CN107194955B - Adaptive big data management method - Google Patents
Adaptive big data management method Download PDFInfo
- Publication number
- CN107194955B CN107194955B CN201710470471.2A CN201710470471A CN107194955B CN 107194955 B CN107194955 B CN 107194955B CN 201710470471 A CN201710470471 A CN 201710470471A CN 107194955 B CN107194955 B CN 107194955B
- Authority
- CN
- China
- Prior art keywords
- image
- equipment
- window
- filtering
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/292—Multi-camera tracking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/262—Analysis of motion using transform domain methods, e.g. Fourier domain methods
-
- 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
- G06T2207/10141—Special mode during image acquisition
-
- 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
- G06T2207/20004—Adaptive image processing
-
- 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]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of adaptive big data management method, including：Outdoor scene image capture is carried out to window outer scene, to obtain outdoor scene image；The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis result；The analysis result is received, different filtering strategies perform the outdoor scene image based on the analysis result, and export filtered image；The filtered image is received, the personage the track whether analysis of personage's track detection belongs to dangerous person with the personage track judged in the filtered image is performed to the filtered image；When personage track in the filtered image belongs to the personage track of dangerous person, automatic window-closing operation is performed to window.
Description
Technical field
The present invention relates to switch control field, more particularly to a kind of adaptive big data management method.
Background technology
Building, i.e. building, mansion, for literal meaning, building refer to tall and big building construction.Each building are all integrated
Multiple windows carry out building structure constructions.Window refers to the hole built on wall or roof on architecture, to make light
Or air is got in.Can bidirectional opening modern window by window frame, middle transparent part and movable part (hinge, handle,
Pulley etc.) three parts composition.Window frame is responsible for supporting the main structure of forms, can be timber, metal, ceramics or plastic material, thoroughly
Bright part is attached on window frame, can be paper, cloth, silk or glass material.Movable part is mainly based on metal material, in people
The place that hand touches may also be wrapped up with heat-insulating materials such as plastics.
In the prior art, a bad actor gets in from window in order to prevent, and indoor occupant is damaged or property is made
Into loss, the people of lowrise layer generally install the strick precaution that burglary-resisting window carries out a bad actor additional, so, on the one hand add economy
Cost, on the other hand causes limited view, while housing appearance is adversely affected.
The content of the invention
To solve the above-mentioned problems, the present invention provides a kind of adaptive big data management method, image can be performed
Personage's track detection analyzes the personage track for whether belonging to dangerous person with the personage track judged in described image, in the figure
When personage track as in belongs to the personage track of dangerous person, automatic window-closing operation is performed to window, it is even more important that also
Having used includes the image capture device of multiple photoelectric conversion units in parallel, and the analysis for the personage track of dangerous person provides
High-definition view data.
According to an aspect of the present invention, there is provided a kind of adaptive big data management method, the described method includes：
Outdoor scene image capture is carried out to window outer scene, to obtain outdoor scene image；
The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis knot
Fruit；
The analysis result is received, different filtering plans performs the outdoor scene image based on the analysis result
Slightly, and filtered image is exported；
The filtered image is received, the analysis of personage's track detection is performed to the filtered image to judge the filtering
Whether the personage track in image belongs to the personage track of dangerous person afterwards；
When personage track in the filtered image belongs to the personage track of dangerous person, automatic close is performed to window
Window operates.
More specifically, in the adaptive big data management method, further include：
The current state of window is detected to judge that window is in closed state or being not turned off state；
Wherein, when the personage track in the filtered image belongs to the personage track of dangerous person, window is performed
Automatic window-closing operation includes：When it is to be in closed state to judge window, stop operating the automatic window-closing that window performs.
More specifically, in the adaptive big data management method, further include：
When personage track in the filtered image belongs to the personage track of dangerous person, broadcasting connects with dangerous person
The corresponding audio alert file of nearly window；
When personage track in the filtered image belongs to the personage track of dangerous person, it will be approached with dangerous person
The position of the corresponding text information of window and window is together sent on the mobile terminal that householder holds.
More specifically, in the adaptive big data management method：
Outdoor scene image capture is carried out to window outer scene, is included with obtaining outdoor scene image：
Using the sub- equipment of the light-intensity test being arranged on window outline border, the irradiation luminous intensity outside window is detected to export
Real-time lighting intensity
Sub- equipment detection real-time lighting intensity is detected using the light intensity change rate being connected with the sub- equipment of the light-intensity test
Change rate, when the change rate of real-time lighting intensity is more than or equal to default change rate threshold value, sends change rate and crosses high RST, no
Then, change rate normal signal is sent；
Use the sub- equipment of adaptive sensing being connected with the sub- equipment of light intensity change rate detection being arranged on window outline border
Outdoor scene image capture is carried out to window outer scene, to obtain outdoor scene image；The sub- equipment bag of adaptive sensing
Include pixel data readout equipment and each pixel data obtains equipment, each pixel data obtains equipment including multiple in parallel
Photoelectric conversion unit, when receiving the change rate normal signal, each pixel data obtain equipment only using it is multiple simultaneously
Data that a photoelectric conversion unit in the photoelectric conversion unit of connection is sensed and output, when receiving, the change rate is excessive
During signal, each pixel data acquisition equipment merges the data that multiple photoelectric conversion units in parallel are sensed rear defeated
Go out, the pixel data readout equipment obtains equipment with each pixel data respectively and is connected, for reading each pixel number respectively
The outdoor scene figure is formed using the pixel value as each pixel, the pixel value of each pixel according to the data for obtaining equipment output
Picture；
The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis knot
Fruit includes：
The outdoor scene image, each pixel based on outdoor scene image are received using the sub- equipment of the first image initial survey
The pixel value of point determines the mean square deviation of outdoor scene image pixel value to be exported as target mean square error；
Outdoor scene image is received using the sub- equipment of the second image initial survey, noise analysis is carried out to outdoor scene image, with
The main noise signal of noise amplitude maximum and the secondary noise signal that noise amplitude time is big are obtained, based on main noise signal, secondary noise
Signal and outdoor scene image determine the signal-to-noise ratio of outdoor scene image to be exported as target signal to noise ratio；
The analysis result is received, different filtering plans performs the outdoor scene image based on the analysis result
Omit, and export filtered image to include：
Following filtering process is performed using the first sub- equipment of filtering：Docking received image uses length small for 6 HARR
Ripple base carries out 5 grades and decomposes the filtering image for laying equal stress on structure to obtain the sub- equipment output of the first filtering；
Following filtering process is performed using the second sub- equipment of filtering：For each pixel of the image received, use
Various filter windows carry out the pixel acquisition of corresponding various block of pixels centered on the pixel, determine each block of pixels
In gray value variance, select pixel of the correspondence filter window as target filter window to the pixel of gray value variance minimum
Value carries out medium filtering to obtain its filtered pixel value, and the filtered pixel values of all pixels based on the image received obtains the
The filtering image of the sub- equipment output of two filtering；
Following filtering process is performed using the sub- equipment of adaptive recursive filtering：Dock received image and carry out adaptive recurrence
Filtering process is to obtain the filtering image of adaptive recursive filtering equipment output；
Use equipment sub- with adaptive recursive filtering respectively, the sub- equipment of the first filtering, the sub- equipment of the second filtering, the first image
The sub- equipment of initial survey and the adaptively selected sub- equipment of the sub- equipment connection of the second image initial survey receive target mean square error and target letter
Make an uproar and compare, and be more than or equal to default mean deviation threshold less than or equal to default snr threshold and target mean square error in target signal to noise ratio
When, filtering process is performed to outdoor scene image using the sub- equipment of the first filtering and the second filtering sub- equipment to be handled successively
Image, when target signal to noise ratio is less than or equal to default snr threshold and target mean square error is less than default mean deviation threshold, uses
The second sub- equipment of filtering performs outdoor scene image filtering process to obtain processing image, is more than default letter in target signal to noise ratio
Make an uproar when being more than or equal to default mean deviation threshold than threshold value and target mean square error, using the first sub- equipment of filtering to outdoor scene image
Filtering process is performed to obtain processing image, is more than default snr threshold in target signal to noise ratio and target mean square error is less than default
During mean deviation threshold, filtering process is performed to outdoor scene image using adaptive recursive filtering equipment to obtain processing figure
Picture；
Wherein, the filtered image is received, the analysis of personage's track detection is performed to the filtered image to judge
Stating the personage track in filtered image and whether belonging to the personage track of dangerous person includes：The processing image is received, to institute
State processing image and perform whether the analysis of personage's track detection belongs to dangerous people with the personage track judged in the filtered image
The personage track of thing.
More specifically, in the adaptive big data management method, further include：Prestore default snr threshold and
Default mean deviation threshold.
More specifically, in the adaptive big data management method：
Perform the analysis of personage's track detection to the processing image is with the personage track judged in the filtered image
The no personage track for belonging to dangerous person includes：Personage's rail is performed to the processing image based on dangerous person's baseline locomotor track
Mark tests and analyzes the personage track for whether belonging to dangerous person with the personage track judged in the filtered image.
Brief description of the drawings
Embodiment of the present invention is described below with reference to attached drawing, wherein：
Fig. 1 is the block diagram of the adaptive window switching tube platform according to embodiment of the present invention.
Fig. 2 is the step flow chart of the adaptive big data management method according to embodiment of the present invention.
Reference numeral：1 image capture device；2 image analysis equipments；3 filtering selection equipment；4 personage's track detection equipment；
5 automatic window-closing equipment；S101 carries out outdoor scene image capture to window outer scene, to obtain outdoor scene image；S102
The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis result；S103
The analysis result is received, different filtering strategies perform the outdoor scene image based on the analysis result, and export
Filtered image；S104 receives the filtered image, and the analysis of personage's track detection is performed to the filtered image to judge
Whether the personage track in the filtered image belongs to the personage track of dangerous person；S105 is in the filtered image
When personage track belongs to the personage track of dangerous person, automatic window-closing operation is performed to window
Embodiment
Embodiment of the present invention is described in detail below with reference to accompanying drawings.
The material of window, generally speaking, is divided into three major types：Wooden, plastic-steel, aluminium alloy.Three has his own strong points.
Plastic-steel：Because being plastic material, weight is small, and heat-proof quality is good, and price is relatively low.Because often
In face of exposing to wind and rain, the sun shines, so most allowing people to concern the anti-aging problem of plastic-steel window.In fact, the plastic-steel window of high-quality
Service life up to 100 years or so.
Aluminium alloy：Be not in problem of aging because being metal material, and firm, impact resistance, intensity are big.But
The weakness that aluminum window is easiest to be attacked is exactly heat-proof quality because metal be heat good conductor, it is extraneous with it is indoor
Temperature can be with the frame transmission of window.
But it is worth query, the ratio on a fan window shared by frame is simultaneously not very big, and window is not one block of metal
Plate, but the glass of frame is inlayed, the heat transmitted by frame edge strip actually can produce the indoor temperature for having warmer, air-conditioning
Great influence
But that precautions averts perils is good, this problem, employs " bridge cut-off " skill on the aluminium alloy window having in order to prevent
Art, i.e., add one layer of resin material in aluminum alloy window frame, thoroughly broken off the approach of heat conduction.
It is wooden：Comparatively, wooden should be forms frame material the most perfect, no matter from heat-insulated, sound insulation angularly
For have obvious advantage, and inherent texture and natural figure are more moving.It is actual although wooden
On some be used to doing the solid wood of window frame and have already been through processing special layer by layer, not only without moisture, it is desirable to which higher is even
Fat is sucked, so, so-called wooden actually as fossil, solid wood after treatment, is only protected
The appearance of timber has been stayed, quality is but completely different, will not cracking deformation, less suffers insect bite with worry, is corroded, moreover,
Intensity also greatly increases.
In addition, also a kind of frame structure is referred to as aluminium Bao Mu, the outdoor part of wood frame is one layer of aluminium alloy structure,
In fact, this be combine wood frame thermal insulation it is good and the advantages of intensity of aluminum alloy is high, integrate, maximize favourable factors and minimize unfavourable ones.
The wooden unique shortcoming of window is exactly that cost is too high.
Currently, lack effective testing mechanism to a bad actor outside window, cause people to need that burglary-resisting window or real-time is installed
The observation to environment outside window and pedestrian is kept, brings no small financial burden or stress.In order to overcome it is above-mentioned not
Foot, the present invention have built a kind of adaptive window switching tube platform and method, can solve above-mentioned technical problem.
Fig. 1 is the block diagram of the adaptive window switching tube platform according to embodiment of the present invention, described
Platform includes：
Image capture device, for carrying out outdoor scene image capture to window outer scene, to obtain outdoor scene figure
Picture；
Image analysis equipment, is connected with described image capture device, for receiving the outdoor scene image, and to described
Outdoor scene image carries out picture signal analysis to export analysis result；
Filtering selection equipment, is connected with described image analytical equipment, for receiving the analysis result, based on the analysis
As a result different filtering strategies are performed to the outdoor scene image, and export filtered image；
Personage's track detection equipment, is connected with the filtering selection equipment, for receiving the filtered image, to described
Filtered image performs whether the analysis of personage's track detection belongs to dangerous people with the personage track judged in the filtered image
The personage track of thing；
Automatic window-closing equipment, is connected with personage's track detection equipment, for the personage in the filtered image
When track belongs to the personage track of dangerous person, automatic window-closing operation is performed to window.
Then, continue that the concrete structure of the adaptive window switching tube platform of the present invention is further detailed.
It can also include in the adaptive window switching tube platform：
Window condition detecting device, for detect the current state of window with judge window be in closed state or
It is not turned off state；
Wherein, the automatic window-closing equipment is also connected with the window condition detecting device, and the automatic window-closing equipment exists
When personage track in the filtered image belongs to the personage track of dangerous person, automatic window-closing operation bag is performed to window
Include：When it is to be in closed state to judge window, stop operating the automatic window-closing that window performs.
It can also include in the adaptive window switching tube platform：
Audio alert equipment, is connected with the automatic window-closing equipment, for the personage track in the filtered image
When belonging to the personage track of dangerous person, play with dangerous person close to the corresponding audio alert file of window；
Wireless telecom equipment, is connected with the automatic window-closing equipment, for the personage track in the filtered image
When belonging to the personage track of dangerous person, by with dangerous person close to the position one of the corresponding text information of window and window
It is same to be sent on the mobile terminal that householder holds.
In the adaptive window switching tube platform：
Described image capture device includes：
The sub- equipment of light-intensity test, is arranged on window outline border, for detecting the irradiation luminous intensity outside window to export
Real-time lighting intensity；
Light intensity change rate detects sub- equipment, is connected with the sub- equipment of the light-intensity test, for detecting real-time lighting intensity
Change rate, when the change rate of real-time lighting intensity is more than or equal to default change rate threshold value, sends change rate and crosses high RST, no
Then, change rate normal signal is sent；
The sub- equipment of adaptive sensing, is arranged on window outline border, and detecting sub- equipment with light intensity change rate is connected, for pair
Window outer scene carries out outdoor scene image capture, to obtain outdoor scene image；The sub- equipment of adaptive sensing includes
Pixel data readout equipment and each pixel data obtain equipment, each pixel data, which obtains equipment, includes multiple light in parallel
Electric converting unit, when receiving the change rate normal signal, each pixel data obtains equipment and only uses multiple parallel connections
Photoelectric conversion unit in the data that are sensed of a photoelectric conversion unit and output, when receiving the excessive letter of the change rate
Number when, each pixel data obtain equipment the data that multiple photoelectric conversion units in parallel are sensed are merged it is rear defeated
Go out, the pixel data readout equipment obtains equipment with each pixel data respectively and is connected, for reading each pixel number respectively
The outdoor scene figure is formed using the pixel value as each pixel, the pixel value of each pixel according to the data for obtaining equipment output
Picture；
Described image analytical equipment includes：
The first sub- equipment of image initial survey, for receiving the outdoor scene image, each picture based on outdoor scene image
The pixel value of vegetarian refreshments determines the mean square deviation of outdoor scene image pixel value to be exported as target mean square error；
The second sub- equipment of image initial survey, for receiving outdoor scene image, noise analysis is carried out to outdoor scene image, with
The main noise signal of noise amplitude maximum and the secondary noise signal that noise amplitude time is big are obtained, based on main noise signal, secondary noise
Signal and outdoor scene image determine the signal-to-noise ratio of outdoor scene image to be exported as target signal to noise ratio；
The filtering selection equipment includes：
The first sub- equipment of filtering, for performing following filtering process：Docking received image uses length small for 6 HARR
Ripple base carries out 5 grades and decomposes the filtering image for laying equal stress on structure to obtain the sub- equipment output of the first filtering；
The second sub- equipment of filtering, for performing following filtering process：For each pixel of the image received, use
Various filter windows carry out the pixel acquisition of corresponding various block of pixels centered on the pixel, determine each block of pixels
In gray value variance, select pixel of the correspondence filter window as target filter window to the pixel of gray value variance minimum
Value carries out medium filtering to obtain its filtered pixel value, and the filtered pixel values of all pixels based on the image received obtains the
The filtering image of the sub- equipment output of two filtering；
The sub- equipment of adaptive recursive filtering, for performing following filtering process：Received image is docked adaptively to be passed
Return filtering process to obtain the filtering image of the sub- equipment output of adaptive recursive filtering；
Adaptively selected sub- equipment, equipment sub- with adaptive recursive filtering, the sub- equipment of the first filtering, the second filtering are sub respectively
Equipment, the sub- equipment of the first image initial survey and the sub- equipment connection of the second image initial survey, for receiving target mean square error and target letter
Make an uproar and compare, and be more than or equal to default mean deviation threshold less than or equal to default snr threshold and target mean square error in target signal to noise ratio
When, filtering process is performed to outdoor scene image using the sub- equipment of the first filtering and the second filtering sub- equipment to be handled successively
Image, when target signal to noise ratio is less than or equal to default snr threshold and target mean square error is less than default mean deviation threshold, uses
The second sub- equipment of filtering performs outdoor scene image filtering process to obtain processing image, is more than default letter in target signal to noise ratio
Make an uproar when being more than or equal to default mean deviation threshold than threshold value and target mean square error, using the first sub- equipment of filtering to outdoor scene image
Filtering process is performed to obtain processing image, is more than default snr threshold in target signal to noise ratio and target mean square error is less than default
During mean deviation threshold, filtering process is performed to outdoor scene image using adaptive recursive filtering equipment to obtain processing figure
Picture；
Wherein, personage's track detection equipment is connected with the adaptively selected sub- equipment, for receiving the processing
Image, performs whether the analysis of personage's track detection is belonged to the personage track judged in the filtered image to the processing image
In the personage track of dangerous person.
It can also include in the adaptive window switching tube platform：
Data storage device, is connected with the adaptively selected sub- equipment, for prestore default snr threshold and
Default mean deviation threshold.
In the adaptive window switching tube platform：
Personage's track detection equipment performs the analysis of personage's track detection to judge the filtering to the processing image
Whether the personage track in image belongs to the personage track of dangerous person and includes afterwards：Based on dangerous person's baseline locomotor track to institute
State processing image and perform whether the analysis of personage's track detection belongs to dangerous people with the personage track judged in the filtered image
The personage track of thing.
Fig. 2 is the step flow chart of adaptive big data management method according to embodiment of the present invention, the side
Method includes：
Outdoor scene image capture is carried out to window outer scene, to obtain outdoor scene image；
The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis knot
Fruit；
The analysis result is received, different filtering plans performs the outdoor scene image based on the analysis result
Slightly, and filtered image is exported；
The filtered image is received, the analysis of personage's track detection is performed to the filtered image to judge the filtering
Whether the personage track in image belongs to the personage track of dangerous person afterwards；
When personage track in the filtered image belongs to the personage track of dangerous person, automatic close is performed to window
Window operates.
Then, continue that the specific steps of the adaptive big data management method of the present invention are further detailed.
The adaptive big data management method can also include：
The current state of window is detected to judge that window is in closed state or being not turned off state；
Wherein, when the personage track in the filtered image belongs to the personage track of dangerous person, window is performed
Automatic window-closing operation includes：When it is to be in closed state to judge window, stop operating the automatic window-closing that window performs.
The adaptive big data management method can also include：
When personage track in the filtered image belongs to the personage track of dangerous person, broadcasting connects with dangerous person
The corresponding audio alert file of nearly window；
When personage track in the filtered image belongs to the personage track of dangerous person, it will be approached with dangerous person
The position of the corresponding text information of window and window is together sent on the mobile terminal that householder holds.
In the adaptive big data management method：
Outdoor scene image capture is carried out to window outer scene, is included with obtaining outdoor scene image：
Using the sub- equipment of the light-intensity test being arranged on window outline border, the irradiation luminous intensity outside window is detected to export
Real-time lighting intensity
Sub- equipment detection real-time lighting intensity is detected using the light intensity change rate being connected with the sub- equipment of the light-intensity test
Change rate, when the change rate of real-time lighting intensity is more than or equal to default change rate threshold value, sends change rate and crosses high RST, no
Then, change rate normal signal is sent；
Use the sub- equipment of adaptive sensing being connected with the sub- equipment of light intensity change rate detection being arranged on window outline border
Outdoor scene image capture is carried out to window outer scene, to obtain outdoor scene image；The sub- equipment bag of adaptive sensing
Include pixel data readout equipment and each pixel data obtains equipment, each pixel data obtains equipment including multiple in parallel
Photoelectric conversion unit, when receiving the change rate normal signal, each pixel data obtain equipment only using it is multiple simultaneously
Data that a photoelectric conversion unit in the photoelectric conversion unit of connection is sensed and output, when receiving, the change rate is excessive
During signal, each pixel data acquisition equipment merges the data that multiple photoelectric conversion units in parallel are sensed rear defeated
Go out, the pixel data readout equipment obtains equipment with each pixel data respectively and is connected, for reading each pixel number respectively
The outdoor scene figure is formed using the pixel value as each pixel, the pixel value of each pixel according to the data for obtaining equipment output
Picture；
The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis knot
Fruit includes：
The outdoor scene image, each pixel based on outdoor scene image are received using the sub- equipment of the first image initial survey
The pixel value of point determines the mean square deviation of outdoor scene image pixel value to be exported as target mean square error；
Outdoor scene image is received using the sub- equipment of the second image initial survey, noise analysis is carried out to outdoor scene image, with
The main noise signal of noise amplitude maximum and the secondary noise signal that noise amplitude time is big are obtained, based on main noise signal, secondary noise
Signal and outdoor scene image determine the signal-to-noise ratio of outdoor scene image to be exported as target signal to noise ratio；
The analysis result is received, different filtering plans performs the outdoor scene image based on the analysis result
Omit, and export filtered image to include：
Following filtering process is performed using the first sub- equipment of filtering：Docking received image uses length small for 6 HARR
Ripple base carries out 5 grades and decomposes the filtering image for laying equal stress on structure to obtain the sub- equipment output of the first filtering；
Following filtering process is performed using the second sub- equipment of filtering：For each pixel of the image received, use
Various filter windows carry out the pixel acquisition of corresponding various block of pixels centered on the pixel, determine each block of pixels
In gray value variance, select pixel of the correspondence filter window as target filter window to the pixel of gray value variance minimum
Value carries out medium filtering to obtain its filtered pixel value, and the filtered pixel values of all pixels based on the image received obtains the
The filtering image of the sub- equipment output of two filtering；
Following filtering process is performed using the sub- equipment of adaptive recursive filtering：Dock received image and carry out adaptive recurrence
Filtering process is to obtain the filtering image of adaptive recursive filtering equipment output；
Use equipment sub- with adaptive recursive filtering respectively, the sub- equipment of the first filtering, the sub- equipment of the second filtering, the first image
The sub- equipment of initial survey and the adaptively selected sub- equipment of the sub- equipment connection of the second image initial survey receive target mean square error and target letter
Make an uproar and compare, and be more than or equal to default mean deviation threshold less than or equal to default snr threshold and target mean square error in target signal to noise ratio
When, filtering process is performed to outdoor scene image using the sub- equipment of the first filtering and the second filtering sub- equipment to be handled successively
Image, when target signal to noise ratio is less than or equal to default snr threshold and target mean square error is less than default mean deviation threshold, uses
The second sub- equipment of filtering performs outdoor scene image filtering process to obtain processing image, is more than default letter in target signal to noise ratio
Make an uproar when being more than or equal to default mean deviation threshold than threshold value and target mean square error, using the first sub- equipment of filtering to outdoor scene image
Filtering process is performed to obtain processing image, is more than default snr threshold in target signal to noise ratio and target mean square error is less than default
During mean deviation threshold, filtering process is performed to outdoor scene image using adaptive recursive filtering equipment to obtain processing figure
Picture；
Wherein, the filtered image is received, the analysis of personage's track detection is performed to the filtered image to judge
Stating the personage track in filtered image and whether belonging to the personage track of dangerous person includes：The processing image is received, to institute
State processing image and perform whether the analysis of personage's track detection belongs to dangerous people with the personage track judged in the filtered image
The personage track of thing.
The adaptive big data management method can also include：Prestore default snr threshold and default mean square deviation
Threshold value.
In the adaptive big data management method：
Perform the analysis of personage's track detection to the processing image is with the personage track judged in the filtered image
The no personage track for belonging to dangerous person includes：Personage's rail is performed to the processing image based on dangerous person's baseline locomotor track
Mark tests and analyzes the personage track for whether belonging to dangerous person with the personage track judged in the filtered image.
In addition, the photoelectric conversion unit can be realized by cmos sensor or ccd sensor.Cmos image sensor is
A kind of typical solid state image sensor, has common historical origin with CCD.Cmos image sensor is usually by image-sensitive unit
Several portions such as array, line driver, row driver, time sequence control logic, a/d converter, data/address bus output interface, control interface
It is grouped into, this several part is usually all integrated on same silicon chip.Its course of work generally can be divided into reset, opto-electronic conversion,
Integration, read four parts.
Other digital signal processing circuits can also be integrated on cmos image sensor chip, such as a/d converter, automatically
Light exposure control, nonuniform compensation, white balance processing, black level control, Gamma correction etc., even can in order to quickly calculate
The DSP devices with programmable functions to be integrated with cmos device, so as to form at single-chip digital camera and image
Reason system.
Morrison in 1963, which has been delivered, can calculate sensor, this is that one kind can utilize photoconductive effect measure facula position
Structure, become cmos image sensor development beginning.The CMOS CMOS active pixel sensor single-chip digitals of nineteen ninety-five low noise
Camera succeeds.
Cmos image sensor has the advantages that following：1), random window reading capability.Window read operation is at random
Cmos image sensor is functionally better than the one side of CCD, and also referred to as area-of-interest is chosen.In addition, cmos image
The highly integrated characteristic of sensor is easily achieved it while opens the function of multiple tracking windows.2), capability of resistance to radiation.Total comes
Say, the potential radiation resistance of cmos image sensor has important enhancing relative to CCD performances.3), system complexity and can
By property.System hardware structure can be greatly simplified using cmos image sensor.4), non-destructive data read-out mode.5)、
The spectrum assignment of optimization.Significantly, since the reason for multiple functional transistors are integrated with pixel structure, CMOS figures
As sensor, there is also some shortcomings, mainly noise and filling rate two indices.In view of cmos image sensor is relatively excellent
Performance more so that cmos image sensor is widely used in every field.
Adaptive window switching tube platform and method using the present invention, it is dangerous to outdoor for being difficult in the prior art
The technical problem that molecule is effectively detected, by introduce include the image capture devices of multiple photoelectric conversion units in parallel with
High-definition view data is provided, the image processing equipment for introducing multiple customizations carries out special project to high-definition view data
Analysis with judge outdoor a bad actor track presence so that for automatic window-closing with take precautions against the implementation of a bad actor provide it is important
Reference data.
It is understood that although the present invention is disclosed as above with preferred embodiment, but above-described embodiment and it is not used to
Limit the present invention.For any those skilled in the art, without departing from the scope of the technical proposal of the invention,
Many possible changes and modifications are all made to technical solution of the present invention using the technology contents of the disclosure above, or are revised as
With the equivalent embodiment of change.Therefore, every content without departing from technical solution of the present invention, the technical spirit pair according to the present invention
Any simple modifications, equivalents, and modifications made for any of the above embodiments, still fall within the scope of technical solution of the present invention protection
It is interior.
Claims (3)
1. a kind of adaptive big data management method, the described method includes：
Outdoor scene image capture is carried out to window outer scene, to obtain outdoor scene image；
The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis result；
The analysis result is received, different filtering strategies perform the outdoor scene image based on the analysis result, and
Export filtered image；
The filtered image is received, the analysis of personage track detection is performed to the filtered image to scheme after judging the filtering
Whether the personage track as in belongs to the personage track of dangerous person；
When personage track in the filtered image belongs to the personage track of dangerous person, automatic window-closing behaviour is performed to window
Make；
It is characterized in that, further include：
The current state of window is detected to judge that window is in closed state or being not turned off state；
Wherein, when the personage track in the filtered image belongs to the personage track of dangerous person, window is performed automatic
Closing window operation includes：When it is to be in closed state to judge window, stop operating the automatic window-closing that window performs；
When personage track in the filtered image belongs to the personage track of dangerous person, play with dangerous person close to window
The corresponding audio alert file in family；
, will be with dangerous person close to window when personage track in the filtered image belongs to the personage track of dangerous person
The position of corresponding text information and window is together sent on the mobile terminal that householder holds；
Outdoor scene image capture is carried out to window outer scene, is included with obtaining outdoor scene image：
Using the sub- equipment of the light-intensity test being arranged on window outline border, the irradiation luminous intensity outside window is detected to export in real time
Intensity of illumination；
The change of sub- equipment detection real-time lighting intensity is detected using the light intensity change rate being connected with the sub- equipment of the light-intensity test
Rate, when the change rate of real-time lighting intensity is more than or equal to default change rate threshold value, sends change rate and crosses high RST, otherwise, hair
Go out change rate normal signal；
Using the sub- equipment of adaptive sensing being connected with the sub- equipment of light intensity change rate detection being arranged on window outline border to window
Family outer scene carries out outdoor scene image capture, to obtain outdoor scene image；The sub- equipment of adaptive sensing includes picture
Plain data taking equipment and each pixel data obtain equipment, each pixel data, which obtains equipment, includes multiple photoelectricity in parallel
Converting unit, when receiving the change rate normal signal, each pixel data obtains equipment only using multiple in parallel
Data that a photoelectric conversion unit in photoelectric conversion unit is sensed and output, high RST is crossed when receiving the change rate
When, each pixel data is obtained after equipment merges the data that multiple photoelectric conversion units in parallel are sensed and exported,
The pixel data readout equipment obtains equipment with each pixel data respectively and is connected, and is obtained for reading each pixel data respectively
The data of taking equipment output form the outdoor scene image using the pixel value as each pixel, the pixel value of each pixel；
The outdoor scene image is received, and picture signal analysis is carried out to the outdoor scene image to export analysis result bag
Include：
The outdoor scene image is received using the sub- equipment of the first image initial survey, each pixel based on outdoor scene image
Pixel value determines the mean square deviation of outdoor scene image pixel value to be exported as target mean square error；
Outdoor scene image is received using the sub- equipment of the second image initial survey, noise analysis is carried out to outdoor scene image, to obtain
The secondary big secondary noise signal of the main noise signal and noise amplitude of noise amplitude maximum, based on main noise signal, secondary noise signal
And outdoor scene image determines the signal-to-noise ratio of outdoor scene image to be exported as target signal to noise ratio；
The analysis result is received, different filtering strategies perform the outdoor scene image based on the analysis result, and
Output filtered image includes：
Following filtering process is performed using the first sub- equipment of filtering：Dock received image use length for 6 HARR wavelet basis
Carry out 5 grades and decompose the filtering image for laying equal stress on structure to obtain the sub- equipment output of the first filtering；
Following filtering process is performed using the second sub- equipment of filtering：It is various for each pixel of the image received, use
Filter window carries out the pixel acquisition of corresponding various block of pixels centered on the pixel, determines that each pixel is in the block
Gray value variance, select the correspondence filter window of gray value variance minimum as target filter window to the pixel value of the pixel into
For row medium filtering to obtain its filtered pixel value, the filtered pixel value of all pixels based on the image received obtains the second filter
The filtering image of marble equipment output；
Following filtering process is performed using the sub- equipment of adaptive recursive filtering：Dock received image and carry out adaptive recursive filtering
Handle to obtain the filtering image of the sub- equipment output of adaptive recursive filtering；
Use equipment sub- with adaptive recursive filtering respectively, the sub- equipment of the first filtering, the sub- equipment of the second filtering, the first image initial survey
Sub- equipment and the adaptively selected sub- equipment of the sub- equipment connection of the second image initial survey receive target mean square error and target signal to noise ratio,
And when target signal to noise ratio is less than or equal to default snr threshold and target mean square error is more than or equal to default mean deviation threshold, successively
Filtering process is performed to obtain processing image to outdoor scene image using the sub- equipment of the first filtering and the second filtering sub- equipment,
When target signal to noise ratio is less than or equal to preset snr threshold and target mean square error is less than default mean deviation threshold, the second filtering is used
Sub- equipment performs outdoor scene image filtering process to obtain processing image, is more than default snr threshold in target signal to noise ratio
And target mean square error performs filtering when being more than or equal to default mean deviation threshold using the first sub- equipment of filtering to outdoor scene image
Processing is more than default snr threshold in target signal to noise ratio and target mean square error is less than default mean square deviation threshold to obtain processing image
During value, filtering process is performed to outdoor scene image using adaptive recursive filtering equipment to obtain processing image；
Wherein, the filtered image is received, the analysis of personage's track detection is performed to the filtered image to judge the filter
The personage track whether the personage track after ripple in image belongs to dangerous person includes：The processing image is received, to the place
Reason image performs whether the analysis of personage's track detection belongs to dangerous person's with the personage track judged in the filtered image
Personage track；
Wherein, the photoelectric conversion unit is realized by cmos image sensor, and the cmos image sensor is a kind of solid
Imaging sensor, including image-sensitive cell array, line driver, row driver, time sequence control logic, a/d converter, data/address bus
Output interface and control interface, image-sensitive cell array, line driver, row driver, time sequence control logic, a/d converter, data
Bus output interface and control interface are integrated on same silicon chip；The course of work of the cmos image sensor is divided into multiple
Position, four parts of opto-electronic conversion, integration and reading.
2. adaptive big data management method as claimed in claim 1, it is characterised in that further include：
Prestore default snr threshold and default mean deviation threshold.
3. adaptive big data management method as claimed in claim 2, it is characterised in that：
Whether the analysis of personage's track detection is belonged to the personage track judged in the filtered image is performed to the processing image
Include in the personage track of dangerous person：The inspection of personage track is performed to the processing image based on dangerous person's baseline locomotor track
Survey the personage track whether analysis belongs to dangerous person with the personage track judged in the filtered image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710470471.2A CN107194955B (en) | 2017-06-20 | 2017-06-20 | Adaptive big data management method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710470471.2A CN107194955B (en) | 2017-06-20 | 2017-06-20 | Adaptive big data management method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107194955A CN107194955A (en) | 2017-09-22 |
| CN107194955B true CN107194955B (en) | 2018-04-13 |
Family
ID=59878549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710470471.2A Active CN107194955B (en) | 2017-06-20 | 2017-06-20 | Adaptive big data management method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107194955B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107706788A (en) * | 2017-11-16 | 2018-02-16 | 宁波隆翔环保科技有限公司 | A kind of box-type substation of wireless temperature measurement |
| CN107910773A (en) * | 2017-11-24 | 2018-04-13 | 余姚市荣事特电子有限公司 | A kind of box-type substation |
| CN108006562A (en) * | 2017-11-24 | 2018-05-08 | 余姚市荣事特电子有限公司 | A kind of solar street light |
| CN107965164A (en) * | 2017-11-24 | 2018-04-27 | 余姚市荣事特电子有限公司 | A kind of gas station's awning |
| CN107995364A (en) * | 2017-12-01 | 2018-05-04 | 庄宿龙 | A kind of method for alleviating puerpera's eye fatigue |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201828915U (en) * | 2010-06-30 | 2011-05-11 | 李立明 | Spatial information monitoring system |
| CN104424749A (en) * | 2013-08-29 | 2015-03-18 | 上海翼旋电子科技有限公司 | Rain automatic alarm system suitable for household |
| CN104504623A (en) * | 2014-12-29 | 2015-04-08 | 深圳市宇恒互动科技开发有限公司 | Method, system and device for scene recognition according to motion sensing |
| CN104632061A (en) * | 2015-01-30 | 2015-05-20 | 福建固美金属有限公司 | Intelligent roller blind type window |
| CN105913559A (en) * | 2016-04-06 | 2016-08-31 | 南京华捷艾米软件科技有限公司 | Motion sensing technique based bank ATM intelligent monitoring method |
| CN105929696A (en) * | 2015-12-18 | 2016-09-07 | 许昌学院 | Intelligent home security method |
| CN106056035A (en) * | 2016-04-06 | 2016-10-26 | 南京华捷艾米软件科技有限公司 | Motion-sensing technology based kindergarten intelligent monitoring method |
| CN106341625A (en) * | 2016-08-31 | 2017-01-18 | 宁波优而雅电器有限公司 | Self-contained light supplement non-refrigeration focal plane infrared shooting system |
| CN106337619A (en) * | 2016-08-31 | 2017-01-18 | 余姚市泗门印刷厂 | Opening method for composite window body |
| CN106437401A (en) * | 2016-08-31 | 2017-02-22 | 余姚市西蒙房车电器有限公司 | Intelligent and integrated window body driving system |
| CN106555546A (en) * | 2016-10-25 | 2017-04-05 | 彭晓梅 | The filtering of image content-based analysis selects platform |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8989687B2 (en) * | 2011-01-28 | 2015-03-24 | Broadcom Corporation | Communication receiver enhancements using multi-signal capture |
-
2017
- 2017-06-20 CN CN201710470471.2A patent/CN107194955B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201828915U (en) * | 2010-06-30 | 2011-05-11 | 李立明 | Spatial information monitoring system |
| CN104424749A (en) * | 2013-08-29 | 2015-03-18 | 上海翼旋电子科技有限公司 | Rain automatic alarm system suitable for household |
| CN104504623A (en) * | 2014-12-29 | 2015-04-08 | 深圳市宇恒互动科技开发有限公司 | Method, system and device for scene recognition according to motion sensing |
| CN104632061A (en) * | 2015-01-30 | 2015-05-20 | 福建固美金属有限公司 | Intelligent roller blind type window |
| CN105929696A (en) * | 2015-12-18 | 2016-09-07 | 许昌学院 | Intelligent home security method |
| CN105913559A (en) * | 2016-04-06 | 2016-08-31 | 南京华捷艾米软件科技有限公司 | Motion sensing technique based bank ATM intelligent monitoring method |
| CN106056035A (en) * | 2016-04-06 | 2016-10-26 | 南京华捷艾米软件科技有限公司 | Motion-sensing technology based kindergarten intelligent monitoring method |
| CN106341625A (en) * | 2016-08-31 | 2017-01-18 | 宁波优而雅电器有限公司 | Self-contained light supplement non-refrigeration focal plane infrared shooting system |
| CN106337619A (en) * | 2016-08-31 | 2017-01-18 | 余姚市泗门印刷厂 | Opening method for composite window body |
| CN106437401A (en) * | 2016-08-31 | 2017-02-22 | 余姚市西蒙房车电器有限公司 | Intelligent and integrated window body driving system |
| CN106555546A (en) * | 2016-10-25 | 2017-04-05 | 彭晓梅 | The filtering of image content-based analysis selects platform |
Non-Patent Citations (4)
| Title |
|---|
| 基于IAP15F2K61S2 单片机的多传感器新型智能窗控制***;何国渊 等;《机床与液压》;20150430;第43卷(第8期);148-151 * |
| 基于轨迹特征分析的行人异常行为识别;周维柏 等;《人工智能及识别技术》;20101231(第12期);摘要，正文第2.1、2.2、3.3、5节 * |
| 多功能家庭防盗控制器的设计;金广锋 等;《科技广场》;20071231(第3期);200-201 * |
| 智能视频分析技术在高校安全防卫中的应用;张海 等;《产业与科技论坛》;20161231;第15卷(第5期);61-62 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107194955A (en) | 2017-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107194955B (en) | Adaptive big data management method | |
| CN107301393B (en) | Self-adaptive switch big data management platform | |
| US10171029B2 (en) | Soiling measurement device for photovoltaic arrays employing microscopic imaging | |
| CN100507479C (en) | Methods and systems for detecting flash rate of surrounding lights for controlling camcorder frame rate | |
| KR20060136229A (en) | Method and Device for measuring environmental information by employing camera module | |
| CN106401359B (en) | Forms report control platform based on infrared photography | |
| CN106974540B (en) | Curtain state transition platform based on wind-force detection | |
| CN106555546A (en) | The filtering of image content-based analysis selects platform | |
| CN106382090B (en) | Interior exterior window combined control system | |
| CN108432232A (en) | Safe camera system | |
| CN106368550A (en) | Window control method | |
| CN107256623B (en) | Signal data receives and processing system | |
| CN106401438B (en) | Interior exterior window integral intelligent control platform | |
| CN106444484A (en) | Combined control method for internal and external windows | |
| CN106131392A (en) | Image processor, display image processing method and related electronic device | |
| CN108019143A (en) | Curtain automatic rolls up system | |
| US20090195677A1 (en) | Method and apparatus for storing location address of defective pixel | |
| CN107217977B (en) | Curtain state transition method based on wind-force detection | |
| CN107091048B (en) | Auto reeling method | |
| CN107454264A (en) | A kind of intelligent audio driving method | |
| EP3662655B1 (en) | Sky monitoring system | |
| CN107360373B (en) | Charge of oil vehicle oil gas collection platform | |
| CN109138732B (en) | Automatic cleaning method for intelligent household device | |
| CN109842837B (en) | Self-adaptive volume adjusting method for radio | |
| CN107240870A (en) | A kind of power distribution cabinet for base station Self Adaptive Control |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Bin Inventor before: Qin Ling |
|
| CB03 | Change of inventor or designer information | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180316 Address after: 230088 middle east part of C3 building, No. 800 Innovation Industrial Park, West Road, Wangjiang West Road, Hefei high tech Zone, Anhui Applicant after: ANHUI JOYFULL INFORMATION TECHNOLOGY CO., LTD. Address before: 201203 Shanghai Guo Shou Jing Road, Pudong New Area Zhangjiang hi tech Park No. 433 Applicant before: Qin Ling |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |