CN106157302A - High in the clouds based on wavelet filtering image segmentation system - Google Patents
High in the clouds based on wavelet filtering image segmentation system Download PDFInfo
- Publication number
- CN106157302A CN106157302A CN201610477885.3A CN201610477885A CN106157302A CN 106157302 A CN106157302 A CN 106157302A CN 201610477885 A CN201610477885 A CN 201610477885A CN 106157302 A CN106157302 A CN 106157302A
- Authority
- CN
- China
- Prior art keywords
- image
- value
- pixel
- clouds
- equipment
- 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.)
- Withdrawn
Links
Classifications
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
Landscapes
- Image Processing (AREA)
Abstract
The present invention relates to a kind of high in the clouds based on wavelet filtering image segmentation system, including ccd image sensing equipment, high in the clouds image-preprocessing device, wavelet filtering equipment and sub screen area splitting equipment, ccd image sensing equipment, high in the clouds image-preprocessing device and wavelet filtering equipment realize the image acquisition to target image, Image semantic classification and image filtering successively, sub screen area splitting equipment is connected with wavelet filtering equipment, for the screen subimage of each mobile terminal that filtering image based on the output of wavelet filtering equipment is partitioned in filtering image.By means of the invention it is possible to improve the speed of image segmentation.
Description
Technical field
The present invention relates to image segmentation field, particularly relate to a kind of high in the clouds based on wavelet filtering image segmentation system.
Background technology
Cloud computing is another great change after the 1980's mainframe computer to the big change of client-server.
Cloud computing (Cloud Computing) is Distributed Calculation (Distributed Computing), parallel computation (Parallel
Computing), effectiveness calculates (Utility Computing), [5] network storage (Network Storage
Technologies), virtualization (Virtualization), load balancing (Load Balance), hot-standby redundancy (High
Etc. Available) product that traditional computer and network technical development merge.Cloud computing starts progressively to enter with its computing editor
Enter each application.
The existing anti-electronic jamming device taken on the sly is that a kind of infrared facility is placed in the back side of curtain or stage,
Infrared facility sends the sightless infrared ray of human eye, if the person of taking on the sly is moving terminal or video camera shooting, the most infrared
Line by the video camera imaging of the interference person of taking on the sly, causes image or lower picture quality that the person of taking on the sly obtains, it is impossible to commercially sell
Sell and propagate on network, thus effectively safeguarding copyright interest.
But, this mode needs to launch infrared ray always, and ultrared transmitting is large-scale, needs from each
Infrared ray, relatively costly and poor feasibility are all launched in position.Therefore on-the-spot electronics prevents that the equipment of taking on the sly also needs to find other
Break through direction.
To this end, the present invention proposes the electronic equipment prevented taking on the sly in a kind of new scene, it is possible to by audience seat
The behavioral value of spectators and by the parameter extraction of site environment being determined whether there is the person of taking on the sly in audience seat, and
In time the position at the person of taking on the sly place can be positioned, in order to take to orient warning measure accordingly, remind the person of taking on the sly to put
Abandon the behavior of taking on the sly.
Summary of the invention
In order to solve the problems referred to above, the invention provides a kind of high in the clouds based on wavelet filtering image segmentation system, existing
Field arranges auditorium image capture device and completes the extraction to auditorium image, introduces various image processing equipment and graphical analysis
Equipment judges whether there is the suspicious person of taking on the sly in the image of auditorium, and in the case of there is the suspicious person of taking on the sly, extracting can
The numbering at the seat, the person of taking on the sly place doubted, detects in real time to ambient parameter or current performance content the most at the scene, with
Information and Site Detection result according to the suspicious person of taking on the sly determine the person that really do not takes on the sly, it is even more important that establish a set of energy
Enough mechanisms being oriented prompting according to the person of taking on the sly position, thus completely automatically realize the detection of the electronic type to the person of taking on the sly and police
Accuse.
According to an aspect of the present invention, it is provided that a kind of high in the clouds based on wavelet filtering image segmentation system, described system
Including ccd image sensing equipment, high in the clouds image-preprocessing device, wavelet filtering equipment and sub screen area splitting equipment, CCD schemes
As sensing equipment, high in the clouds image-preprocessing device and wavelet filtering equipment realize the image acquisition to target image, image successively
Pretreatment and image filtering, sub screen area splitting equipment is connected with wavelet filtering equipment, for defeated based on wavelet filtering equipment
The screen subimage of each mobile terminal that the filtering image gone out is partitioned in filtering image.
More specifically, in the image segmentation system of described high in the clouds based on wavelet filtering, including: projection room content acquisition sets
Standby, it is arranged on projection room draw curtain opposite, the scene image on Real-time Collection projection room draw curtain is using as real-time scene image
Output;Content analysis apparatus, the Background control being arranged on projection room is indoor, respectively with sub screen area splitting equipment and projection room
Content acquisition equipment connects, and is used for receiving real-time scene image, is additionally operable to receive the seat of each screen subimage and its correspondence
Bit number, carries out content matching to determine content matching percentage ratio, when interior by each screen subimage and real-time scene image
When holding match-percentage more than or equal to preset percentage threshold value, seat numbering corresponding for screen subimage is compiled as targeted seat
Number output;Ling Yang SPCE061A chip, the Background control being arranged on projection room is indoor, is connected with Content analysis apparatus, for base
The drive control signal including that targeted seat is numbered is sent in targeted seat numbering;Remote control equipment, is just being arranged on projection room draw curtain
The position, roof of top, controls for being connected the driving sent to receive Ling Yang SPCE061A chip with Ling Yang SPCE061A chip
Signal, and resolve drive control signal to determine that targeted seat is numbered, determine remote signal based on targeted seat numbering;
Seat warning lamp, each spectators seat arranges one, is used for receiving remote signal alert to carry out HONGGUANG based on remote signal
Show;Ccd image sensing equipment, is arranged on middle position, projection room roof, for carrying out the auditorium on projection room draw curtain opposite
Image acquisition is to export target image;Data splitting equipment, is positioned at high in the clouds, is connected with ccd image sensing equipment, for by target
Image is divided into N number of image block, and N is the natural number more than 1;N number of high in the clouds storage device, is positioned at high in the clouds, sets with data segmentation
Standby connection, for storing N number of image block respectively;Command channel, high in the clouds management equipment, is positioned at high in the clouds, for entering high in the clouds order
Row of channels manages;Cloud service central apparatus, is positioned at high in the clouds, is connected with N number of high in the clouds storage device, is used for concentrating N number of image to divide
Block, and N number of image block is distributed to M high in the clouds application apparatus, M is that the natural number more than 1 and M are less than or equal to N;M high in the clouds
Application apparatus, is positioned at high in the clouds, is connected with cloud service central apparatus, for receiving be assigned to, more than one image block,
Each high in the clouds application apparatus includes: gray processing process subset, including channel parameters extraction unit, weighted value memory element and
Gray value computing unit, channel parameters extraction unit is used for receiving each image block, extracts each in image block
The R passage pixel value of pixel, G passage pixel value and channel B pixel value, weighted value memory element is used for having prestored R and leads to
Higher, G channel weighting value and channel B weighted value, gray value computing unit respectively with channel parameters extraction unit and weighting
Value memory element connects, for each pixel in image block, by R passage pixel value and the product of R channel weighting value, G
The product addition of the product of passage pixel value and G channel weighting value and channel B pixel value and channel B weighted value with obtain for
The gray value of pixel, and obtain, based on the gray value of each pixel in image block, the gray processing figure that image block is corresponding
Picture;Wherein, R channel weighting value value is 0.298839, and G channel weighting value value is 0.586811, and channel B weighted value value is
0.114350;Histogram distribution detection subset, processes subset with gray processing and is connected, be used for receiving gray level image, and right
Gray level image carries out grey level histogram and processes to obtain corresponding histogram image, presents bimodal distribution at histogram image
Time, send global threshold and select signal, otherwise, send non-global threshold and select signal;Threshold value selects subset, divides with rectangular histogram
Cloth detection subset connects, for when receiving global threshold and selecting signal, using defeated as threshold data for global threshold 128
Go out, when receiving non-global threshold and selecting signal, neighbor pixel gray difference threshold 40 is exported as threshold data;Two-value
Change processes subset, selects subset and histogram distribution detection subset to be connected, for receiving the overall situation respectively with threshold value
When threshold value selects signal, for each pixel in gray level image, when gray value is more than or equal to threshold data, by pin
To pixel be set to white level pixel, when gray value less than threshold data time, by for pixel be set to black appliances
Flat pixel, and the binary image that output gray level image is corresponding;Binary conversion treatment subset be additionally operable to receive non-entirely
When office's threshold value selects signal, for each pixel in gray level image, calculate vertical direction upwards apart from its 3 pixels
The gray value of the pixel of point as upper grey scale pixel value, calculates vertical direction downwards apart from the pixel of its 3 pixels
Gray value as lower grey scale pixel value, calculated level direction to the left apart from the gray value of pixel of its 3 pixels as a left side
Grey scale pixel value, calculated level direction to the right apart from the gray value of pixel of its 3 pixels as right grey scale pixel value, when
The absolute value of the difference of upper grey scale pixel value and lower grey scale pixel value is less than or equal to threshold data and left grey scale pixel value and right pixel
The absolute value of the difference of gray value less than or equal to threshold data time, by for pixel be set to white level pixel, when upper picture
The absolute value of the difference of element gray value and lower grey scale pixel value more than threshold data or left grey scale pixel value and right grey scale pixel value it
Difference absolute value more than threshold data time, by for pixel be set to black level pixel, and output gray level image pair
The binary image answered;Picture smooth treatment subset, is connected with binary conversion treatment subset, is used for receiving binary image,
For each pixel in binary image, when adjacent all pixels exist the trip point of more than half, then
By for the gray value of pixel retain, otherwise, by for the gray value of pixel be set to white level pixel, and defeated
Go out the smoothed image that binary image is corresponding;Data merge equipment, are positioned at high in the clouds, are connected with M high in the clouds application apparatus, and being used for will
The smoothed image of the picture smooth treatment subset output of each high in the clouds application apparatus carries out splicing to obtain smooth integration figure
Picture;Wavelet filtering equipment, the Background control being arranged on projection room is indoor, merges equipment with data and is connected, and is used for receiving smooth whole
Close image, smooth integration image is performed 2-d wavelet Filtering Processing to obtain filtering image;Sub screen area splitting equipment, if
Put the Background control at projection room indoor, be connected to obtain filtering image with wavelet filtering subset, by filtering image with in advance
The screen pattern of the various mobile terminals of storage carries out mating each screen subimage to detect and be partitioned in filtering image,
For each screen subimage, for its position in filtering image determine for seat corresponding to screen subimage compile
Number;Wherein, for seat corresponding to screen subimage numbered hold for mobile terminal corresponding to screen subimage
The projection room seat numbering at spectators place.More specifically, in the image segmentation system of described high in the clouds based on wavelet filtering: content
It is concrete to determine content matching percentage ratio that each screen subimage and real-time scene image are carried out content matching by analytical equipment
Including: each pixel in real-time scene image is carried out pixel value with each pixel of each screen subimage and mates, as
The quantity of the element value pixel that the match is successful is the most, and the content matching percentage ratio that screen subimage is corresponding is the highest.
More specifically, in the image segmentation system of described high in the clouds based on wavelet filtering: the gray value of black level pixel
It is 0.
More specifically, in the image segmentation system of described high in the clouds based on wavelet filtering: the gray value of white level pixel
It is 255.
More specifically, in the image segmentation system of described high in the clouds based on wavelet filtering, also include: static storage device,
For prestoring R channel weighting value, G channel weighting value and channel B weighted value.
More specifically, in the image segmentation system of described high in the clouds based on wavelet filtering: static storage device is arranged on to be put
The Background control reflecting room is indoor.
More specifically, in the image segmentation system of described high in the clouds based on wavelet filtering: the pixel value pixel that the match is successful
The relation of quantity and content matching percentage ratio be non-linear dependence.
Accompanying drawing explanation
Below with reference to accompanying drawing, embodiment of the present invention are described, wherein:
Fig. 1 is the structure square frame according to high in the clouds based on the wavelet filtering image segmentation system shown in embodiment of the present invention
Figure.
Reference: 1CCD image sensing apparatus;2 high in the clouds image-preprocessing device;3 wavelet filtering equipment;4 sub-screen districts
Regional partition equipment
Detailed description of the invention
Below with reference to accompanying drawings the embodiment of high in the clouds based on the wavelet filtering image segmentation system of the present invention is carried out
Describe in detail.
Cloud computing, i.e. cloud computing, is increase, use and the delivery mode of related service based on the Internet,
It is usually directed to provide the most easily extension and the most virtualized resource by the Internet.Cloud is the one of network, the Internet
Metaphor saying.Past the most often represents telecommunications network with cloud, is the most also used for representing the Internet and underlying infrastructure
Abstract.Therefore, cloud computing even can allow you experience the operational capability of 10 TFlops per second, has the most powerful computing capability
Can be with simulated-nuclear explosion, Prediction of Climate Change and market trend.User is accessed by modes such as computer, notebook, mobile phones
Data center, carries out computing by the demand of oneself.
Definition to cloud computing has multiple saying.For what is cloud computing on earth, 100 kinds of explanations at least can be found.
Present stage is well accepted is National Institute of Standards and Technology (NIST) definition: cloud computing is a kind of by usage amount paying
Pattern, this pattern provide available, easily, on-demand network access, enter configurable computing resource sharing pond (money
Source includes network, server, storage, application software, service), these resources can quickly be provided, and only need to put into little pipe
Science and engineering is made, or carries out little mutual with service supplier.
The most substantial amounts of camera interference method mainly make use of imaging sensor to respond ultrared principle.In order to disturb shifting
Moving video camera and generate light in the picture of shooting, infrared emitter is installed in cinema.Except those emitters
Etc. extra cost price and regular job expense, this scheme can be by enclosing suitable filter on camera lens.But it is this
Mode is easy to be avoided because of the selection of the person of taking on the sly.
In order to overcome above-mentioned deficiency, the present invention has built a kind of high in the clouds based on wavelet filtering image segmentation system, to sight
Many seat images carry out electronic analysis, scene performance environment is carried out electronic analysis, and carries out, based on two kinds of analysis results, the person of taking on the sly
Judge, it addition, the person of taking on the sly is reminded by the positional information being additionally based upon the person of taking on the sly, thus realize the orientation of the on-the-spot person of taking on the sly of performance
Detection and orientation warning.
Fig. 1 is the structure square frame according to high in the clouds based on the wavelet filtering image segmentation system shown in embodiment of the present invention
Figure, described system includes that ccd image sensing equipment, high in the clouds image-preprocessing device, wavelet filtering equipment and sub screen area are divided
Cutting equipment, ccd image sensing equipment, high in the clouds image-preprocessing device and wavelet filtering equipment realize the figure to target image successively
As collection, Image semantic classification and image filtering, sub screen area splitting equipment is connected with wavelet filtering equipment, for based on small echo
The screen subimage of each mobile terminal that the filtering image of filter apparatus output is partitioned in filtering image.
Then, continue the concrete structure of high in the clouds based on the wavelet filtering image segmentation system of the present invention is carried out further
Explanation.
Described system includes: projection room content acquisition equipment, is arranged on projection room draw curtain opposite, shows for Real-time Collection
Scene image on the draw curtain of room is to export as real-time scene image;Content analysis apparatus, is arranged on the Background control of projection room
Indoor, are connected with sub screen area splitting equipment and projection room content acquisition equipment respectively, are used for receiving real-time scene image, also
For receiving the seat numbering of each screen subimage and its correspondence, each screen subimage is entered with real-time scene image
Row content matching is to determine content matching percentage ratio, when content matching percentage ratio is more than or equal to preset percentage threshold value, and will screen
The seat numbering that curtain subimage is corresponding exports as targeted seat numbering.
Described system includes: Ling Yang SPCE061A chip, and the Background control being arranged on projection room is indoor, sets with content analysis
Standby connection, for sending, based on targeted seat numbering, the drive control signal including that targeted seat is numbered;Remote control equipment, is arranged on
Position, roof directly over projection room draw curtain, sends out with reception Ling Yang SPCE061A chip for being connected with Ling Yang SPCE061A chip
The drive control signal come, and resolve drive control signal to determine that targeted seat is numbered, number based on targeted seat
Determine remote signal;Seat warning lamp, each spectators seat arranges one, is used for receiving remote signal to believe based on remote control
Number carry out HONGGUANG warning;Ccd image sensing equipment, is arranged on middle position, projection room roof, for projection room draw curtain opposite
Auditorium carry out image acquisition to export target image.
Described system includes: data splitting equipment, is positioned at high in the clouds, is connected with ccd image sensing equipment, for by target figure
As being divided into N number of image block, N is the natural number more than 1;N number of high in the clouds storage device, is positioned at high in the clouds, with data splitting equipment
Connect, for storing N number of image block respectively.
Described system includes: command channel, high in the clouds management equipment, is positioned at high in the clouds, for high in the clouds order is carried out channel
Reason;Cloud service central apparatus, is positioned at high in the clouds, is connected with N number of high in the clouds storage device, is used for concentrating N number of image block, and by N
Individual image block is distributed to M high in the clouds application apparatus, and M is that the natural number more than 1 and M are less than or equal to N;M high in the clouds application apparatus,
It is positioned at high in the clouds, is connected with cloud service central apparatus, for receiving be assigned to, more than one image block.
Each high in the clouds application apparatus includes: gray processing processes subset, deposits including channel parameters extraction unit, weighted value
Storage unit and gray value computing unit, channel parameters extraction unit is used for receiving each image block, extracts image block
In the R passage pixel value of each pixel, G passage pixel value and channel B pixel value, weighted value memory element is for depositing in advance
Stored up R channel weighting value, G channel weighting value and channel B weighted value, gray value computing unit respectively with channel parameters extraction unit
Connect with weighted value memory element, for each pixel in image block, by R passage pixel value and R channel weighting value
The product addition of the product of product, G passage pixel value and G channel weighting value and channel B pixel value and channel B weighted value is to obtain
Take for the gray value of pixel, and obtain, based on the gray value of each pixel in image block, the ash that image block is corresponding
Degreeization image;Wherein, R channel weighting value value is 0.298839, and G channel weighting value value is 0.586811, channel B weighted value
Value is 0.114350.
Each high in the clouds application apparatus includes: histogram distribution detection subset, processes subset with gray processing and is connected, uses
In receiving gray level image, and gray level image is carried out the grey level histogram process histogram image with acquisition correspondence, directly
When side's figure image presents bimodal distribution, send global threshold and select signal, otherwise, send non-global threshold and select signal;Threshold value
Select subset, be connected with histogram distribution detection subset, for when receiving global threshold and selecting signal, by overall situation threshold
Value 128 exports as threshold data, when receiving non-global threshold and selecting signal, neighbor pixel gray difference threshold 40 is made
Export for threshold data;Binary conversion treatment subset, selects subset and histogram distribution detection subset even respectively with threshold value
Connect, for when receiving global threshold and selecting signal, for each pixel in gray level image, when gray value is more than
During equal to threshold data, by for pixel be set to white level pixel, when gray value is less than threshold data, will be for
Pixel be set to black level pixel, and the binary image that output gray level image is corresponding.
Each high in the clouds application apparatus includes: binary conversion treatment subset is additionally operable to receiving non-global threshold selection letter
Number time, for each pixel in gray level image, calculate vertical direction upwards apart from the pixel of its 3 pixels
Gray value as upper grey scale pixel value, calculate vertical direction downwards apart from its 3 pixels pixel gray value as under
Grey scale pixel value, calculated level direction to the left apart from the gray value of pixel of its 3 pixels as left grey scale pixel value, meter
Calculate horizontal direction to the right apart from the gray value of pixel of its 3 pixels as right grey scale pixel value, when upper grey scale pixel value
With the difference that the absolute value of the difference of lower grey scale pixel value is less than or equal to threshold data and left grey scale pixel value and right grey scale pixel value
Absolute value less than or equal to threshold data time, by for pixel be set to white level pixel, when upper grey scale pixel value and under
The absolute value of the difference of grey scale pixel value is big more than the absolute value of threshold data or the difference of left grey scale pixel value and right grey scale pixel value
When threshold data, by for pixel be set to black level pixel, and the binary picture that output gray level image is corresponding
Picture.
Each high in the clouds application apparatus includes: picture smooth treatment subset, is connected with binary conversion treatment subset, is used for
Receive binary image, for each pixel in binary image, when adjacent all pixels exist half with
On trip point time, then by for the gray value of pixel retain, otherwise, by for the gray value of pixel be set to white
Level pixel point, and export the smoothed image that binary image is corresponding.
Described platform includes: data merge equipment, are positioned at high in the clouds, are connected with M high in the clouds application apparatus, for by each
The smoothed image of the picture smooth treatment subset output of high in the clouds application apparatus carries out splicing to obtain smooth integration image;Small echo
Filter apparatus, the Background control being arranged on projection room is indoor, merges equipment with data and is connected, is used for receiving smooth integration image,
Smooth integration image is performed 2-d wavelet Filtering Processing to obtain filtering image.
Described platform includes: sub screen area splitting equipment, and the Background control being arranged on projection room is indoor, with wavelet filtering
Subset connects to obtain filtering image, the screen pattern of filtering image with the various mobile terminals prestored is mated
With each screen subimage detected and be partitioned in filtering image, for each screen subimage, for it at filtering figure
Position in Xiang determine for screen subimage corresponding seat numbering.
Wherein, for seat corresponding to screen subimage numbered hold for mobile end corresponding to screen subimage
The projection room seat numbering at the spectators place of end.
Alternatively, in the system: each screen subimage is carried out by Content analysis apparatus with real-time scene image
Content matching is to determine that content matching percentage ratio specifically includes: by each pixel in real-time scene image and each screen
Each pixel of image carries out pixel value coupling, and the quantity of the pixel value pixel that the match is successful is the most, and screen subimage is corresponding
Content matching percentage ratio is the highest;The gray value of black level pixel is 0;The gray value of white level pixel is 255;Also include:
Static storage device, is used for prestoring R channel weighting value, G channel weighting value and channel B weighted value;Static storage device sets
Put the Background control at projection room indoor;And the relation of the quantity of the pixel value pixel that the match is successful and content matching percentage ratio
It is can be with right and wrong linear dependence.
It addition, small echo (Wavelet) this term, as the term suggests, " small echo " is exactly little waveform.So-called " little " refers to him
There is Decay Rate;It is referred to as " ripple " and then refers to its undulatory property, the concussion form of its amplitude alternate positive and negative.Become with Fourier
Commutation ratio, wavelet transformation is the localization analysis of time (space) frequency, he by flexible shift operations to signal (function) by
Step carries out multi-scale refinement, is finally reached high frequency treatment time subdivision, and at low frequency, frequency segmentation, can adapt to time frequency signal analysis automatically
Requirement, thus any details of signal can be focused on, solve the difficult problem of Fourier conversion, become the Fourier that continues and become
Important breakthrough in scientific method since changing.Someone calls " school microscop " wavelet transformation.
The application of wavelet analysis is that the theoretical research with wavelet analysis is bound tightly together ground.He believes in science and technology
Breath industrial field achieves the achievement attracted people's attention.Electronic information technology is a field important in six big new and high technologies, he
Importance be image and signal processing.Now, signal processing has become as the pith of contemporary science and technology work, letter
Number purpose processed is exactly: analyze accurately, diagnose, compression coding and quantization, quickly transmission or storage, accurately reconstruct (or
Recover).From the point of view of mathematically angle, signal and image procossing can be unified to regard that (image can be regarded as two to signal processing as
Dimensional signal), in the many many application analyzed in wavelet analysis ground, signal processing problems can be attributed to.For its character
Being to stablize constant signal in time, the ideal tools of process remains Fourier analysis.But in actual applications absolutely big
Majority signal is non-stable, and the instrument being particularly well-suited to astable signal is exactly wavelet analysis.
Use high in the clouds based on the wavelet filtering image segmentation system of the present invention, cannot stealing for performance for prior art
Clap provide effective detection of electrons technical problem, by use targetedly, high-precision a series of images processing equipment
With image analysis equipment, the audience status of auditorium is analyzed, is analyzed, upper also by the environment on-the-spot to performance
On the basis of stating analysis, the behavior of pirates shoot person in auditorium is accurately detected and location recognition, finally takes aposematic mechanism
The person of taking on the sly is carried out electronic reminding, thus while avoiding interference other spectators viewing, effectively reduces the generation of behavior of taking on the sly.
Although it is understood that the present invention discloses as above with preferred embodiment, but above-described embodiment being not used to
Limit the present invention.For any those of ordinary skill in the art, without departing under technical solution of the present invention ambit,
Technical solution of the present invention is made many possible variations and modification by the technology contents that all may utilize the disclosure above, or is revised as
Equivalent embodiments with change.Therefore, every content without departing from technical solution of the present invention, according to the technical spirit pair of the present invention
Any simple modification made for any of the above embodiments, equivalent variations and modification, all still fall within the scope of technical solution of the present invention protection
In.
Claims (8)
1. high in the clouds based on a wavelet filtering image segmentation system, described system includes ccd image sensing equipment, high in the clouds image
Pre-processing device, wavelet filtering equipment and sub screen area splitting equipment, ccd image sensing equipment, high in the clouds Image semantic classification set
Standby and wavelet filtering equipment realizes the image acquisition to target image, Image semantic classification and image filtering, sub screen area successively
Splitting equipment is connected with wavelet filtering equipment, is partitioned in filtering image for filtering image based on the output of wavelet filtering equipment
The screen subimage of each mobile terminal.
2. high in the clouds based on wavelet filtering as claimed in claim 1 image segmentation system, it is characterised in that described system bag
Include:
Projection room content acquisition equipment, is arranged on projection room draw curtain opposite, the scene graph on Real-time Collection projection room draw curtain
As to export as real-time scene image;
Content analysis apparatus, the Background control being arranged on projection room is indoor, respectively with sub screen area splitting equipment and projection room
Content acquisition equipment connects, and is used for receiving real-time scene image, is additionally operable to receive the seat of each screen subimage and its correspondence
Bit number, carries out content matching to determine content matching percentage ratio, when interior by each screen subimage and real-time scene image
When holding match-percentage more than or equal to preset percentage threshold value, seat numbering corresponding for screen subimage is compiled as targeted seat
Number output;
Ling Yang SPCE061A chip, the Background control being arranged on projection room is indoor, is connected with Content analysis apparatus, for based on mesh
Standard seat bit number sends the drive control signal including that targeted seat is numbered;
Remote control equipment, is arranged on the position, roof directly over projection room draw curtain, for being connected to connect with Ling Yang SPCE061A chip
Receive the drive control signal that Ling Yang SPCE061A chip is sent, and resolve drive control signal to determine that targeted seat is compiled
Number, determine remote signal based on targeted seat numbering;
Seat warning lamp, each spectators seat arranges one, is used for receiving remote signal to carry out red based on remote signal
Light warns;
Ccd image sensing equipment, is arranged on middle position, projection room roof, for carrying out the auditorium on projection room draw curtain opposite
Image acquisition is to export target image;
Data splitting equipment, is positioned at high in the clouds, is connected with ccd image sensing equipment, for being divided into N number of image to divide target image
Block, N is the natural number more than 1;
N number of high in the clouds storage device, is positioned at high in the clouds, is connected with data splitting equipment, for storing N number of image block respectively;
Command channel, high in the clouds management equipment, is positioned at high in the clouds, for high in the clouds order is carried out channel management;
Cloud service central apparatus, is positioned at high in the clouds, is connected with N number of high in the clouds storage device, is used for concentrating N number of image block, and by N
Individual image block is distributed to M high in the clouds application apparatus, and M is that the natural number more than 1 and M are less than or equal to N;
M high in the clouds application apparatus, is positioned at high in the clouds, is connected with cloud service central apparatus, for receive be assigned to, more than one
Image block, each high in the clouds application apparatus includes:
Gray processing processes subset, including channel parameters extraction unit, weighted value memory element and gray value computing unit, passage
Parameter extraction unit is used for receiving each image block, extract each pixel in image block R passage pixel value,
G passage pixel value and channel B pixel value, weighted value memory element is used for having prestored R channel weighting value, G channel weighting value
With channel B weighted value, gray value computing unit is connected with channel parameters extraction unit and weighted value memory element respectively, for figure
As each pixel in piecemeal, by R passage pixel value and the product of R channel weighting value, G passage pixel value and G channel weighting
The product of value and the product addition of channel B pixel value and channel B weighted value with obtain for the gray value of pixel, and base
In image block, the gray value of each pixel obtains the gray level image that image block is corresponding;Wherein, R channel weighting value takes
Value is 0.298839, and G channel weighting value value is 0.586811, and channel B weighted value value is 0.114350;
Histogram distribution detection subset, processes subset with gray processing and is connected, be used for receiving gray level image, and to gray processing
Image carries out grey level histogram and processes to obtain corresponding histogram image, when histogram image presents bimodal distribution, sends
Global threshold selects signal, otherwise, sends non-global threshold and selects signal;
Threshold value selects subset, is connected with histogram distribution detection subset, is used for when receiving global threshold and selecting signal,
Global threshold 128 is exported as threshold data, when receiving non-global threshold and selecting signal, by neighbor pixel gray scale difference
Threshold value 40 exports as threshold data;
Binary conversion treatment subset, selects subset and histogram distribution detection subset to be connected, for connecing respectively with threshold value
When receiving global threshold selection signal, for each pixel in gray level image, when gray value is more than or equal to number of threshold values
According to time, by for pixel be set to white level pixel, when gray value less than threshold data time, by for pixel set
It is set to black level pixel, and the binary image that output gray level image is corresponding;Binary conversion treatment subset is additionally operable to connecing
When receiving non-global threshold selection signal, for each pixel in gray level image, calculate vertical direction upwards distance
The gray value of the pixel of its 3 pixels, as upper grey scale pixel value, calculates vertical direction downwards apart from its 3 pixels
The gray value of pixel is as lower grey scale pixel value, and calculated level direction is to the left apart from the gray scale of pixel of its 3 pixels
Value as left grey scale pixel value, calculated level direction to the right apart from the gray value of pixel of its 3 pixels as right pixel
Gray value, when the absolute value of upper grey scale pixel value and the difference of lower grey scale pixel value is less than or equal to threshold data and left grey scale pixel value
During with the absolute value of the difference of right grey scale pixel value less than or equal to threshold data, by for pixel be set to white level pixel
Point, when the absolute value of upper grey scale pixel value and the difference of lower grey scale pixel value is more than threshold data or left grey scale pixel value and right pixel
The absolute value of the difference of gray value more than threshold data time, by for pixel be set to black level pixel, and output gray level
Change the binary image that image is corresponding;
Picture smooth treatment subset, is connected with binary conversion treatment subset, is used for receiving binary image, for binary picture
Each pixel in Xiang, when adjacent all pixels exist the trip point of more than half, then by for pixel
The gray value of point retains, otherwise, by for the gray value of pixel be set to white level pixel, and export binary image
Corresponding smoothed image;
Data merge equipment, are positioned at high in the clouds, are connected with M high in the clouds application apparatus, for by the figure of each high in the clouds application apparatus
As the smoothed image of smoothing processing subset output carries out splicing to obtain smooth integration image;
Wavelet filtering equipment, the Background control being arranged on projection room is indoor, merges equipment with data and is connected, and is used for receiving smooth whole
Close image, smooth integration image is performed 2-d wavelet Filtering Processing to obtain filtering image;
Sub screen area splitting equipment, the Background control being arranged on projection room is indoor, is connected with wavelet filtering subset to obtain
Filtering image, carries out the screen pattern of filtering image with the various mobile terminals prestored mating with detection and being partitioned into filter
Each screen subimage in ripple image, for each screen subimage, determines pin for its position in filtering image
To screen subimage corresponding seat numbering;Wherein, for seat corresponding to screen subimage numbered hold for
The projection room seat numbering at the spectators place of the mobile terminal that screen subimage is corresponding.
3. high in the clouds based on wavelet filtering as claimed in claim 2 image segmentation system, it is characterised in that:
Each screen subimage and real-time scene image are carried out content matching to determine content matching hundred by Content analysis apparatus
Proportion by subtraction specifically includes: each pixel of each pixel in real-time scene image Yu each screen subimage is carried out pixel value
Coupling, the quantity of the pixel value pixel that the match is successful is the most, and the content matching percentage ratio that screen subimage is corresponding is the highest.
4. high in the clouds based on wavelet filtering as claimed in claim 2 image segmentation system, it is characterised in that:
The gray value of black level pixel is 0.
5. high in the clouds based on wavelet filtering as claimed in claim 2 image segmentation system, it is characterised in that:
The gray value of white level pixel is 255.
6. high in the clouds based on wavelet filtering as claimed in claim 2 image segmentation system, it is characterised in that also include:
Static storage device, is used for prestoring R channel weighting value, G channel weighting value and channel B weighted value.
7. high in the clouds based on wavelet filtering as claimed in claim 6 image segmentation system, it is characterised in that:
The Background control that static storage device is arranged on projection room is indoor.
8. high in the clouds based on the wavelet filtering image segmentation system as described in claim 2-7 is arbitrary, it is characterised in that:
The quantity of the pixel value pixel that the match is successful and the relation of content matching percentage ratio are non-linear dependence.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610477885.3A CN106157302A (en) | 2016-06-27 | 2016-06-27 | High in the clouds based on wavelet filtering image segmentation system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610477885.3A CN106157302A (en) | 2016-06-27 | 2016-06-27 | High in the clouds based on wavelet filtering image segmentation system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106157302A true CN106157302A (en) | 2016-11-23 |
Family
ID=57349339
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610477885.3A Withdrawn CN106157302A (en) | 2016-06-27 | 2016-06-27 | High in the clouds based on wavelet filtering image segmentation system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106157302A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107368869A (en) * | 2017-08-29 | 2017-11-21 | 无锡北斗星通信息科技有限公司 | Personnel characteristics' advanced treating platform |
CN109605429A (en) * | 2018-12-06 | 2019-04-12 | 宁波宝尼尔厨具电器有限公司 | Type elbow type capable of washing removes Hu knife |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040062348A1 (en) * | 2002-09-27 | 2004-04-01 | Bor-Shenn Jeng | Image-based container defects detector |
CN105676741A (en) * | 2016-03-06 | 2016-06-15 | 张希梅 | Logistics channel remote control system based on Internet of Things technology |
-
2016
- 2016-06-27 CN CN201610477885.3A patent/CN106157302A/en not_active Withdrawn
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040062348A1 (en) * | 2002-09-27 | 2004-04-01 | Bor-Shenn Jeng | Image-based container defects detector |
CN105676741A (en) * | 2016-03-06 | 2016-06-15 | 张希梅 | Logistics channel remote control system based on Internet of Things technology |
Non-Patent Citations (1)
Title |
---|
朱江峰: "汽车牌照自动识别系统的研究", 《中国优秀硕士学位论文全文数据库——信息科技辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107368869A (en) * | 2017-08-29 | 2017-11-21 | 无锡北斗星通信息科技有限公司 | Personnel characteristics' advanced treating platform |
CN107368869B (en) * | 2017-08-29 | 2018-03-20 | 极客站企业孵化器管理南京股份有限公司 | Personnel characteristics' advanced treating platform |
CN109605429A (en) * | 2018-12-06 | 2019-04-12 | 宁波宝尼尔厨具电器有限公司 | Type elbow type capable of washing removes Hu knife |
CN109605429B (en) * | 2018-12-06 | 2020-10-09 | 蒋玉素 | Cleanable curved shaver |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105868748A (en) | Data service platform on basis of cloud processing | |
CN106027849B (en) | High in the clouds high-definition image processing platform | |
CN106101622A (en) | A kind of big data-storage system | |
CN105868747B (en) | Cloud computing system | |
CN106027850A (en) | Cloud median filtering system | |
CN106162076A (en) | Big data image gray processing processing means | |
CN110706328B (en) | Three-dimensional scene virtual generation method and system based on GAN network | |
CN106096665A (en) | Dual pathways cloud data management platform | |
CN106157302A (en) | High in the clouds based on wavelet filtering image segmentation system | |
CN105858377A (en) | Elevator safety control system | |
CN105872115A (en) | Remote monitoring system based on cloud computing | |
CN107545575A (en) | High in the clouds image segmentation system based on wavelet filtering | |
CN106162077A (en) | IN service based on the Internet management system | |
CN107545212A (en) | Binary channels cloud data management platform | |
EP4148379A1 (en) | Visual positioning method and apparatus | |
CN105955058A (en) | Wireless intelligent household system | |
CN107545211A (en) | High in the clouds data handling system | |
CN106067010A (en) | Data combination system based on distributed treatment | |
CN107257449B (en) | A kind of remote supervision system based on cloud computing | |
CN107547845A (en) | Big data processing method and big data processing unit | |
CN106855875A (en) | Data combination system based on distributed treatment | |
CN106157260B (en) | parallel image smoothing processing platform | |
CN106210623A (en) | High in the clouds histogram distribution detection platform | |
CN106067011A (en) | It is positioned at the high-definition image parallel processing platform in high in the clouds | |
CN106204757A (en) | Smart city cloud computing parking management information system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20161123 |
|
WW01 | Invention patent application withdrawn after publication |