
As the process of Feature Points Matching are as follows: the detection of splicing regions characteristic point, Feature Points Matching and match point screening.
5. a kind of dynamic panorama mosaic method suitable for trailer as claimed in claim 4, which is characterized in that the characteristic point
Detection method is SIFT, SURF or ORB.
6. a kind of dynamic panorama mosaic method suitable for trailer as described in claim 4 or 5, which is characterized in that the spy
Sign point screening technique is RANSAC, consistent symmetry is screened, arest neighbors or secondary neighbour's ratio screen.
7. a kind of dynamic panorama mosaic method suitable for trailer as described in claim 1, which is characterized in that the setting is adjacent
The size in domain is 3 × 3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710325495.9A CN107154022B (en) | 2017-05-10 | 2017-05-10 | A kind of dynamic panorama mosaic method suitable for trailer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710325495.9A CN107154022B (en) | 2017-05-10 | 2017-05-10 | A kind of dynamic panorama mosaic method suitable for trailer |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107154022A CN107154022A (en) | 2017-09-12 |
CN107154022B true CN107154022B (en) | 2019-08-27 |
Family
ID=59794275
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710325495.9A Active CN107154022B (en) | 2017-05-10 | 2017-05-10 | A kind of dynamic panorama mosaic method suitable for trailer |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107154022B (en) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102017123452A1 (en) * | 2017-10-10 | 2019-04-11 | Connaught Electronics Ltd. | Method for generating an output image, camera system and motor vehicle showing a motor vehicle and a surrounding area of the motor vehicle in a predetermined target view |
CN108093188B (en) * | 2017-12-25 | 2019-01-25 | 北京航空航天大学 | A method of the big visual field video panorama splicing based on hybrid projection transformation model |
CN108263283B (en) * | 2018-01-25 | 2020-01-14 | 长沙立中汽车设计开发股份有限公司 | Method for calibrating and splicing panoramic all-round looking system of multi-marshalling variable-angle vehicle |
CN108665501A (en) * | 2018-07-12 | 2018-10-16 | 合肥晟泰克汽车电子股份有限公司 | Automobile viewing system three-dimensional scaling scene and the scaling method for using the scene |
CN109688329B (en) * | 2018-12-24 | 2020-12-11 | 天津天地伟业信息系统集成有限公司 | Anti-shake method for high-precision panoramic video |
CN110223226B (en) * | 2019-05-07 | 2021-01-15 | 中国农业大学 | Panoramic image splicing method and system |
CN112184545A (en) * | 2019-07-05 | 2021-01-05 | 杭州海康威视数字技术股份有限公司 | Vehicle-mounted ring view generating method, device and system |
CN110706282B (en) * | 2019-10-31 | 2020-09-15 | 镁佳(北京)科技有限公司 | Automatic calibration method and device for panoramic system, readable storage medium and electronic equipment |
CN110719411B (en) * | 2019-12-16 | 2020-04-03 | 长沙智能驾驶研究院有限公司 | Panoramic all-around view image generation method of vehicle and related equipment |
CN111161404B (en) * | 2019-12-23 | 2023-05-09 | 华中科技大学鄂州工业技术研究院 | Annular scanning morphology three-dimensional reconstruction method, device and system |
CN111546986A (en) * | 2020-04-30 | 2020-08-18 | 北京大椽科技有限公司 | Trailer panoramic looking-around method |
CN111959511B (en) * | 2020-08-26 | 2022-06-03 | 腾讯科技(深圳)有限公司 | Vehicle control method and device |
CN112669219A (en) * | 2021-01-21 | 2021-04-16 | 深圳市爱培科技术股份有限公司 | Three-dimensional 360-degree panoramic image splicing method and device |
CN112669218A (en) * | 2021-01-21 | 2021-04-16 | 深圳市爱培科技术股份有限公司 | Planar 360-degree panoramic image splicing method and device |
CN114187334B (en) * | 2021-10-12 | 2022-09-13 | 武汉兰丁云医学检验实验室有限公司 | Adjacent slice image superposition and alignment method based on HE staining, Ki67 and P16 combination |
CN115205110A (en) * | 2022-05-30 | 2022-10-18 | 浙江大华技术股份有限公司 | Panoramic image splicing method and device, electronic device and storage medium |
CN115239724B (en) * | 2022-09-22 | 2022-11-22 | 鹰驾科技(深圳)有限公司 | 360-degree panoramic stitching image analysis and evaluation method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105447865A (en) * | 2015-11-23 | 2016-03-30 | 深圳进化动力数码科技有限公司 | Method and device for evaluating static splicing quality of panoramic splicing algorithm |
CN105469387A (en) * | 2015-11-13 | 2016-04-06 | 深圳进化动力数码科技有限公司 | Quantification method and quantification device for splicing quality |
CN105957007A (en) * | 2016-05-05 | 2016-09-21 | 电子科技大学 | Image stitching method based on characteristic point plane similarity |
-
2017
- 2017-05-10 CN CN201710325495.9A patent/CN107154022B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105469387A (en) * | 2015-11-13 | 2016-04-06 | 深圳进化动力数码科技有限公司 | Quantification method and quantification device for splicing quality |
CN105447865A (en) * | 2015-11-23 | 2016-03-30 | 深圳进化动力数码科技有限公司 | Method and device for evaluating static splicing quality of panoramic splicing algorithm |
CN105957007A (en) * | 2016-05-05 | 2016-09-21 | 电子科技大学 | Image stitching method based on characteristic point plane similarity |
Non-Patent Citations (1)
Title |
---|
图像拼接质量评价方法;万国挺 等;《通信学报》;20130831;第34卷(第8期);第76-81页 * |
Also Published As
Publication number | Publication date |
---|---|
CN107154022A (en) | 2017-09-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107154022B (en) | A kind of dynamic panorama mosaic method suitable for trailer | |
KR102003015B1 (en) | Creating an intermediate view using an optical flow | |
Aires et al. | Optical flow using color information: preliminary results | |
US20170345214A1 (en) | High Resolution (HR) Panorama Generation Without Ghosting Artifacts Using Multiple HR Images Mapped to a Low-Resolution 360-Degree Image | |
CN102507592B (en) | Fly-simulation visual online detection device and method for surface defects | |
CN109407547A (en) | Multi-cam assemblage on-orbit test method and system towards panoramic vision perception | |
CN108230245B (en) | Image splicing method, image splicing device and electronic equipment | |
CN111462128B (en) | Pixel-level image segmentation system and method based on multi-mode spectrum image | |
CN105957007A (en) | Image stitching method based on characteristic point plane similarity | |
CN112184604B (en) | Color image enhancement method based on image fusion | |
CN103679749A (en) | Moving target tracking based image processing method and device | |
WO2021098567A1 (en) | Method and apparatus for generating panorama having depth information, and storage medium | |
CN109934772A (en) | A kind of image interfusion method, device and portable terminal | |
CN105869115B (en) | A kind of depth image super-resolution method based on kinect2.0 | |
CN106709894B (en) | Image real-time splicing method and system | |
CN206249426U (en) | A kind of image restoration system | |
CN105488777A (en) | System and method for generating panoramic picture in real time based on moving foreground | |
CN106780297A (en) | Image high registration accuracy method under scene and Varying Illumination | |
CN109803172A (en) | A kind of processing method of live video, device and electronic equipment | |
Asmare et al. | Image enhancement by fusion in contourlet transform | |
Lv et al. | An integrated enhancement solution for 24-hour colorful imaging | |
Fan et al. | Multiscale cross-connected dehazing network with scene depth fusion | |
CN111292228A (en) | Lens defect detection method | |
CN110472527A (en) | A kind of track line detecting method and system | |
CN109886195A (en) | Skin identification method based on depth camera near-infrared single color gradation figure |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |