Invention content
To solve the above-mentioned problems, the present invention provides a kind of smart homes based on sleep state detection to play control system
System and method, for being TV opening state in the current state of TV and receiving the sleep of sleep state identification equipment transmission
When controlling signal, the switchgear of chain road direction TV, which is sent, by radio communication closes drive signal, effectively prevents when viewing
The people of TV enter under slumber to still happening the TV of broadcasting is uncontrollable.
According to an aspect of the present invention, a kind of smart home broadcasting control system detected based on sleep state is provided,
The system comprises:
Tv state detection device is connect with the switchgear of TV, for the switch shape of the switchgear based on TV
State determines the current state of TV, and the current state of TV includes TV opening state and TV closed state;Image Acquisition is set
Standby, for carrying out image data acquiring to TV Week enveloping field scape to obtain high definition scene image, the high definition scene image divides
Resolution is related to the complexity of TV Week enveloping field scape;Object detection apparatus, for determining mesh therein based on high definition scene image
Mark whether the corresponding target of subgraph is human body target;Sleep state identification equipment, connect with object detection apparatus, for receiving
It is determined as each target subgraph of human body target, the amplitude of variation of each target subgraph is determined based on time shaft, works as change
When change amplitude is less than default variation amplitude, the current state for determining the human body target of corresponding target subgraph is sleep state, when
When all current states for being determined as human body target are all sleep state, sleep control signal is sent out, otherwise sends out non-sleep control
Signal processed;Wireless control apparatus is connect respectively with sleep state identification equipment and tv state detection device, in TV
Current state is for TV opening state and when receiving sleep control signal, the switchgear of chain road direction TV by radio communication
It sends and closes drive signal.
More specifically, described based in the smart home broadcasting control system of sleep state detection:The high definition scene
The resolution ratio of image it is related to the complexity of TV Week enveloping field scape including:The complexity of TV Week enveloping field scape is higher, the high definition
The resolution ratio of scene image is bigger, and the complexity of TV Week enveloping field scape is lower, and the resolution ratio of the high definition scene image is smaller;
Wherein, the resolution bits of the high definition scene image are between resolution ratio upper limit threshold and resolution limit threshold value.
More specifically, it is further included described based in the smart home broadcasting control system of sleep state detection:
Noise spot analytical equipment for receiving high definition scene image, carries out high definition scene image scene and judges to determine
The distribution situation of pixel pixel value in high definition scene image, based on the distribution situation to each picture in high definition scene image
Vegetarian refreshments is analyzed to determine whether into row pixel value as noise spot;
Noise region analytical equipment is connect with noise spot analytical equipment, for receiving point of the noise spot analytical equipment
Analysis as a result, each noise spot in high definition scene image is formed multiple noise regions, determine each noise region area and
Shape, and by the Area summing of each noise region to obtain the noise region gross area;
Starting device is filtered, is connect respectively with noise region analytical equipment and adaptive-filtering equipment, in noise range
When the domain gross area is more than or equal to preset area threshold value, start adaptive-filtering equipment, be less than default face in the noise region gross area
During product threshold value, adaptive-filtering equipment is closed, high definition scene image is input to image smoothing directly as pending image and is set
It is standby;
Adaptive-filtering equipment is connect respectively with filtering starting device and image smoothing equipment, for starting by filtering
After equipment starts, adaptive-filtering processing is carried out to high definition scene image to obtain pending image, and pending image is defeated
Enter to image smoothing equipment;
The adaptive-filtering equipment includes region split cells, module selecting unit, filtering execution unit and image group
Close unit, region split cells is used to receiving the shape of each noise region, the shape based on each noise region it is several
What feature, multiple benchmark subregions are split by each noise region;Image smoothing equipment, respectively with adaptive-filtering equipment
It is connected with filtering starting device, for obtaining smoothed image to the pending image received progress picture smooth treatment;
Wherein, the object detection apparatus is connect with image smoothing equipment, for determining multiple targets in smoothed image
Subgraph determines the area of each target subgraph, by the area of each target subgraph divided by the area of smoothed image
To obtain target area percentage, when target area percentage is fallen in the range of default human body area percentage, the mesh is determined
It is human body target to mark the corresponding target of subgraph;
Wherein, in high definition scene image, the region except multiple noise regions is non-noise region;Wherein, height is cleared out a gathering place
The distribution situation of pixel pixel value is included for region of each pixel in high definition scene image in scape image
Determine the pixel value range that the pixel pixel value should belong to;
Wherein, the module selecting unit is connect with the region split cells, for each noise region, for
Its each benchmark subregion after being split selects corresponding medium filtering template, the filtering execution unit and the module
Selecting unit connects, for each noise region, each benchmark subregion after being split for it to be taken in selection
Value filtering template performs medium filtering, to obtain each sub-regions filter pattern, and each sub-regions filter pattern is combined into
Filtered noise region subgraph;Described image assembled unit is connect with the filtering execution unit, for receiving each make an uproar
Sound area domain subgraph, and non-noise region combined with each noise region subgraph to obtain pending image;
Wherein, the shape of each benchmark subregion is rectangular, round or linear, when shape is rectangular, each base
The size of quasi- subregion is 3 × 3,5 × 5 or 7 × 7, when shape is round, the radius of each benchmark subregion is 3 pixels,
5 pixels or 7 pixels, when shape is linear, each benchmark subregion is an one-dimensional pixel set.
More specifically, it is further included described based in the smart home broadcasting control system of sleep state detection:Static state is deposited
Equipment is stored up, is connect with described image collecting device, for prestoring the resolution ratio upper limit threshold and the resolution limit
Threshold value.
More specifically, it is further included described based in the smart home broadcasting control system of sleep state detection:Use shape
Condition analytical equipment is connect with the switchgear of TV, for the switch based on the switchgear of TV in daily each period
State determines the frequency of use of daily each period of TV;
Wherein, the behaviour in service analytical equipment is ranked up the frequency of use of daily each period, will use frequency
Rate is more than or equal to the period of predeterminated frequency threshold value as frequent usage time interval, and frequency of use is less than predeterminated frequency threshold value
Period matches as idle usage time interval, and by current time with daily each period, when current time and certain
During one idle usage time interval matching, closing pre-warning signal is sent out;
Wherein, the screen of the behaviour in service analytical equipment and TV connects, and will close pre-warning signal and includes in TV
To wait for the confirmation of TV user or deny on screen, the switchgear of TV is displayed up to default display in closing pre-warning signal
Time threshold and cannot get TV user the input signal denied when, closing television.
According to another aspect of the present invention, it additionally provides a kind of smart home detected based on sleep state and plays controlling party
Method, the method includes:
The current state of TV is determined based on the on off state of the switchgear of TV, the current state of TV includes TV
Opening state and TV closed state;
Image data acquiring is carried out to TV Week enveloping field scape to obtain high definition scene image, point of the high definition scene image
Resolution is related to the complexity of TV Week enveloping field scape;
Determine whether the corresponding target of target subgraph therein is human body target based on high definition scene image;
The each target subgraph for being determined as human body target is received, the change of each target subgraph is determined based on time shaft
Change amplitude, when amplitude of variation is less than default variation amplitude, the current state for determining the human body target of corresponding target subgraph is
Sleep state when all current states for being determined as human body target are all sleep state, send out sleep control signal, otherwise sends out
Go out non-sleep control signal;
TV current state for TV opening state and when receiving sleep control signal, link by radio communication
It is sent to the switchgear of TV and closes drive signal.
More specifically, described based in the smart home control method for playing back of sleep state detection:
The resolution ratio of the high definition scene image it is related to the complexity of TV Week enveloping field scape including:TV Week enveloping field scape
Complexity is higher, and the resolution ratio of the high definition scene image is bigger, and the complexity of TV Week enveloping field scape is lower, the high definition scene
The resolution ratio of image is smaller;
Wherein, the resolution bits of the high definition scene image are between resolution ratio upper limit threshold and resolution limit threshold value.
More specifically, it is further included described based in the smart home control method for playing back of sleep state detection:
High definition scene image is received, scene is carried out to high definition scene image and judges to determine pixel in high definition scene image
The distribution situation of pixel value, based on the distribution situation to each pixel in high definition scene image into row pixel value analyze with
Determine whether for noise spot;
The analysis result of the noise spot analytical equipment is received, noise spot each in high definition scene image is formed into multiple make an uproar
Sound area domain, determines the area and shape of each noise region, and by the Area summing of each noise region to obtain noise range
The domain gross area;
When the noise region gross area is more than or equal to preset area threshold value, start adaptive-filtering equipment, in noise region
When the gross area is less than preset area threshold value, adaptive-filtering equipment is closed, by high definition scene image directly as pending image
It is input to image smoothing equipment;
Using adaptive-filtering equipment, after being activated, adaptive-filtering processing is carried out to high definition scene image to obtain
Pending image;The adaptive-filtering equipment includes region split cells, module selecting unit, filtering execution unit and image
Assembled unit, region split cells are used to receive the shape of each noise region, the shape based on each noise region
Each noise region is split into multiple benchmark subregions by geometric properties;
Picture smooth treatment is carried out to pending image to obtain smoothed image;Wherein, it is described to determine target therein
Whether the corresponding target of image is that human body target includes:It determines multiple target subgraphs in smoothed image, determines each mesh
The area of subgraph is marked, by the area of the area of each target subgraph divided by smoothed image to obtain target area percentage
Than when target area percentage is fallen in the range of default human body area percentage, determining the corresponding target of target subgraph
For human body target;
Wherein, in high definition scene image, the region except multiple noise regions is non-noise region;High definition scene image
The distribution situation of interior pixel pixel value includes determining this for region of each pixel in high definition scene image
The pixel value range that pixel pixel value should belong to;The module selecting unit is connect with the region split cells, for pair
Each noise region, each benchmark subregion after being split for it, selects corresponding medium filtering template, the filtering
Execution unit is connect with the module selecting unit, for each noise region, each benchmark after being split for it
Subregion takes the medium filtering template of selection to perform medium filtering, to obtain each sub-regions filter pattern, and by each height
Region filter pattern is combined into filtered noise region subgraph;Described image assembled unit connects with the filtering execution unit
It connects, for receiving each noise region subgraph, and non-noise region is combined to be treated with each noise region subgraph
Handle image;
Wherein, the shape of each benchmark subregion is rectangular, round or linear, when shape is rectangular, each base
The size of quasi- subregion is 3 × 3,5 × 5 or 7 × 7, when shape is round, the radius of each benchmark subregion is 3 pixels,
5 pixels or 7 pixels, when shape is linear, each benchmark subregion is an one-dimensional pixel set.
More specifically, it is further included described based in the smart home control method for playing back of sleep state detection:It deposits in advance
Store up the resolution ratio upper limit threshold and the resolution limit threshold value.
More specifically, it is further included described based in the smart home control method for playing back of sleep state detection:
Daily each period of TV is determined based on the on off state of the switchgear of TV in daily each period
Frequency of use;
The frequency of use of daily each period is ranked up, by frequency of use be more than or equal to predeterminated frequency threshold value when
Between section as frequent usage time interval, frequency of use is less than period of predeterminated frequency threshold value as free time usage time interval,
And match current time with daily each period, when being matched when current time with some idle usage time interval,
Send out closing pre-warning signal;
Pre-warning signal will be closed to include on the screen of TV waiting for the confirmation of TV user or denying, the switch of TV
When equipment is displayed up to default display time threshold in closing pre-warning signal and cannot get the input signal denied of TV user,
Closing television.
Specific embodiment
Below with reference to accompanying drawings to the implementation of the smart home broadcasting control system based on sleep state detection of the present invention
Scheme is described in detail.
Telephotography is to focus on the light image of scenery on photosensitive (or light guide) target surface of pick-up tube, the photoelectricity of target surface each point
The excitation or photoconductive situation of change of son are different with the brightness of light image each point.When being scanned with electron beam to target surface, that is, generate
One amplitude proportional is transmitted to the scanning electricity for making crt screen in television receiver in the electric signal of each point scene light image brightness
Beamlet becomes with the power of input signal.When with transmitting terminal synchronous scanning, show the original of transmission on the screen of kinescope
Image.
The process of television signal transmission distribution, for relaying the fact in other cities, generally from video camera, TV
The heart or mobile control room, then through microwave relay link, transmitting station, finally to user's television receiver.In addition, telecasting satellite and electricity
Cable television is also national respectively and the effective means of urban area property television transmission distribution.
People are frequently encountered such situation when watching TV:Because the dull or people itself of TV programme
It is tired out, during TV is watched, people will inadvertently get into it is shallow sleep state even slumber, and at this moment, TV by
In lacking state of the closing control still in lasting broadcasting, in this way, by entering the shallow people for sleeping state or even slumber
Ceaselessly impact visually and acoustically is carried out, has seriously affected the normal rest of people, consumes unnecessary electric power money
Source.
In order to overcome above-mentioned deficiency, the present invention has been built a kind of smart home detected based on sleep state and has played control system
Whether system enters dormant detection by the detection of the state switched to TV and to the people for watching TV, realizes
TV is automatically closed, so as to efficiently solve above-mentioned technical problem.
Fig. 1 is the smart home broadcasting control system based on sleep state detection according to embodiment of the present invention
Block diagram, the system comprises:
Tv state detection device is connect with the switchgear of TV, for the switch shape of the switchgear based on TV
State determines the current state of TV, and the current state of TV includes TV opening state and TV closed state;
Image capture device, for obtaining high definition scene image, institute to TV Week enveloping field scape progress image data acquiring
The resolution ratio for stating high definition scene image is related to the complexity of TV Week enveloping field scape;Object detection apparatus is cleared out a gathering place for being based on height
Scape image determines whether the corresponding target of target subgraph therein is human body target;
Sleep state identification equipment, connect with object detection apparatus, for receiving each target for being determined as human body target
Subgraph determines the amplitude of variation of each target subgraph based on time shaft, when amplitude of variation is less than default variation amplitude,
The current state for determining the human body target of corresponding target subgraph is sleep state, when all current shapes for being determined as human body target
When state is all sleep state, sleep control signal is sent out, otherwise sends out non-sleep control signal;
Wireless control apparatus is connect respectively with sleep state identification equipment and tv state detection device, in TV
Current state for TV opening state and when receiving sleep control signal, the switch of chain road direction TV is set by radio communication
Preparation send closing drive signal.
Then, continue to the present invention the concrete structure of smart home broadcasting control system detected based on sleep state into
Row further instruction.
Described based in the smart home broadcasting control system of sleep state detection:The resolution of the high definition scene image
Rate it is related to the complexity of TV Week enveloping field scape including:The complexity of TV Week enveloping field scape is higher, the high definition scene image
Resolution ratio is bigger, and the complexity of TV Week enveloping field scape is lower, and the resolution ratio of the high definition scene image is smaller;
Wherein, the resolution bits of the high definition scene image are between resolution ratio upper limit threshold and resolution limit threshold value.
Described based in the smart home broadcasting control system of sleep state detection, further include:
Noise spot analytical equipment for receiving high definition scene image, carries out high definition scene image scene and judges to determine
The distribution situation of pixel pixel value in high definition scene image, based on the distribution situation to each picture in high definition scene image
Vegetarian refreshments is analyzed to determine whether into row pixel value as noise spot;
Noise region analytical equipment is connect with noise spot analytical equipment, for receiving point of the noise spot analytical equipment
Analysis as a result, each noise spot in high definition scene image is formed multiple noise regions, determine each noise region area and
Shape, and by the Area summing of each noise region to obtain the noise region gross area;
Starting device is filtered, is connect respectively with noise region analytical equipment and adaptive-filtering equipment, in noise range
When the domain gross area is more than or equal to preset area threshold value, start adaptive-filtering equipment, be less than default face in the noise region gross area
During product threshold value, adaptive-filtering equipment is closed, high definition scene image is input to image smoothing directly as pending image and is set
It is standby;
Adaptive-filtering equipment is connect respectively with filtering starting device and image smoothing equipment, for starting by filtering
After equipment starts, adaptive-filtering processing is carried out to high definition scene image to obtain pending image, and pending image is defeated
Enter to image smoothing equipment;
The adaptive-filtering equipment includes region split cells, module selecting unit, filtering execution unit and image group
Close unit, region split cells is used to receiving the shape of each noise region, the shape based on each noise region it is several
What feature, multiple benchmark subregions are split by each noise region;Image smoothing equipment, respectively with adaptive-filtering equipment
It is connected with filtering starting device, for obtaining smoothed image to the pending image received progress picture smooth treatment;
Wherein, the object detection apparatus is connect with image smoothing equipment, for determining multiple targets in smoothed image
Subgraph determines the area of each target subgraph, by the area of each target subgraph divided by the area of smoothed image
To obtain target area percentage, when target area percentage is fallen in the range of default human body area percentage, the mesh is determined
It is human body target to mark the corresponding target of subgraph;
Wherein, in high definition scene image, the region except multiple noise regions is non-noise region;Wherein, height is cleared out a gathering place
The distribution situation of pixel pixel value is included for region of each pixel in high definition scene image in scape image
Determine the pixel value range that the pixel pixel value should belong to;
Wherein, the module selecting unit is connect with the region split cells, for each noise region, for
Its each benchmark subregion after being split selects corresponding medium filtering template, the filtering execution unit and the module
Selecting unit connects, for each noise region, each benchmark subregion after being split for it to be taken in selection
Value filtering template performs medium filtering, to obtain each sub-regions filter pattern, and each sub-regions filter pattern is combined into
Filtered noise region subgraph;Described image assembled unit is connect with the filtering execution unit, for receiving each make an uproar
Sound area domain subgraph, and non-noise region combined with each noise region subgraph to obtain pending image;
Wherein, the shape of each benchmark subregion is rectangular, round or linear, when shape is rectangular, each base
The size of quasi- subregion is 3 × 3,5 × 5 or 7 × 7, when shape is round, the radius of each benchmark subregion is 3 pixels,
5 pixels or 7 pixels, when shape is linear, each benchmark subregion is an one-dimensional pixel set.
Described based in the smart home broadcasting control system of sleep state detection, further include:Static storage device, with
Described image collecting device connects, for prestoring the resolution ratio upper limit threshold and the resolution limit threshold value.
Described based in the smart home broadcasting control system of sleep state detection, further include:Behaviour in service analysis is set
It is standby, it is connect with the switchgear of TV, for being determined based on the on off state of the switchgear of TV in daily each period
The frequency of use of daily each period of TV;
Wherein, the behaviour in service analytical equipment is ranked up the frequency of use of daily each period, will use frequency
Rate is more than or equal to the period of predeterminated frequency threshold value as frequent usage time interval, and frequency of use is less than predeterminated frequency threshold value
Period matches as idle usage time interval, and by current time with daily each period, when current time and certain
During one idle usage time interval matching, closing pre-warning signal is sent out;
Wherein, the screen of the behaviour in service analytical equipment and TV connects, and will close pre-warning signal and includes in TV
To wait for the confirmation of TV user or deny on screen, the switchgear of TV is displayed up to default display in closing pre-warning signal
Time threshold and cannot get TV user the input signal denied when, closing television.
Fig. 2 is the smart home control method for playing back based on sleep state detection according to embodiment of the present invention
Flow chart of steps, the method includes:
The current state of TV is determined based on the on off state of the switchgear of TV, the current state of TV includes TV
Opening state and TV closed state;
Image data acquiring is carried out to TV Week enveloping field scape to obtain high definition scene image, point of the high definition scene image
Resolution is related to the complexity of TV Week enveloping field scape;
Determine whether the corresponding target of target subgraph therein is human body target based on high definition scene image;
The each target subgraph for being determined as human body target is received, the change of each target subgraph is determined based on time shaft
Change amplitude, when amplitude of variation is less than default variation amplitude, the current state for determining the human body target of corresponding target subgraph is
Sleep state when all current states for being determined as human body target are all sleep state, send out sleep control signal, otherwise sends out
Go out non-sleep control signal;
TV current state for TV opening state and when receiving sleep control signal, link by radio communication
It is sent to the switchgear of TV and closes drive signal.
Then, continue to the present invention the specific steps of smart home control method for playing back detected based on sleep state into
Row further instruction.
Described based in the smart home control method for playing back of sleep state detection:
The resolution ratio of the high definition scene image it is related to the complexity of TV Week enveloping field scape including:TV Week enveloping field scape
Complexity is higher, and the resolution ratio of the high definition scene image is bigger, and the complexity of TV Week enveloping field scape is lower, the high definition scene
The resolution ratio of image is smaller;
Wherein, the resolution bits of the high definition scene image are between resolution ratio upper limit threshold and resolution limit threshold value.
Described based in the smart home control method for playing back of sleep state detection, further include:
High definition scene image is received, scene is carried out to high definition scene image and judges to determine pixel in high definition scene image
The distribution situation of pixel value, based on the distribution situation to each pixel in high definition scene image into row pixel value analyze with
Determine whether for noise spot;
The analysis result of the noise spot analytical equipment is received, noise spot each in high definition scene image is formed into multiple make an uproar
Sound area domain, determines the area and shape of each noise region, and by the Area summing of each noise region to obtain noise range
The domain gross area;
When the noise region gross area is more than or equal to preset area threshold value, start adaptive-filtering equipment, in noise region
When the gross area is less than preset area threshold value, adaptive-filtering equipment is closed, by high definition scene image directly as pending image
It is input to image smoothing equipment;
Using adaptive-filtering equipment, after being activated, adaptive-filtering processing is carried out to high definition scene image to obtain
Pending image;The adaptive-filtering equipment includes region split cells, module selecting unit, filtering execution unit and image
Assembled unit, region split cells are used to receive the shape of each noise region, the shape based on each noise region
Each noise region is split into multiple benchmark subregions by geometric properties;
Picture smooth treatment is carried out to pending image to obtain smoothed image;Wherein, it is described to determine target therein
Whether the corresponding target of image is that human body target includes:It determines multiple target subgraphs in smoothed image, determines each mesh
The area of subgraph is marked, by the area of the area of each target subgraph divided by smoothed image to obtain target area percentage
Than when target area percentage is fallen in the range of default human body area percentage, determining the corresponding target of target subgraph
For human body target;
Wherein, in high definition scene image, the region except multiple noise regions is non-noise region;High definition scene image
The distribution situation of interior pixel pixel value includes determining this for region of each pixel in high definition scene image
The pixel value range that pixel pixel value should belong to;The module selecting unit is connect with the region split cells, for pair
Each noise region, each benchmark subregion after being split for it, selects corresponding medium filtering template, the filtering
Execution unit is connect with the module selecting unit, for each noise region, each benchmark after being split for it
Subregion takes the medium filtering template of selection to perform medium filtering, to obtain each sub-regions filter pattern, and by each height
Region filter pattern is combined into filtered noise region subgraph;Described image assembled unit connects with the filtering execution unit
It connects, for receiving each noise region subgraph, and non-noise region is combined to be treated with each noise region subgraph
Handle image;
Wherein, the shape of each benchmark subregion is rectangular, round or linear, when shape is rectangular, each base
The size of quasi- subregion is 3 × 3,5 × 5 or 7 × 7, when shape is round, the radius of each benchmark subregion is 3 pixels,
5 pixels or 7 pixels, when shape is linear, each benchmark subregion is an one-dimensional pixel set.
Described based in the smart home control method for playing back of sleep state detection, further include:Prestore described point
Resolution upper limit threshold and the resolution limit threshold value.
Described based in the smart home control method for playing back of sleep state detection, further include:
Daily each period of TV is determined based on the on off state of the switchgear of TV in daily each period
Frequency of use;
The frequency of use of daily each period is ranked up, by frequency of use be more than or equal to predeterminated frequency threshold value when
Between section as frequent usage time interval, frequency of use is less than period of predeterminated frequency threshold value as free time usage time interval,
And match current time with daily each period, when being matched when current time with some idle usage time interval,
Send out closing pre-warning signal;
Pre-warning signal will be closed to include on the screen of TV waiting for the confirmation of TV user or denying, the switch of TV
When equipment is displayed up to default display time threshold in closing pre-warning signal and cannot get the input signal denied of TV user,
Closing television.
In addition, image filtering, i.e., press down the noise of target image under conditions of image detail feature is retained as possible
System, is indispensable operation in image preprocessing, and the quality for the treatment of effect will directly influence subsequent image processing and divide
The validity and reliability of analysis.
Not perfect due to imaging system, transmission medium and recording equipment etc., digital picture is in its formation, transmission log mistake
It is often polluted in journey by a variety of noises.In addition, certain links in image procossing when the picture object inputted and are not so good as pre-
Also noise can be introduced when thinking in result images.These noises often show as one and cause the isolated of stronger visual effect on the image
Pixel or block of pixels.Generally, noise signal it is uncorrelated to the object to be studied it occur with useless message form, upset figure
The observable information of picture.For data image signal, psophometer is either large or small extreme value, these extreme values are acted on by plus-minus
On the true gray value of image pixel, the interference of bright, dim spot is caused to image, greatly reduces picture quality, influence image restoration,
The progress of the subsequent work such as segmentation, feature extraction, image identification.A kind of effective wave filter for inhibiting noise is constructed to must take into consideration
Two basic problems:The noise in target and background can effectively be removed;Meanwhile can protect well image object shape,
Size and specific geometry and topological features.
One kind in common image filtering pattern is nonlinear filter, it is, in general, that when signal spectrum and noise frequency
Such as the noise as caused by mission nonlinear or there are non-gaussian to make an uproar when composing aliasing or when containing nonadditivity noise in signal
Sound etc.), traditional linear filter technology, if Fourier converts, while noise is filtered out, always blurred picture is thin in some way
Section (such as edge) and then the extraction property reduction for leading to the positioning accuracy and feature as linear character.And nonlinear filter is
Based on a kind of Nonlinear Mapping relationship to input signal, often a certain specific noise approx can be mapped as zero and retained
Signal wants feature, thus it can overcome the shortcoming of linear filter to a certain extent.
Smart home broadcasting control system using the present invention based on sleep state detection, in the prior art can not
It is automatically closed faced spectators have been enter into the TV under sleep state the technical issues of, by sleep state identification equipment, connects
The each target subgraph for being determined as human body target in acquisition image is received, the change of each target subgraph is determined based on time shaft
Change amplitude, when amplitude of variation is less than default variation amplitude, the current state for determining the human body target of corresponding target subgraph is
Sleep state when all current states for being determined as human body target are all sleep state, send out sleep control signal, otherwise sends out
Go out non-sleep control signal;TV current switch states are detected by tv state detection device;Pass through wireless control apparatus
It is connect respectively with sleep state identification equipment and tv state detection device, for being TV open shape in the current state of TV
State and when receiving sleep control signal, the switchgear of chain road direction TV, which is sent, by radio communication closes drive signal.
It is understood that although the present invention has been disclosed in the preferred embodiments as above, above-described embodiment not 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 variation.Therefore, every content without departing from technical solution of the present invention, 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 range of technical solution of the present invention protection
It is interior.