CN107734294A - Monitoring image recovery system and method - Google Patents
Monitoring image recovery system and method Download PDFInfo
- Publication number
- CN107734294A CN107734294A CN201710880171.1A CN201710880171A CN107734294A CN 107734294 A CN107734294 A CN 107734294A CN 201710880171 A CN201710880171 A CN 201710880171A CN 107734294 A CN107734294 A CN 107734294A
- Authority
- CN
- China
- Prior art keywords
- data
- monitoring image
- point cloud
- cloud
- target point
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 108
- 238000011084 recovery Methods 0.000 title claims abstract description 16
- 238000000034 method Methods 0.000 title claims description 28
- 238000013500 data storage Methods 0.000 claims abstract description 20
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 51
- 230000000007 visual effect Effects 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 5
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000007850 degeneration Effects 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000003384 imaging method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000009792 diffusion process Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 101150067309 bmp4 gene Proteins 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001627 detrimental effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 210000001747 pupil Anatomy 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/207—Analysis of motion for motion estimation over a hierarchy of resolutions
-
- 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
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
- Studio Devices (AREA)
Abstract
The invention discloses a kind of monitoring image recovery system, including：First data acquisition facility, for obtaining continuous monitoring image data；Second data acquisition facility, for obtaining cloud data corresponding to the monitoring image data；First data storage device, for storing the continuous monitoring image data acquired in the first data acquisition module；Second data storage device, for storing the cloud data acquired in the second data acquisition module；Data processor, for selecting target point cloud corresponding to target area and the extraction in monitoring image data, the target point cloud is registering with target area progress, estimate the movement velocity of target point cloud, and calculate the spread function of target point cloud, to non-dots cloud position point spread function interpolation in the target area, the point spread function of the target area continuous position is obtained, so that the target area image is restored.The present invention has the beneficial effect that image restoration result is more true and reliable, reducing power is strong.
Description
Technical field
The present invention relates to computer vision and monitoring technology field, system and side that more particularly to a kind of monitoring image restores
Method.
Background technology
With the rapid development of machine vision technique, video monitoring can accurate recording accident in real time video shadow
Picture, the management of important place is applied to more and more.But limited by the principle of optical lens, monitoring image can not be right
Beyond the target blur-free imaging of field depth, and the picture matter of image border visual field is less than central vision, thus when in image we
When target area interested is beyond field depth or in image border, gratifying imaging effect can not be often obtained
Fruit.For moving faster target in monitored picture, because motion blur will also result in the decline of image quality, it is highly detrimental to
The scene of accident is reduced.
Chinese patent CN 104123566A disclose a kind of method and system of video blur consistency target tracking, can
The methods of training wavelet basis by Edge extraction, deblurring solves monitor video and obscured, but the method is based entirely on monitoring
Image carries out Mathematical treatment, can not characterize deblurring process with physical process, recovery effect is limited.
Chinese patent CN 106127699A disclose a kind of road monitoring random motion blurred picture Fast Restoration emulation system
System, the system have certain image restoration effect just for motion blur, can not restore the defocus blur beyond field depth,
Therefore also there is certain limitation in the application.
Therefore, a kind of restored method of monitoring image is needed badly, being capable of effective restored image.
The content of the invention
It is contemplated that the defects of overcoming above-mentioned prior art, the monitoring capacity of image repair is improved.
To achieve the above object, the present invention uses following technical scheme：A kind of monitoring image recovery system is provided, including：
First data acquisition facility, for obtaining continuous monitoring image data；
Second data acquisition facility, for obtaining cloud data corresponding to the monitoring image data；
First data storage device, for storing the continuous monitoring image data acquired in the first data acquisition module；
Second data storage device, for storing the cloud data acquired in the second data acquisition module；
Data processor, for selecting target point cloud corresponding to target area and the extraction in monitoring image data, by institute
State that target point cloud is registering with target area progress, estimate the movement velocity of target point cloud, and calculate the diffusion of target point cloud
Function, to non-dots cloud position point spread function interpolation in the target area, the point for obtaining the target area continuous position expands
Function is dissipated, so that the target area image is restored.
In some embodiments, first data acquisition facility is at least one monitoring camera, and first data are deposited
Saving is set to DVR, and the monitoring image data that the monitoring camera obtains are transmitted to HD recording through the network switch
Machine is stored.
In some embodiments, second data acquisition facility is at least one laser radar, second data storage
Device is radar data memory, and the laser radar stores the cloud data of measurement to the radar data memory.
In some embodiments, the visual field of the laser radar is more than or equal to the visual field of the monitoring camera.
In some embodiments, data processor estimates the movement velocity of target point cloud, and calculates the diffusion letter of target point cloud
Several detailed processes are：
Obtain the range information of target point cloud position in monitoring image；
Calculate the degenrate function of target point cloud defocus blur；
The movement velocity of target point cloud is estimated according to the monitoring image, calculates the degeneration letter of target point cloud motion blur
Number；
With reference to point spread function, the movement velocity of target point cloud, the target point of each visual field of monitoring image demarcated in advance
Cloud coordinate and target point cloud are registering with the monitoring image, to calculate the point spread function function of target point cloud.
In some embodiments, when calculating the degenrate function of target point cloud motion blur, target point cloud is in monitoring image middle position
The calculation formula of shifting is：
Wherein, f is monitoring camera focal length, and u is object distance, and d is the displacement of target point cloud, and d passes through laser radar
Cloud data obtains.
In some embodiments, the point spread function of each visual field of the monitoring image of monitoring camera is demarcated and deposited in advance
Storage.
In some embodiments, the point spread function calibration process of the monitoring image acquired in the monitoring camera is：Will
Image is divided into multiple circle ring areas by visual field, and the discrete point spread function in each region is surveyed using knife-edge method or point impulse method
Go forward side by side row interpolation, the point spread function of whole target area is finally combined into multiple circle ring areas.
To achieve the above object, the present invention also uses following technical scheme：A kind of monitoring image restored method is provided, including
Following steps：
First data acquisition facility obtains continuous monitoring image data；
Second data acquisition facility obtains cloud data corresponding to the monitoring image data；
First data storage device stores the continuous monitoring image data acquired in the first data acquisition module；
Second data storage device stores the cloud data acquired in the second data acquisition module；
Target point cloud corresponding to target area and extraction in data processor selection monitoring image data, by the target
Point cloud is registering with target area progress, estimates the movement velocity of target point cloud, and calculates the spread function of target point cloud, right
Non-dots cloud position point spread function interpolation in the target area, obtains the point spread function of the target area continuous position,
So that the target area image is restored.
In some embodiments, first data acquisition facility is at least one monitoring camera, and first data are deposited
Saving is set to DVR, and the monitoring image data that the monitoring camera obtains are transmitted to HD recording through the network switch
Machine is stored；
Second data acquisition facility is at least one laser radar, and second data storage device is radar data
Memory, the laser radar store the cloud data of measurement to the radar data memory.
The beneficial effects of the present invention are：
1st, the restoration algorithm different from the past only for Digital Image Processing, recuperation of the invention consider defocus mould
The influence of paste, motion blur and different field positions, has a physical significance, and image restoration result is more true and reliable.
2nd, when weather environment dislike very much slightly cause optical monitoring camera can not blur-free imaging when, can still be obtained with laser radar
The cloud data of target is obtained, ensures that system still has certain surveillance coverage under extreme conditions.
Brief description of the drawings
Fig. 1 is the workflow diagram of one embodiment of monitoring image restored method of the present invention.
Fig. 2 is the schematic diagram of monitoring image recovery system one embodiment of the present invention.
Wherein：1- monitoring cameras；2- laser radars；The 3- network switch；4- DVRs；5- radar datas store
Device；6- data processors.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, below in conjunction with accompanying drawing and specific implementation
Example, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only explaining this hair
It is bright, without being construed as limiting the invention.
Monitoring image recovery system of the present invention, including：
First data acquisition facility, for obtaining continuous monitoring image data；
Second data acquisition facility, for obtaining cloud data corresponding to the monitoring image data；
First data storage device, for storing the continuous monitoring image data acquired in the first data acquisition module；
Second data storage device, for storing the cloud data acquired in the second data acquisition module；
Data processor, for selecting target point cloud corresponding to target area and the extraction in monitoring image data, by institute
State that target point cloud is registering with target area progress, estimate the movement velocity of target point cloud, and calculate the diffusion of target point cloud
Function, to non-dots cloud position point spread function interpolation in the target area, the point for obtaining the target area continuous position expands
Function is dissipated, so that the target area image is restored.
Fig. 1 and Fig. 2 are referred to, is the flow chart of monitoring image recovery system one embodiment of the present invention, Fig. 2 shows prison
The schematic diagram of image restoration system one embodiment is controlled, in the present embodiment, including monitoring camera 1, laser radar 2, network
Interchanger 3, DVR 4, radar data memory 5 and data processor 6.
For monitoring camera, it is necessary to demarcate in advance and store the point spread function of the monitoring image of different visual fields, demarcation
Using an impulse method, the point spread function of dispersive target point is obtained first, recycles cubic Hamiltonian symmetrical systems algorithm to be regarded entirely
The point spread function of field.
When system works, monitoring camera transmits the monitoring image data of collection to HD recording by the network switch
Machine is stored, and the cloud data of lidar measurement is stored to radar data memory, when needing to restore monitoring image, data
Processor is aligned to the data time axle in DVR and radar data memory first, reads the mesh of monitoring image
The coordinate of region and corresponding target point cloud is marked, it is registering with the pixel progress of target area in monitoring image to target point cloud,
And according to the degenrate function of the range information of target point cloud position, according to the following formula calculating target point cloud defocus blur：
Wherein, R is blur radius, and R computational methods are：
Wherein, D is monitoring camera entrance pupil bore, and f is monitoring camera focal length, and u is object distance, and v is image distance, and s is image distance
With defocusing amount sum.
The motion blur of pixel in continuous monitoring image estimation target area in monitor video, it is assumed that target area
Pixel linear uniform motion in domain, if the angle of the direction of motion and x-axis is θ, pixel is transported in target area in monitoring image
Dynamic number of pixels integer approximations are L, then the degenrate function of motion blur can be expressed as：
Parameter L can be expressed as：
Wherein, f is monitoring camera focal length, and u is object distance, and d is pixel displacement in target area, can pass through laser
The cloud data of radar obtains.
With reference to the point spread function h for each visual field of monitoring image demarcated in advance0(x, y), estimate by defocus blur and
The point spread function h (x, y) of target point cloud under the influence of motion blur：
To the point spread function interpolation of target area non-dots cloud position, the point spread function of target area continuous position is obtained
Number, its process are：Image is divided into multiple circle ring areas by visual field, each region is surveyed using knife-edge method or point impulse method
Discrete point spread function is gone forward side by side row interpolation, and the point spread function of whole target area is finally combined into multiple circle ring areas.
Using the point spread function of whole target area as initial parameter, target area roll up using Wiener filtering
The long-pending image that is restored.Wiener filtering performance is good and does not have iterative process, and its frequency-domain expression can be reduced to：
Wherein, H (u, v) represents degenrate function, Sη(u, v) represents the power spectrum of noise, Sf(u, v) represents non-degraded image
Power spectrum, G (u, v) represents the frequency-domain expression of non-degraded image.
The beneficial effects of the present invention are：
1st, the restoration algorithm different from the past only for Digital Image Processing, recuperation of the invention consider defocus mould
The influence of paste, motion blur and different field positions, has a physical significance, and image restoration result is more true and reliable.
2nd, when weather environment dislike very much slightly cause optical monitoring camera can not blur-free imaging when, can still be obtained with laser radar
The point cloud chart of target is obtained, ensures that system still has certain surveillance coverage under extreme conditions.
Correspondingly, the present invention also provides a kind of monitoring image restored method, comprises the following steps：
First data acquisition facility obtains continuous monitoring image data；
Second data acquisition facility obtains cloud data corresponding to the monitoring image data；
First data storage device stores the continuous monitoring image data acquired in the first data acquisition module；
Second data storage device stores the cloud data acquired in the second data acquisition module；
Target point cloud corresponding to target area and extraction in data processor selection monitoring image data, by the target
Point cloud is registering with target area progress, estimates the movement velocity of target point cloud, and calculates the spread function of target point cloud, right
Non-dots cloud position point spread function interpolation in the target area, obtains the point spread function of the target area continuous position,
So that the target area image is restored.
Monitoring image restored method have with system parenchyma identical technical characteristic, here is omitted.
The embodiment of present invention described above, is not intended to limit the scope of the present invention..Any basis
Various other corresponding changes and deformation made by the technical concept of the present invention, should be included in the guarantor of the claims in the present invention
In the range of shield.
Claims (10)
- A kind of 1. monitoring image recovery system, it is characterised in that including：First data acquisition facility, for obtaining continuous monitoring image data；Second data acquisition facility, for obtaining cloud data corresponding to the monitoring image data；First data storage device, for storing the continuous monitoring image data acquired in the first data acquisition module；Second data storage device, for storing the cloud data acquired in the second data acquisition module；Data processor, for selecting target point cloud corresponding to target area and the extraction in monitoring image data, by the mesh Punctuate cloud is registering with target area progress, estimates the movement velocity of target point cloud, and calculates the spread function of target point cloud, To non-dots cloud position point spread function interpolation in the target area, the point spread function of the target area continuous position is obtained Number, so that the target area image is restored.
- 2. monitoring image recovery system according to claim 1, it is characterised in that first data acquisition facility is extremely A few monitoring camera, first data storage device is DVR, the monitoring figure that the monitoring camera obtains Transmit to DVR and stored through the network switch as data.
- 3. monitoring image recovery system as claimed in claim 2, it is characterised in that second data acquisition facility is at least One laser radar, second data storage device are radar data memory, and the laser radar is by the point cloud number of measurement According to storing to the radar data memory.
- 4. monitoring image recovery system as claimed in claim 2, it is characterised in that the visual field of the laser radar is more than or equal to The visual field of the monitoring camera.
- 5. monitoring image recovery system as claimed in claim 2, it is characterised in that data processor estimates the fortune of target point cloud Dynamic speed, and the detailed process for calculating the spread function of target point cloud is：Obtain the range information of target point cloud position in monitoring image；Calculate the degenrate function of target point cloud defocus blur；The movement velocity of target point cloud is estimated according to the monitoring image, calculates the degenrate function of target point cloud motion blur；Sat with reference to the point spread function of each visual field of monitoring image, the movement velocity of target point cloud, the target point cloud demarcated in advance It is marked with and target point cloud is registering with the monitoring image, calculates the point spread function function of target point cloud.
- 6. monitoring image recovery system as claimed in claim 5, it is characterised in that calculate the degeneration of target point cloud motion blur During function, target point cloud calculation formula of displacement in monitoring image is：<mrow> <mi>L</mi> <mo>=</mo> <mi>f</mi> <mfrac> <mi>d</mi> <mi>u</mi> </mfrac> </mrow>Wherein, f is monitoring camera focal length, and u is object distance, and d is the displacement of target point cloud, the point cloud that d passes through laser radar Data obtain.
- 7. monitoring image recovery system as claimed in claim 5, it is characterised in that the monitoring image of monitoring camera respectively regards The point spread function of field is demarcated and stored in advance.
- 8. monitoring image recovery system as claimed in claim 7, it is characterised in that the monitoring acquired in the monitoring camera The point spread function calibration process of image is：Image is divided into multiple circle ring areas by visual field, the discrete point that each region is surveyed using knife-edge method or point impulse method is expanded Scattered function is gone forward side by side row interpolation, and the point spread function of whole target area is finally combined into multiple circle ring areas.
- 9. a kind of monitoring image restored method, it is characterised in that comprise the following steps：First data acquisition facility obtains continuous monitoring image data；Second data acquisition facility obtains cloud data corresponding to the monitoring image data；First data storage device stores the continuous monitoring image data acquired in the first data acquisition module；Second data storage device stores the cloud data acquired in the second data acquisition module；Target point cloud corresponding to target area and extraction in data processor selection monitoring image data, by the target point cloud It is registering with target area progress, the movement velocity of target point cloud is estimated, and the spread function of target point cloud is calculated, to described Non-dots cloud position point spread function interpolation in target area, obtains the point spread function of the target area continuous position, so that The target area image is restored.
- 10. a kind of monitoring image restored method, it is characterised in that first data acquisition facility is at least one monitoring camera Head, first data storage device are DVR, and the monitoring image data that the monitoring camera obtains are handed over through network Change planes to transmit to DVR and stored；Second data acquisition facility is at least one laser radar, and second data storage device stores for radar data Device, the laser radar store the cloud data of measurement to the radar data memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710880171.1A CN107734294A (en) | 2017-09-26 | 2017-09-26 | Monitoring image recovery system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710880171.1A CN107734294A (en) | 2017-09-26 | 2017-09-26 | Monitoring image recovery system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107734294A true CN107734294A (en) | 2018-02-23 |
Family
ID=61207980
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710880171.1A Pending CN107734294A (en) | 2017-09-26 | 2017-09-26 | Monitoring image recovery system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107734294A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830805A (en) * | 2018-05-25 | 2018-11-16 | 北京小米移动软件有限公司 | Image processing method, device and readable storage medium storing program for executing, electronic equipment |
| CN109190448A (en) * | 2018-07-09 | 2019-01-11 | 深圳市科脉技术股份有限公司 | Face identification method and device |
| CN112365416A (en) * | 2020-11-10 | 2021-02-12 | 浙江大华技术股份有限公司 | Picture occlusion processing method and device, storage medium and electronic device |
| CN113132626A (en) * | 2021-03-26 | 2021-07-16 | 联想(北京)有限公司 | Image processing method and electronic equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101842661A (en) * | 2007-11-07 | 2010-09-22 | 电子地图有限公司 | Method of and arrangement for mapping range sensor data on image sensor data |
| CN101968881A (en) * | 2010-10-27 | 2011-02-09 | 东南大学 | Motion blurring and defocusing composite blurring image restoration method |
| CN102170526A (en) * | 2011-03-22 | 2011-08-31 | 公安部第三研究所 | Method for calculation of defocus fuzzy core and sharp processing of defocus fuzzy image of defocus fuzzy core |
| CN103426165A (en) * | 2013-06-28 | 2013-12-04 | 吴立新 | Precise registration method of ground laser-point clouds and unmanned aerial vehicle image reconstruction point clouds |
| CN103971341A (en) * | 2014-05-16 | 2014-08-06 | 北京理工大学 | Out-of-focus iris image restoration method based on frequency domain parameter estimation |
| CN104103050A (en) * | 2014-08-07 | 2014-10-15 | 重庆大学 | Real video recovery method based on local strategies |
| CN105389820A (en) * | 2015-11-18 | 2016-03-09 | 成都中昊英孚科技有限公司 | Infrared image definition evaluating method based on cepstrum |
| CN106485685A (en) * | 2016-08-30 | 2017-03-08 | 重庆大学 | A kind of vehicle-mounted high-quality imaging method being restored based on two steps |
-
2017
- 2017-09-26 CN CN201710880171.1A patent/CN107734294A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101842661A (en) * | 2007-11-07 | 2010-09-22 | 电子地图有限公司 | Method of and arrangement for mapping range sensor data on image sensor data |
| CN101968881A (en) * | 2010-10-27 | 2011-02-09 | 东南大学 | Motion blurring and defocusing composite blurring image restoration method |
| CN102170526A (en) * | 2011-03-22 | 2011-08-31 | 公安部第三研究所 | Method for calculation of defocus fuzzy core and sharp processing of defocus fuzzy image of defocus fuzzy core |
| CN103426165A (en) * | 2013-06-28 | 2013-12-04 | 吴立新 | Precise registration method of ground laser-point clouds and unmanned aerial vehicle image reconstruction point clouds |
| CN103971341A (en) * | 2014-05-16 | 2014-08-06 | 北京理工大学 | Out-of-focus iris image restoration method based on frequency domain parameter estimation |
| CN104103050A (en) * | 2014-08-07 | 2014-10-15 | 重庆大学 | Real video recovery method based on local strategies |
| CN105389820A (en) * | 2015-11-18 | 2016-03-09 | 成都中昊英孚科技有限公司 | Infrared image definition evaluating method based on cepstrum |
| CN106485685A (en) * | 2016-08-30 | 2017-03-08 | 重庆大学 | A kind of vehicle-mounted high-quality imaging method being restored based on two steps |
Non-Patent Citations (1)
| Title |
|---|
| 陶恂: ""一种基于块配准的运动模糊图像复原算法"", 《计算机应用与软件》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830805A (en) * | 2018-05-25 | 2018-11-16 | 北京小米移动软件有限公司 | Image processing method, device and readable storage medium storing program for executing, electronic equipment |
| CN109190448A (en) * | 2018-07-09 | 2019-01-11 | 深圳市科脉技术股份有限公司 | Face identification method and device |
| CN112365416A (en) * | 2020-11-10 | 2021-02-12 | 浙江大华技术股份有限公司 | Picture occlusion processing method and device, storage medium and electronic device |
| CN113132626A (en) * | 2021-03-26 | 2021-07-16 | 联想(北京)有限公司 | Image processing method and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108242062B (en) | Target tracking method, system, terminal and medium based on depth feature flow | |
| US8437551B2 (en) | Devices and methods for restoring low-resolution text images | |
| CN107734294A (en) | Monitoring image recovery system and method | |
| US9589328B2 (en) | Globally dominant point spread function estimation | |
| US20110013028A1 (en) | Video stabilizing method and system using dual-camera system | |
| US11093778B2 (en) | Method and system for selecting image region that facilitates blur kernel estimation | |
| CN104303208A (en) | Image-processing apparatus for removing haze contained in video, and method therefor | |
| CN107403451B (en) | Self-adaptive binary characteristic monocular vision odometer method, computer and robot | |
| CN111738071B (en) | Inverse perspective transformation method based on motion change of monocular camera | |
| EP3293700A1 (en) | 3d reconstruction for vehicle | |
| CN110827321B (en) | Multi-camera collaborative active target tracking method based on three-dimensional information | |
| Gao et al. | Atmospheric turbulence removal using convolutional neural network | |
| Duan et al. | Guided event filtering: Synergy between intensity images and neuromorphic events for high performance imaging | |
| CN108732570A (en) | The nowcasting method of disastrous convection weather based on particle filter blending algorithm | |
| CN115240089A (en) | Vehicle detection method of aerial remote sensing image | |
| Jin et al. | Practical speed measurement for an intelligent vehicle based on double radon transform in urban traffic scenarios | |
| US9826162B2 (en) | Method and apparatus for restoring motion blurred image | |
| Du et al. | A motion-blurred restoration method for surface damage detection of wind turbine blades | |
| CN114648639B (en) | Target vehicle detection method, system and device | |
| CN115984124A (en) | Method and device for de-noising and super-resolution of neuromorphic pulse signals | |
| CN115512263A (en) | Dynamic visual monitoring method and device for falling object | |
| CN111161304B (en) | Remote sensing video target track tracking method for rapid background estimation | |
| CN114170267A (en) | Target tracking method, device, equipment and computer readable storage medium | |
| CN102542530A (en) | Motion-blurred image defining device | |
| CN111899287A (en) | Ghost high dynamic range image fusion method for automatic driving |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180223 |