CN108921776A - A kind of image split-joint method and device based on unmanned plane - Google Patents
A kind of image split-joint method and device based on unmanned plane Download PDFInfo
- Publication number
- CN108921776A CN108921776A CN201810556862.0A CN201810556862A CN108921776A CN 108921776 A CN108921776 A CN 108921776A CN 201810556862 A CN201810556862 A CN 201810556862A CN 108921776 A CN108921776 A CN 108921776A
- Authority
- CN
- China
- Prior art keywords
- image
- point
- characteristic point
- spliced
- unmanned plane
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000009466 transformation Effects 0.000 claims abstract description 37
- 239000011159 matrix material Substances 0.000 claims abstract description 32
- 238000003708 edge detection Methods 0.000 claims abstract description 19
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 230000007246 mechanism Effects 0.000 claims abstract description 8
- 239000000284 extract Substances 0.000 claims abstract description 7
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 1
- 230000007613 environmental effect 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
- G06T3/14—Transformations for image registration, e.g. adjusting or mapping for alignment of images
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to air vehicle technique fields, disclose a kind of image split-joint method and device based on unmanned plane, and this method includes：Image to be spliced is obtained, the image to be spliced includes the first image and the second image；Extract the characteristic point of the image to be spliced；To the characteristic point of the first image and the second image to slightly being matched；Edge detection is carried out using edge detection algorithm；Remove the characteristic point pair of error hiding；Splice the first image and the second image using the syncretizing mechanism gradually gone out is fade-in；It is matched by the characteristics of image of extraction, estimates the transformation matrix between image, be then aligned image using transformation matrix, effectively eliminate " ghost " phenomenon of image, matching speed and splicing speed are improved, improves splicing accuracy rate, hsrdware requirements are lower.
Description
Technical field
The present invention relates to air vehicle technique field more particularly to a kind of image split-joint methods and device based on unmanned plane.
Background technique
Due to being influenced by environmental factor when shooting, the limited area that the image of unmanned plane shooting is covered, image
It is second-rate, people are unable to satisfy to high-resolution, the demand of wide viewing angle, and the hardware device for obtaining panoramic picture is very high
It is expensive.Therefore, it is necessary to splice to image, range of widening the vision increases the information content of image, improves its clarity.And it is existing
Technology splicing speed is lower, and accuracy rate is not high.
Summary of the invention
It is a primary object of the present invention to propose a kind of image split-joint method and device based on unmanned plane, pass through extraction
Characteristics of image is matched, and the transformation matrix between image is estimated, and is then aligned image using transformation matrix, is effectively eliminated
" ghost " phenomenon of image improves matching speed and splicing speed, improves splicing accuracy rate, hsrdware requirements are lower.
To achieve the above object, a kind of image split-joint method based on unmanned plane provided by the invention, including：
Image to be spliced is obtained, the image to be spliced includes the first image and the second image；
Extract the characteristic point of the image to be spliced；
To the characteristic point of the first image and the second image to slightly being matched；
Edge detection is carried out using edge detection algorithm；
Remove the characteristic point pair of error hiding；
Splice the first image and the second image using the syncretizing mechanism gradually gone out is fade-in.
Optionally, the characteristic point for extracting the image to be spliced includes：
Gray processing is carried out to the first image and the second image；
To the first image and the second image progress Multidimensional Discrete wavelet transformation after gray processing, low-frequency data and height are extracted
Frequency evidence；
The low-frequency data is divided into several image blocks of default size, by each pixel of each image block
Rgb value is as characteristic information.
Optionally, the characteristic point to the first image and the second image includes to thick matching is carried out：
The matching similarity of the first image and the second image correspondence image block is calculated, the highest region of similarity is selected
As picture registration region；
Characteristic point is carried out using BRISK algorithm slightly to match.
Optionally, the characteristic point of the removal error hiding is to including：
3 characteristic points pair not on the same line are randomly selected, transformation matrix is calculated；
It calculates each characteristic point and passes through distance of the matrixing to character pair point, the distance obtained using adaptive approach
Threshold value is by characteristic point to being divided into interior point and exterior point；
When the quantity of interior point is greater than preset amount threshold, the transformation matrix is re-evaluated with all interior points；
When the quantity of interior point is not more than the preset amount threshold, according to the interior points and error rate estimation
Transformation matrix.
As another aspect of the present invention, a kind of image splicing device based on unmanned plane provided, including：
Image collection module, for obtaining image to be spliced, the image to be spliced includes the first image and second
Image；
Feature point extraction module, for extracting the characteristic point of the image to be spliced；
Matching module, for the characteristic point to the first image and the second image to slightly being matched；
Edge detection module, for carrying out edge detection using edge detection algorithm；
Module is removed, for removing the characteristic point pair of error hiding；
Splicing module, for using the syncretizing mechanism splicing the first image and the second image being fade-in gradually out.
Optionally, the feature point extraction module includes：
Gray processing unit, for carrying out gray processing to the first image and the second image；
Wavelet transform unit, for after gray processing the first image and the second image carry out Multidimensional Discrete wavelet transformation,
Extract low-frequency data and high-frequency data；
Extraction unit, for the low-frequency data to be divided into several image blocks of default size, by each image block
The rgb value of each pixel is as characteristic information.
Optionally, the matching module includes：
Similarity calculated, for calculating the matching similarity of the first image and the second image correspondence image block,
Select the highest region of similarity as picture registration region；
Matching unit is slightly matched for carrying out characteristic point using BRISK algorithm.
Optionally, the removal module includes：
Selection unit calculates transformation matrix for randomly selecting 3 characteristic points pair not on the same line；
Metrics calculation unit, for calculating distance of each characteristic point by matrixing to character pair point, using certainly
The distance threshold that adaptive method obtains is by characteristic point to being divided into interior point and exterior point；
First evaluation unit, for being re-evaluated with all interior points when the quantity of interior point is greater than preset amount threshold
The transformation matrix；
Second evaluation unit, for when the quantity of interior point be not more than the preset amount threshold when, according to the interior point
Several and error rate estimates the transformation matrix.
A kind of image split-joint method and device based on unmanned plane proposed by the present invention, this method include：It obtains to be spliced
Image, the image to be spliced include the first image and the second image；Extract the characteristic point of the image to be spliced；It is right
The first image and the characteristic point of the second image are to slightly being matched；Edge detection is carried out using edge detection algorithm；Removal
The characteristic point pair of error hiding；Splice the first image and the second image using the syncretizing mechanism gradually gone out is fade-in；Pass through extraction
Characteristics of image is matched, and the transformation matrix between image is estimated, and is then aligned image using transformation matrix, is effectively eliminated
" ghost " phenomenon of image improves matching speed and splicing speed, improves splicing accuracy rate, hsrdware requirements are lower.
Detailed description of the invention
Fig. 1 is a kind of flow chart for image split-joint method based on unmanned plane that the embodiment of the present invention one provides；
Fig. 2 is the method flow diagram of step S20 in Fig. 1；
Fig. 3 is the method flow diagram of step S30 in Fig. 1；
Fig. 4 is the method flow diagram of step S50 in Fig. 1；
Fig. 5 is a kind of demonstrative structure frame of the image splicing device based on unmanned plane provided by Embodiment 2 of the present invention
Figure；
Fig. 6 is the exemplary block diagram of feature point extraction module in Fig. 5；
Fig. 7 is the exemplary block diagram of matching module in Fig. 5；
Fig. 8 is the exemplary block diagram that module is removed in Fig. 5.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Embodiment
As shown in Figure 1, in the present embodiment, a kind of image split-joint method based on unmanned plane, including：
S10, image to be spliced is obtained, the image to be spliced includes the first image and the second image；
S20, the characteristic point for extracting the image to be spliced；
S30, to the characteristic point of the first image and the second image to slightly being matched；
S40, edge detection is carried out using edge detection algorithm；
S50, the characteristic point pair for removing error hiding；
S60, the first image and the second image are spliced using the syncretizing mechanism being fade-in gradually out.
In the present embodiment, it is matched by the characteristics of image of extraction, estimates the transformation matrix between image, it is then sharp
It is aligned image with transformation matrix, effectively eliminates " ghost " phenomenon of image, matching speed and splicing speed is improved, improves
Splicing accuracy rate, hsrdware requirements are lower.
In the present embodiment, image to be spliced is two width or multiple image of Same Scene, when image to be spliced is two width
When, spliced according to above-mentioned joining method, when image to be spliced is multiple image, is carried out two-by-two according to above-mentioned joining method
Splicing.
As shown in Fig. 2, in the present embodiment, the step S20 includes：
S21, gray processing is carried out to the first image and the second image；
S22, to after gray processing the first image and the second image carry out Multidimensional Discrete wavelet transformation, extract low-frequency data
And high-frequency data；
S23, several image blocks that the low-frequency data is divided into default size, by each pixel of each image block
Rgb value as characteristic information.
In the present embodiment, low-frequency data is the content information for characterizing image, and high-frequency data is the edge letter for characterizing image
Breath is mainly divided into fixed size (comprising 1*M picture to low-frequency data using the unequal matching similarity method based on region
Element) n image block, using the rgb value of each piece of each pixel as characteristic information.
As shown in figure 3, in the present embodiment, the step S30 includes：
S31, the matching similarity for calculating the first image and the second image correspondence image block select similarity highest
Region is as picture registration region；
S32, it is slightly matched using BRISK algorithm progress characteristic point.
In the present embodiment, BRISK algorithm is on ICCV in 2011《BRISK:Binary Robust Invariant
Scalable Keypoints》A kind of feature extraction algorithm and a kind of binary feature description put forward in article is calculated
Son.It has preferable rotational invariance, scale invariability, preferable robustness etc..In image registration application, speed ratio
Compared with：SIFT<SURF<BRISK<FREAK<ORB, when to there is larger fuzzy image registration, BRISK algorithm shows most wherein
It is outstanding.
As shown in figure 4, in the present embodiment, the step S50 includes：
S51,3 characteristic points pair not on the same line are randomly selected, calculates transformation matrix；
S52, each characteristic point is calculated by matrixing to the distance of character pair point, obtained using adaptive approach
Distance threshold is by characteristic point to being divided into interior point and exterior point；
S53, when the quantity of interior point be greater than preset amount threshold when, re-evaluate the transformation matrix with all interior points；
S54, when the quantity of interior point be not more than the preset amount threshold when, according to it is described it is interior points and error rate estimate
The transformation matrix.
In the present embodiment, shown characteristic point is to match point is referred to as, if the feature of the first image and the second image
Point is to respectively：X=[x, y]TWith X '=[x ', y ']T, the two meets formula：
X '=HX
H is the transformation matrix of 6 parameters in formula, then above formula can be expressed as again：
Wherein,For transformation matrix H.
Embodiment two
As shown in figure 5, in the present embodiment, a kind of image splicing device based on unmanned plane, including：
Image collection module 10, for obtaining image to be spliced, the image to be spliced includes the first image and the
Two images；
Feature point extraction module 20, for extracting the characteristic point of the image to be spliced；
Matching module 30, for the characteristic point to the first image and the second image to slightly being matched；
Edge detection module 40, for carrying out edge detection using edge detection algorithm；
Module 50 is removed, for removing the characteristic point pair of error hiding；
Splicing module 60, for using the syncretizing mechanism splicing the first image and the second image being fade-in gradually out.
In the present embodiment, it is matched by the characteristics of image of extraction, estimates the transformation matrix between image, it is then sharp
It is aligned image with transformation matrix, effectively eliminates " ghost " phenomenon of image, matching speed and splicing speed is improved, improves
Splicing accuracy rate, hsrdware requirements are lower.
In the present embodiment, image to be spliced is two width or multiple image of Same Scene, when image to be spliced is two width
When, spliced according to above-mentioned joining method, when image to be spliced is multiple image, is carried out two-by-two according to above-mentioned joining method
Splicing.
As shown in fig. 6, in the present embodiment, the feature point extraction module includes：
Gray processing unit 21, for carrying out gray processing to the first image and the second image；
Wavelet transform unit 22, for after gray processing the first image and the second image carry out the change of Multidimensional Discrete small echo
It changes, extracts low-frequency data and high-frequency data；
Extraction unit 23, for the low-frequency data to be divided into several image blocks of default size, by each image block
Each pixel rgb value as characteristic information.
In the present embodiment, low-frequency data is the content information for characterizing image, and high-frequency data is the edge letter for characterizing image
Breath is mainly divided into fixed size (comprising 1*M picture to low-frequency data using the unequal matching similarity method based on region
Element) n image block, using the rgb value of each piece of each pixel as characteristic information.
As shown in fig. 7, in the present embodiment, the matching module includes：
Similarity calculated 31, it is similar with the matching of the second image correspondence image block for calculating the first image
Degree, selects the highest region of similarity as picture registration region；
Matching unit 32 is slightly matched for carrying out characteristic point using BRISK algorithm.
In the present embodiment, BRISK algorithm is on ICCV in 2011《BRISK:Binary Robust Invariant
Scalable Keypoints》A kind of feature extraction algorithm and a kind of binary feature description put forward in article is calculated
Son.It has preferable rotational invariance, scale invariability, preferable robustness etc..In image registration application, speed ratio
Compared with：SIFT<SURF<BRISK<FREAK<ORB, when to there is larger fuzzy image registration, BRISK algorithm shows most wherein
It is outstanding.
As shown in figure 8, in the present embodiment, the removal module includes：
Selection unit 51 calculates transformation matrix for randomly selecting 3 characteristic points pair not on the same line；
Metrics calculation unit 52 is used for calculating distance of each characteristic point by matrixing to character pair point
The distance threshold that adaptive approach obtains is by characteristic point to being divided into interior point and exterior point；
First evaluation unit 53, for being estimated again with all interior points when the quantity of interior point is greater than preset amount threshold
Calculate the transformation matrix；
Second evaluation unit 54, for when the quantity of interior point be not more than the preset amount threshold when, according to described interior
Points and error rate estimate the transformation matrix.
In the present embodiment, shown characteristic point is to match point is referred to as, if the feature of the first image and the second image
Point is to respectively：X=[x, y]TWith X '=[x ', y ']T, the two meets formula：
X '=HX
H is the transformation matrix of 6 parameters in formula, then above formula can be expressed as again：
Wherein,For transformation matrix H.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the device that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or device institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or device.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (8)
1. a kind of image split-joint method based on unmanned plane, which is characterized in that including：
Image to be spliced is obtained, the image to be spliced includes the first image and the second image；
Extract the characteristic point of the image to be spliced；
To the characteristic point of the first image and the second image to slightly being matched；
Edge detection is carried out using edge detection algorithm；
Remove the characteristic point pair of error hiding；
Splice the first image and the second image using the syncretizing mechanism gradually gone out is fade-in.
2. a kind of image split-joint method based on unmanned plane according to claim 1, which is characterized in that described in the extraction
The characteristic point of image to be spliced includes：
Gray processing is carried out to the first image and the second image；
To the first image and the second image progress Multidimensional Discrete wavelet transformation after gray processing, low-frequency data and high frequency are extracted
According to；
The low-frequency data is divided into several image blocks of default size, by the rgb value of each pixel of each image block
As characteristic information.
3. a kind of image split-joint method based on unmanned plane according to claim 2, which is characterized in that described to described
To progress, slightly matching includes the characteristic point of one image and the second image：
Calculate the first image and the second image correspondence image block matching similarity, select the highest region of similarity as
Picture registration region；
Characteristic point is carried out using BRISK algorithm slightly to match.
4. a kind of image split-joint method based on unmanned plane according to claim 3, which is characterized in that the removal mistake
The characteristic point matched is to including：
3 characteristic points pair not on the same line are randomly selected, transformation matrix is calculated；
It calculates each characteristic point and passes through distance of the matrixing to character pair point, the distance threshold obtained using adaptive approach
By characteristic point to being divided into interior point and exterior point；
When the quantity of interior point is greater than preset amount threshold, the transformation matrix is re-evaluated with all interior points；
When the quantity of interior point is not more than the preset amount threshold, the transformation is estimated according to the interior points and error rate
Matrix.
5. a kind of image splicing device based on unmanned plane, which is characterized in that including：
Image collection module, for obtaining image to be spliced, the image to be spliced includes the first image and the second image；
Feature point extraction module, for extracting the characteristic point of the image to be spliced；
Matching module, for the characteristic point to the first image and the second image to slightly being matched；
Edge detection module, for carrying out edge detection using edge detection algorithm；
Module is removed, for removing the characteristic point pair of error hiding；
Splicing module, for using the syncretizing mechanism splicing the first image and the second image being fade-in gradually out.
6. a kind of image splicing device based on unmanned plane according to claim 5, which is characterized in that the characteristic point mentions
Modulus block includes：
Gray processing unit, for carrying out gray processing to the first image and the second image；
Wavelet transform unit, for extracting to the first image and the second image progress Multidimensional Discrete wavelet transformation after gray processing
Low-frequency data and high-frequency data out；
Extraction unit, for the low-frequency data to be divided into several image blocks of default size, by each of each image block
The rgb value of pixel is as characteristic information.
7. a kind of image splicing device based on unmanned plane according to claim 6, which is characterized in that the matching module
Including：
Similarity calculated is selected for calculating the matching similarity of the first image and the second image correspondence image block
The highest region of similarity is as picture registration region；
Matching unit is slightly matched for carrying out characteristic point using BRISK algorithm.
8. a kind of image splicing device based on unmanned plane according to claim 7, which is characterized in that the removal module
Including：
Selection unit calculates transformation matrix for randomly selecting 3 characteristic points pair not on the same line；
Metrics calculation unit, for calculating distance of each characteristic point by matrixing to character pair point, using adaptive
The distance threshold that method obtains is by characteristic point to being divided into interior point and exterior point；
First evaluation unit, for being re-evaluated with all interior points described when the quantity of interior point is greater than preset amount threshold
Transformation matrix；
Second evaluation unit, for when the quantity of interior point be not more than the preset amount threshold when, according to it is described it is interior points and
Error rate estimates the transformation matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810556862.0A CN108921776A (en) | 2018-05-31 | 2018-05-31 | A kind of image split-joint method and device based on unmanned plane |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810556862.0A CN108921776A (en) | 2018-05-31 | 2018-05-31 | A kind of image split-joint method and device based on unmanned plane |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108921776A true CN108921776A (en) | 2018-11-30 |
Family
ID=64420014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810556862.0A Pending CN108921776A (en) | 2018-05-31 | 2018-05-31 | A kind of image split-joint method and device based on unmanned plane |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108921776A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020135394A1 (en) * | 2018-12-28 | 2020-07-02 | 清华大学 | Video splicing method and device |
| CN113592929A (en) * | 2021-08-04 | 2021-11-02 | 北京优翼科科技有限公司 | Real-time splicing method and system for aerial images of unmanned aerial vehicle |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130208997A1 (en) * | 2010-11-02 | 2013-08-15 | Zte Corporation | Method and Apparatus for Combining Panoramic Image |
| CN104134200A (en) * | 2014-06-27 | 2014-11-05 | 河海大学 | Mobile scene image splicing method based on improved weighted fusion |
| WO2017107700A1 (en) * | 2015-12-21 | 2017-06-29 | 努比亚技术有限公司 | Image registration method and terminal |
| CN107274346A (en) * | 2017-06-23 | 2017-10-20 | 中国科学技术大学 | Real-time panoramic video splicing system |
-
2018
- 2018-05-31 CN CN201810556862.0A patent/CN108921776A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130208997A1 (en) * | 2010-11-02 | 2013-08-15 | Zte Corporation | Method and Apparatus for Combining Panoramic Image |
| CN104134200A (en) * | 2014-06-27 | 2014-11-05 | 河海大学 | Mobile scene image splicing method based on improved weighted fusion |
| WO2017107700A1 (en) * | 2015-12-21 | 2017-06-29 | 努比亚技术有限公司 | Image registration method and terminal |
| CN107274346A (en) * | 2017-06-23 | 2017-10-20 | 中国科学技术大学 | Real-time panoramic video splicing system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020135394A1 (en) * | 2018-12-28 | 2020-07-02 | 清华大学 | Video splicing method and device |
| US11538177B2 (en) | 2018-12-28 | 2022-12-27 | Tsinghua University | Video stitching method and device |
| CN113592929A (en) * | 2021-08-04 | 2021-11-02 | 北京优翼科科技有限公司 | Real-time splicing method and system for aerial images of unmanned aerial vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Juan et al. | A comparison of sift, pca-sift and surf | |
| US8351662B2 (en) | System and method for face verification using video sequence | |
| US7174044B2 (en) | Method for character recognition based on gabor filters | |
| Trulls et al. | Dense segmentation-aware descriptors | |
| CN110546651B (en) | Method, system and computer readable medium for identifying objects | |
| CN107967482A (en) | Icon-based programming method and device | |
| Zhang et al. | Detecting and extracting the photo composites using planar homography and graph cut | |
| US20040190787A1 (en) | Image noise reduction | |
| CN111104943A (en) | Color image region-of-interest extraction method based on decision-level fusion | |
| WO2019204577A1 (en) | System and method for multimedia analytic processing and display | |
| CN111126412A (en) | Image key point detection method based on characteristic pyramid network | |
| Kim et al. | Adaptive weighted fusion with new spatial and temporal fingerprints for improved video copy detection | |
| Swami et al. | Candy: Conditional adversarial networks based fully end-to-end system for single image haze removal | |
| CN108921776A (en) | A kind of image split-joint method and device based on unmanned plane | |
| CN108805139A (en) | A kind of image similarity computational methods based on frequency-domain visual significance analysis | |
| CN113469216B (en) | Retail terminal poster identification and integrity judgment method, system and storage medium | |
| CN110633691A (en) | Binocular in-vivo detection method based on visible light and near-infrared camera | |
| Bhullar et al. | DWT and SIFT based passive copy-move forgery detection | |
| CN106611417B (en) | Method and device for classifying visual elements into foreground or background | |
| US20030044067A1 (en) | Apparatus and methods for pattern recognition based on transform aggregation | |
| Kumar et al. | A review of image features extraction techniques and their applications in image forensic | |
| Shah et al. | Copy-move forgery detection using hybrid transform and K-means clustering technique | |
| Chen et al. | Forensics of blurred images based on no-reference image quality assessment | |
| CN109101969A (en) | A kind of image processing method and device based on character recognition | |
| Kakar | Passive approaches for digital image forgery detection |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181130 |
|
| RJ01 | Rejection of invention patent application after publication |