CN106757976B - Washing machine and its control method of washing and device based on image recognition clothing volume - Google Patents
Washing machine and its control method of washing and device based on image recognition clothing volume Download PDFInfo
- Publication number
- CN106757976B CN106757976B CN201710047571.4A CN201710047571A CN106757976B CN 106757976 B CN106757976 B CN 106757976B CN 201710047571 A CN201710047571 A CN 201710047571A CN 106757976 B CN106757976 B CN 106757976B
- Authority
- CN
- China
- Prior art keywords
- volume
- clothing
- image
- laundry item
- washing
- 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
-
- D—TEXTILES; PAPER
- D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
- D06F—LAUNDERING, DRYING, IRONING, PRESSING OR FOLDING TEXTILE ARTICLES
- D06F33/00—Control of operations performed in washing machines or washer-dryers
-
- D06F2202/10—
Abstract
The invention discloses a kind of washing machine and its control method of washing and device based on image recognition clothing volume, it the described method comprises the following steps: obtaining the image of multiple laundry items；The image of multiple laundry items is handled, to establish laundry item volume-based model；Obtain the neural network parameter of description clothing volume characteristic；The volume of laundry item is determined according to laundry item volume-based model and neural network parameter, and washing parameter is generated according to the volume of laundry item, and washing machine is controlled according to washing parameter and is washed.According to the method for the present invention, using image recognition and neural network, the volume of laundry item can be obtained automatically, is washed so as to intelligently control washing machine.
Description
Technical field
The present invention relates to field of home appliance technology, in particular to a kind of washing machine washing based on image recognition clothing volume
Wash control method, a kind of washing machine controlling arrangement in washing and a kind of washing machine based on image recognition clothing volume.
Background technique
With the continuous improvement of people's living standards, washing machine comes into huge numbers of families, become people's daily life
In it is essential as household electrical appliance.When user carries out washing clothing using washing machine, the weight of the clothing in washing machine drum
Amount can become the factor for influencing washing effect with volume.Due to the difference of laundry, same weight may be such that not
There may be very big difference with volume corresponding to clothing, for example, down jackets just belong to light-weight but bulky type.
Currently, the scheme that washing machine is usually taken is to determine mode of washing by being weighed to clothing, and have ignored
The volume of clothing.Therefore, washing machine how is made intelligently to judge the problem of volume of laundry item is current urgent need to resolve.
Summary of the invention
The present invention is directed to solve one of the technical problem in above-mentioned technology at least to a certain extent.For this purpose, of the invention
One purpose is to propose a kind of control method of washing of the washing machine based on image recognition clothing volume, can obtain automatically to be washed
The volume of clothing is washed so as to intelligently control washing machine.
Second object of the present invention is to propose a kind of washing control dress of the washing machine based on image recognition clothing volume
It sets.
Third object of the present invention is to propose a kind of washing machine.
In order to achieve the above objectives, first aspect present invention embodiment proposes a kind of washing machine and is based on image recognition clothing body
Long-pending control method of washing, comprising the following steps: obtain the image of multiple laundry items；To the image of the multiple laundry item
It is handled, to establish laundry item volume-based model；Obtain the neural network parameter of description clothing volume characteristic；According to it is described to
Washed volume mould and the neural network parameter determine the volume of the laundry item, and according to the volume of the laundry item
Washing parameter is generated, and the washing machine is controlled according to the washing parameter and is washed.
The control method of washing of washer image identification clothing volume according to an embodiment of the present invention, firstly, obtaining multiple
Then the image of laundry item is handled the image of multiple laundry items, to establish laundry item volume-based model, and obtain
The neural network parameter of description clothing volume characteristic is taken, finally, determining according to laundry item volume-based model and neural network parameter
The volume of laundry item, and according to the volume of laundry item generate washing parameter, and according to washing parameter control washing machine into
Row washing.This method utilizes image recognition and neural network, can obtain the volume of laundry item, automatically so as to intelligently
Control washing machine is washed.
In addition, washing controlling party of the washing machine proposed according to that above embodiment of the present invention based on image recognition clothing volume
Method can also have the following additional technical features:
According to one embodiment of present invention, the neural network parameter of description clothing volume characteristic is obtained, comprising: obtain more
The image of a sample clothing；The image of the multiple sample clothing is handled, to establish sample clothing volume-based model；Pass through
Neural network is trained the sample clothing volume-based model, to obtain the neural network ginseng of the description clothing volume characteristic
Number.
According to one embodiment of present invention, the image of multiple sample clothings or the image of laundry item are obtained, comprising:
After the sample clothing or the laundry item are put into washing machine inner tub, alternately to the sample clothing or described wait do washing
Object is taken pictures, controls the interior bucket rotation to shake the sample clothing or the laundry item, and until taking pictures, number reaches
Predetermined amount, to obtain the image of the sample clothing of the predetermined amount or the image of the laundry item.
Further, the image of the image to multiple sample clothings or the laundry item is handled, to establish
Sample clothing or laundry item volume-based model, comprising: the image of the image or laundry item of each sample clothing of removal
Clothing background, and binaryzation is carried out to the image of the sample clothing or the image of laundry item each of after removal clothing background
Processing, is only retained multiple sample clothing binary images or laundry item binary image in clothing region；By multiple institutes
It states sample clothing binary image or the laundry item binary image is overlapped, to obtain a secondary composograph；It obtains
The gray value of different pixels in the composograph, to establish the sample clothing or the laundry item volume-based model.
According to one embodiment of present invention, the volume of the laundry item is determined, and according to the body of the laundry item
Product generates washing parameter, comprising: determines volume grade locating for the volume of the laundry item, and according to the laundry item
Volume grade locating for volume generates corresponding washing parameter.
In order to achieve the above objectives, second aspect of the present invention embodiment proposes a kind of washing machine and is based on image recognition clothing body
Long-pending controlling arrangement in washing, comprising: photographing module, the photographing module are used to obtain the image of multiple laundry items；Model mould
Block, the modeling module is for handling the image of the multiple laundry item, to establish laundry item volume-based model；It obtains
Modulus block, the neural network parameter for obtaining module and being used to obtain description clothing volume characteristic；Main control module, the master control mould
Block is used to determine the volume of the laundry item according to the laundry item volume-based model and the neural network parameter, and according to
The volume of the laundry item generates washing parameter, and controls the washing machine according to the washing parameter and washed.
Controlling arrangement in washing of the washing machine according to an embodiment of the present invention based on image recognition clothing volume, by imaging mould
Block obtains the image of multiple laundry items, and is handled by image of the modeling module to multiple laundry items, with establish to
Then washed volume-based model obtains the deep neural network parameter of description clothing volume characteristic by obtaining module, finally,
The volume for determining laundry item, and root are obtained according to laundry item volume-based model and deep neural network parameter by main control module
Washing parameter is generated according to the volume of laundry item, and washing machine is controlled according to washing parameter and is washed.The device utilizes figure
As identification and deep neural network, the volume of laundry item can be obtained automatically, carried out so as to intelligently control washing machine
Washing.
In addition, the washing machine proposed according to that above embodiment of the present invention controls dress based on the washing of image recognition clothing volume
Setting can also have the following additional technical features:
According to one embodiment of present invention, the photographing module is also used to obtain the image of multiple sample clothings, described
Modeling module is also used to handle the image of the multiple sample clothing, described to obtain to establish sample clothing volume-based model
Modulus block is trained the sample clothing volume-based model by neural network, to obtain the description clothing volume characteristic
Neural network parameter.
According to one embodiment of present invention, the image of the sample clothing or laundry item are being put into washing machine inner tub
Afterwards, the photographing module is alternately passed through to take pictures to the image or laundry item of the sample clothing, pass through main control module
The interior bucket rotation is controlled to shake the sample clothing or laundry item, number reaches predetermined amount until taking pictures, to obtain
State the image of the sample clothing of predetermined amount or the image of laundry item.
Further, the modeling module is used to remove the image of each sample clothing or the image of laundry item
Clothing background, and binaryzation is carried out to the image of the sample clothing or the image of laundry item each of after removal clothing background
Processing is only retained multiple sample clothing binary images or laundry item binary image and multiple in clothing region
The sample clothing binary image or the laundry item binary image are overlapped, and to obtain a secondary composograph, are obtained
The gray value of pixel different in the composograph is taken, to establish the sample clothing or the laundry item volume mould
Type.
According to one embodiment of present invention, the main control module is specifically used for determining the volume of the laundry item, and
Washing parameter is generated according to the volume of the laundry item, comprising: described main control module determines the volume institute of the laundry item
The volume grade at place, and the volume grade according to locating for the volume of the laundry item generates corresponding washing parameter.
In order to achieve the above objectives, third aspect present invention embodiment proposes a kind of washing machine comprising above-mentioned laundry
Controlling arrangement in washing of the machine based on image recognition clothing volume.
The washing machine of the embodiment of the present invention controls dress based on the washing of image recognition clothing volume by above-mentioned washing machine
It sets, using image recognition and neural network, the volume of laundry item can be obtained, automatically so as to intelligently control washing machine
It is washed.
Detailed description of the invention
Fig. 1 is the process of control method of washing of the washing machine according to an embodiment of the present invention based on image recognition clothing volume
Figure；
Fig. 2 is the flow chart of acquisition neural network parameter accord to a specific embodiment of that present invention；
Fig. 3 is washing controlling party of the washing machine based on image recognition clothing volume accord to a specific embodiment of that present invention
The flow chart of method；
Fig. 4 is the box of controlling arrangement in washing of the washing machine according to an embodiment of the present invention based on image recognition clothing volume
Schematic diagram；And
The block diagram of Fig. 5 washing machine according to an embodiment of the present invention.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to is used to explain the present invention, and is not considered as limiting the invention.
The washing machine of the embodiment of the present invention is described and its washing based on image recognition clothing volume with reference to the accompanying drawing
Control method and device.
Fig. 1 is the process of control method of washing of the washing machine according to an embodiment of the present invention based on image recognition clothing volume
Figure.As shown in Figure 1, control method of washing of the washing machine of the embodiment of the present invention based on image recognition clothing volume includes following step
It is rapid:
S1 obtains the image of multiple laundry items.
Specifically, in can alternately being taken pictures after laundry item is put into washing machine inner tub, being controlled to laundry item
Bucket rotation is to shake laundry item, and until taking pictures, number reaches predetermined amount, to obtain the image of the laundry item of predetermined amount.Its
In, predetermined amount can be demarcated according to the actual situation, here with no restrictions.
Specifically, after laundry item is put into washing machine inner tub by user, it, can during washing machine plus water running
To be taken pictures the washing machine inner tub during current laundry to obtain by the photographic device being arranged in washing machine drum
The image of laundry item.If number of taking pictures is not up to predetermined amount, washing machine inner tub rotation is controlled to shake laundry item, is handed over
It alternately takes pictures to laundry item, until taking pictures, number reaches predetermined amount.
S2 handles the image of multiple laundry items, to establish laundry item volume-based model.
Specifically, can remove the clothing background of the image of each laundry item, and to each of after removal clothing background to
Washed image carries out binary conversion treatment, is only retained multiple laundry item binary images in clothing region, will be multiple
Laundry item binary image is overlapped, and to obtain a secondary composograph, obtains the ash of pixel different in composograph
Angle value, to establish laundry item volume-based model.
Specifically, can remove the clothing of the image of each laundry item after obtaining the image of laundry item of predetermined amount
Object background, how clothing background specifically removes can be not detailed here using common morphologic method etc..It then can be right
The image of each laundry item after removing clothing background carries out binary conversion treatment, is only retained the multiple to be washed of clothing region
Clothing binary image.Wherein, the gray value in settable laundry item region is 1, and the gray value of background area is 0, binaryzation
Image afterwards shows the effect of black and white.Multiple laundry item binary images are overlapped, a secondary composograph can be obtained,
Wherein, the gray value of each pixel in composograph can represent the pixel occur in the composite image clothing region time
Number.Therefore, composograph can reflect out laundry item volume, can will acquire the synthesis of the gray value of different pixels
Image is as laundry item volume-based model.It should be appreciated that the gray value of pixel is maximum in the central region of composograph, close
It is also most bright at the brightness of image, this is because the central region of each laundry item binary image typicallys represent laundry item
Region causes the gray value of the pixel of the central region of composograph larger.In the surrounding of the central region of composograph, as
The gray value of vegetarian refreshments is gradually reduced, this is because the surrounding of the central region of each laundry item binary image has plenty of
Laundry item region, has plenty of background area, if the gray value of the pixel of composograph be 0, then it represents that laundry item from
This region is not appeared in.It can be seen that the maximum region of the gray value of pixel can indicate laundry item one in composograph
Fixed existing range, the biggish region of gray value can indicate the range that laundry item there may be, gray value be 0 can indicate to
This range is not present in washed affirmative.
S3 obtains the neural network parameter of description clothing volume characteristic.
Specifically, the image of multiple sample clothings can be obtained, and the image of multiple sample clothings is handled, to establish
Sample clothing volume-based model, and sample clothing volume-based model is trained by neural network, clothing body is described to obtain
The neural network parameter of product feature.
Specifically, in can alternately being taken pictures after sample clothing is put into washing machine inner tub, being controlled to sample clothing
Bucket rotation is with jitter samples clothing, and until taking pictures, number reaches predetermined amount, to obtain the image of the sample clothing of predetermined amount.
Specifically, before current laundry process, such as can need to obtain description clothing body before washing machine factory
The neural network parameter of product feature.During to sample image training, sample clothing is put into washing machine inner tub by user,
During washing machine adds water running, can by the photographic device that is arranged in washing machine drum to current laundry during
Washing machine inner tub is taken pictures to obtain the image of sample clothing.If number of taking pictures is not up to predetermined amount, washing machine is controlled
Interior bucket is rotated with jitter samples clothing, is alternately taken pictures to sample clothing, until taking pictures, number reaches predetermined amount.
Specifically, it can remove the clothing background of the image of each sample clothing, and to each sample after removal clothing background
The image of this clothing carries out binary conversion treatment, is only retained multiple sample clothing binary images in clothing region, will be multiple
Sample clothing binary image is overlapped, and to obtain a secondary composograph, obtains the ash of pixel different in composograph
Angle value, to establish sample clothing volume-based model.
Specifically, can remove the clothing of the image of each sample clothing after obtaining the image of sample clothing of predetermined amount
Object background, how clothing background specifically removes can be not detailed here using common morphologic method etc..It then can be right
The image of each sample clothing after removing clothing background carries out binary conversion treatment, is only retained multiple samples in clothing region
Clothing binary image.Wherein, the gray value in settable sample clothing region is 1, and the gray value of background area is 0, binaryzation
Image afterwards shows the effect of black and white.Multiple sample clothing binary images are overlapped, a secondary composograph can be obtained,
Wherein, the gray value of each pixel in composograph can represent the pixel occur in the composite image clothing region time
Number.Therefore, composograph can reflect out sample clothing volume, can will acquire the synthesis of the gray value of different pixels
Image is as sample clothing volume-based model.It should be appreciated that the gray value of pixel is maximum in the central region of composograph, close
It is also most bright at the brightness of image, this is because the central region of each sample clothing binary image typicallys represent sample clothing
Region causes the gray value of the pixel of the central region of composograph larger.In the surrounding of the central region of composograph, as
The gray value of vegetarian refreshments is gradually reduced, this is because the surrounding of the central region of each sample clothing binary image has plenty of
Sample clothing region, has plenty of background area, if the gray value of the pixel of composograph be 0, then it represents that sample clothing from
This region is not appeared in.It can be seen that the maximum region of the gray value of pixel can indicate sample clothing one in composograph
Fixed existing range, the biggish region of gray value can indicate the range that sample clothing there may be, and gray value can indicate sample for 0
This range is not present in this clothing certainly.
It further, can be by neural network to the volume-based model of sample clothing after establishing sample clothing volume-based model
It is trained, to obtain the neural network parameter of corresponding relationship between description volume-based model and true volume.
In order to make those skilled in the art more clearly understand the present invention, Fig. 2 is obtaining accord to a specific embodiment of that present invention
Take the flow chart of neural network parameter.As shown in Fig. 2, the acquisition neural network parameter of the specific embodiment of the invention can include:
Sample clothing is put into washing machine inner tub by S301.
S302 takes pictures to interior bucket by photographic device.
S303, judges whether the number taken pictures reaches predetermined amount.If so, executing step S304；If not, executing step
S305。
S304 obtains the image of multiple sample clothings.
S305, control washing machine rotation.
S306 handles the image of multiple sample clothings.
S307 establishes sample clothing volume-based model.
S308 is trained by volume-based model of the neural network to sample.
S309 obtains the neural network parameter of description clothing volume characteristic.
S4 determines the volume of laundry item according to laundry item volume mould and neural network parameter, and according to laundry item
Volume generate washing parameter, and according to washing parameter control washing machine washed.
Specifically, it may be determined that volume grade locating for the volume of laundry item, and according to locating for the volume of laundry item
Volume grade generates corresponding washing parameter.
It specifically, can be according to laundry item volume-based model and neural network parameter after establishing laundry item volume-based model
Determine volume grade locating for the volume of laundry item, and the volume grade according to locating for laundry item generates corresponding washing ginseng
Number, washing machine can intelligently select laundry mode according to washing parameter, and control washing machine and washed.
It should be noted that in practical applications, after obtaining volume grade locating for laundry item, can also pass through setting
Weight sensor on washing machine obtains the weight of laundry item, and the volume grade according to locating for laundry item and wait do washing
The weight of object generates corresponding washing parameter, and washing machine can intelligently select laundry mode according to washing parameter, and control laundry
Machine is washed.
For example, natural feather stock just belongs to clothing light-weight but that volume grade is big, volume locating for down jackets is being obtained
After grade, it is also necessary to which the weight for obtaining down jackets, volume grade and weight according to locating for down jackets produce corresponding washing
Parameter, washing machine can intelligently select the laundry mode of small, the more rinsing of such as revolving speed according to washing parameter, and control washing machine into
Row washing, can reach optimal washing effect in this way.
Further, Fig. 3 is washing machine accord to a specific embodiment of that present invention based on image recognition clothing volume
The flow chart of control method of washing.As shown in figure 3, control method of washing of the washing machine based on image recognition clothing volume may include
Following steps:
Laundry item is put into washing machine inner tub by S101.
S102 takes pictures to interior bucket by photographic device.
S103, judges whether the number taken pictures reaches predetermined amount.If so, executing step S104；If not, executing step
S105。
S104 obtains the image of multiple laundry items.
S105, control washing machine rotation.
S106 handles the image of multiple laundry items.
S107 establishes laundry item volume-based model.
S108 obtains the volume grade of laundry item according to the volume-based model of laundry item and neural network parameter.
S109 obtains the weight of laundry item.
S110 generates washing parameter according to the volume grade and weight of laundry item.
S111 controls washing machine according to washing parameter and is washed.
In conclusion the control method of washing of washer image identification clothing volume according to an embodiment of the present invention, firstly,
The image of multiple laundry items is obtained, then, the image of multiple laundry items is handled, to establish laundry item volume mould
Type, and the neural network parameter of description clothing volume characteristic is obtained, finally, being joined according to laundry item volume-based model and neural network
Number determines the volume of laundry item, and generates washing parameter according to the volume of laundry item, and wash according to washing parameter control
Clothing machine is washed.This method utilizes image recognition and neural network, can obtain the volume of laundry item automatically, so as to
Intelligently control washing machine is washed.
Fig. 4 is the box of controlling arrangement in washing of the washing machine according to an embodiment of the present invention based on image recognition clothing volume
Schematic diagram.As shown in figure 4, controlling arrangement in washing of the washing machine of the embodiment of the present invention based on image recognition clothing volume, comprising:
Photographing module 10, obtains module 30 and main control module 40 at modeling module 20.
Wherein, photographing module 10 is used to obtain the image of multiple laundry items.Modeling module 20 is used for multiple wait do washing
The image of object is handled, to establish laundry item volume-based model.It obtains module 30 and is used to obtain description clothing volume characteristic
Neural network parameter.Main control module 40 is used to determine the body of laundry item according to laundry item volume-based model and neural network parameter
Product, and washing parameter is generated according to the volume of laundry item, and washing machine is controlled according to washing parameter and is washed.
Specifically, photographing module 10 can be alternately passed through to laundry item after laundry item is put into washing machine inner tub
It taken pictures, control interior bucket rotation by main control module 40 to shake laundry item, number reaches predetermined amount until taking pictures, to obtain
Take the image of the laundry item of predetermined amount.Wherein, predetermined amount can be demarcated according to the actual situation, here with no restrictions.
Specifically, after laundry item is put into washing machine inner tub by user, it, can during washing machine plus water running
Taken pictures to washing machine inner tub by the photographing module 10 being arranged in washing machine drum to obtain the image of laundry item.
If number of taking pictures is not up to predetermined amount, washing machine inner tub can be controlled by main control module 40 and is rotated to shake laundry item,
It alternately passes through photographic device 10 to take pictures to laundry item, until taking pictures, number reaches predetermined amount.
Specifically, the clothing background of the image of each laundry item can be can remove by modeling module 20, and to removal clothing
The image of each laundry item after object background carries out binary conversion treatment, is only retained multiple laundry items two in clothing region
Multiple laundry item binary images are overlapped by value image, to obtain a secondary composograph, are obtained in composograph not
The gray value of same pixel, to establish laundry item volume-based model.
Specifically, after obtaining the image of laundry item of predetermined amount by photographic device 10, by establishing module 20
The image of multiple laundry items is handled, can remove the clothing background of the image of each laundry item, clothing background is specific
How removing can be not detailed here using common morphologic method etc..It then can be to every after removal clothing background
The image of a laundry item carries out binary conversion treatment, is only retained multiple laundry item binary images in clothing region.Its
In, the gray value in settable laundry item region is 1, and the gray value of background area is 0, and the image after binaryzation shows black and white
Effect.Multiple laundry item binary images are overlapped, a secondary composograph can be obtained, wherein in composograph
The gray value of each pixel can represent the number that clothing region occurs in the composite image in the pixel.Therefore, composograph
It can reflect out laundry item volume, the composograph of the gray value of different pixels can be will acquire as laundry item
Volume-based model.It should be appreciated that the gray value of pixel is maximum in the central region of composograph, the brightness of composograph is also most
It is bright, this is because the central region of each laundry item binary image typicallys represent the region of laundry item, lead to composite diagram
The gray value of the pixel of the central region of picture is larger.It is in the gray value of the surrounding of the central region of composograph, pixel
It is gradually reduced, this is because the surrounding of the central region of each laundry item binary image has plenty of laundry item region,
Has plenty of background area, if the gray value of the pixel of composograph is 0, then it represents that laundry item is not from appearing in this area
Domain.It can be seen that the maximum region of the gray value of pixel can indicate the range that laundry item certainly exists in composograph, ash
The biggish region of angle value can indicate the range that laundry item there may be, and gray value can indicate that laundry item is not deposited certainly for 0
In this range.
In one embodiment of the invention, photographing module 10 is also used to obtain the image of multiple sample clothings, models mould
Block 20 is also used to handle the image of multiple sample clothings, to establish sample clothing volume-based model, obtains module 30 and passes through
Neural network is trained sample clothing volume-based model, to obtain the neural network parameter of description clothing volume characteristic.
Specifically, photographing module 10 can be alternately passed through to sample clothing after sample clothing is put into washing machine inner tub
It taken pictures, control interior bucket rotation by main control module 40 with jitter samples clothing, number reaches predetermined amount until taking pictures, to obtain
Take the image of the sample clothing of predetermined amount.Wherein, predetermined amount can be demarcated according to the actual situation, here with no restrictions.
Specifically, after sample clothing is put into washing machine inner tub by user, it, can during washing machine plus water running
Taken pictures to washing machine inner tub by the photographing module 10 being arranged in washing machine drum to obtain the image of sample clothing.
If number of taking pictures is not up to predetermined amount, washing machine inner tub can be controlled by main control module 40 and is rotated with jitter samples clothing,
It alternately passes through photographic device 10 to take pictures to sample clothing, until taking pictures, number reaches predetermined amount.
Specifically, the clothing background of the image of each sample clothing can be can remove by modeling module 20, and to removal clothing
The image of each sample clothing after object background carries out binary conversion treatment, is only retained multiple sample clothings two in clothing region
Multiple sample clothing binary images are overlapped by value image, to obtain a secondary composograph, are obtained in composograph not
The gray value of same pixel, to establish sample clothing volume-based model.
Specifically, after obtaining the image of sample clothing of predetermined amount by photographic device 10, by establishing module 20
The image of multiple sample clothings is handled, can remove the clothing background of the image of each sample clothing, clothing background is specific
How removing can be not detailed here using common morphologic method etc..It then can be to every after removal clothing background
The image of a sample clothing carries out binary conversion treatment, is only retained multiple sample clothing binary images in clothing region.Its
In, the gray value in settable sample clothing region is 1, and the gray value of background area is 0, and the image after binaryzation shows black and white
Effect.Multiple sample clothing binary images are overlapped, a secondary composograph can be obtained, wherein in composograph
The gray value of each pixel can represent the number that clothing region occurs in the composite image in the pixel.Therefore, composograph
It can reflect out sample clothing volume, the composograph of the gray value of different pixels can be will acquire as sample clothing
Volume-based model.It should be appreciated that the gray value of pixel is maximum in the central region of composograph, the brightness of composograph is also most
It is bright, this is because the central region of each sample clothing binary image typicallys represent the region of sample clothing, lead to composite diagram
The gray value of the pixel of the central region of picture is larger.It is in the gray value of the surrounding of the central region of composograph, pixel
It is gradually reduced, this is because the surrounding of the central region of each sample clothing binary image has plenty of sample clothing region,
Has plenty of background area, if the gray value of the pixel of composograph is 0, then it represents that sample clothing is not from appearing in this area
Domain.It can be seen that the maximum region of the gray value of pixel can indicate the range that sample clothing certainly exists in composograph, ash
The biggish region of angle value can indicate the range that sample clothing there may be, and gray value can indicate that sample clothing is not deposited certainly for 0
In this range.
Further, it after establishing sample clothing volume-based model by modeling module 20, obtains module 30 and passes through nerve net
Network is trained the volume-based model of sample clothing, to obtain the nerve of corresponding relationship between description volume-based model and true volume
Network parameter.
In one embodiment of the invention, volume locating for the volume of laundry item etc. can be determined by main control module
Grade, and the volume grade according to locating for the volume of laundry item generates corresponding washing parameter.
Specifically, after establishing laundry item volume-based model by modeling module 20, can by main control module 40 according to
Washed volume-based model and neural network parameter determine volume grade locating for the volume of laundry item, and according to laundry item institute
The volume grade at place generates corresponding washing parameter, and washing machine can intelligently select laundry mode according to washing parameter, and control
Washing machine is washed.
It should be noted that in practical applications, after obtaining volume grade locating for laundry item, can also pass through setting
Weight sensor on washing machine obtains the weight of laundry item, and the grade of the volume according to locating for laundry item and to be washed
The weight of clothing generates corresponding washing parameter, and washing machine can intelligently select laundry mode according to washing parameter, and control and wash
Clothing machine is washed.
For example, natural feather stock just belongs to clothing light-weight but that volume grade is big, volume locating for down jackets is being obtained
After grade, it is also necessary to which the weight for obtaining down jackets, volume grade and weight according to locating for down jackets produce corresponding washing
Parameter, washing machine can intelligently select the laundry mode of small, the more rinsing of such as revolving speed according to washing parameter, and control washing machine into
Row washing, can reach optimal washing effect in this way.
Controlling arrangement in washing of the washing machine according to an embodiment of the present invention based on image recognition clothing volume, by imaging mould
Block obtains the image of multiple laundry items, and is handled by image of the modeling module to multiple laundry items, with establish to
Then washed volume-based model obtains the neural network parameter of description clothing volume characteristic by obtaining module, finally, passing through
Main control module determines the volume of laundry item according to laundry item volume-based model and neural network parameter, and according to laundry item
Volume generates washing parameter, and controls washing machine according to washing parameter and washed.The device utilizes image recognition and nerve
Network can obtain the volume of laundry item automatically, be washed so as to intelligently control washing machine.
Based on the above embodiment, the invention also provides a kind of washing machines 1000.
Fig. 5 is the block diagram of washing machine according to an embodiment of the present invention.As shown in figure 5, the embodiment of the present invention is washed
Clothing machine 1000 may include controlling arrangement in washing 100 of the above-mentioned washing machine based on image recognition clothing volume.
It should be noted that the undisclosed details in the washing machine 1000 of the embodiment of the present invention, please refers to of the invention real
Apply the washing machine of example based on details disclosed in the controlling arrangement in washing 100 of image recognition clothing volume, specifically here no longer
It is described in detail.
Washing machine according to an embodiment of the present invention, the washing control by above-mentioned washing machine based on image recognition clothing volume
Device processed can obtain the volume of laundry item automatically, wash so as to intelligently control using image recognition and neural network
Clothing machine is washed.
In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", " length ", " width ",
" thickness ", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside", " up time
The orientation or positional relationship of the instructions such as needle ", " counterclockwise ", " axial direction ", " radial direction ", " circumferential direction " be orientation based on the figure or
Positional relationship is merely for convenience of description of the present invention and simplification of the description, rather than the device or element of indication or suggestion meaning must
There must be specific orientation, be constructed and operated in a specific orientation, therefore be not considered as limiting the invention.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include one or more of the features.In the description of the present invention, the meaning of " plurality " is two or more,
Unless otherwise specifically defined.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc.
Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be mechanical connect
It connects, is also possible to be electrically connected；It can be directly connected, can also can be in two elements indirectly connected through an intermediary
The interaction relationship of the connection in portion or two elements.It for the ordinary skill in the art, can be according to specific feelings
Condition understands the concrete meaning of above-mentioned term in the present invention.
In the present invention unless specifically defined or limited otherwise, fisrt feature in the second feature " on " or " down " can be with
It is that the first and second features directly contact or the first and second features pass through intermediary mediate contact.Moreover, fisrt feature exists
Second feature " on ", " top " and " above " but fisrt feature be directly above or diagonally above the second feature, or be merely representative of
First feature horizontal height is higher than second feature.Fisrt feature can be under the second feature " below ", " below " and " below "
One feature is directly under or diagonally below the second feature, or is merely representative of first feature horizontal height less than second feature.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (11)
1. a kind of control method of washing of washing machine based on image recognition clothing volume, which comprises the following steps:
Obtain the image of multiple laundry items；
The image of the multiple laundry item is handled, to establish laundry item volume-based model；
The neural network parameter of description clothing volume characteristic is obtained, the neural network parameter is by neural network to sample clothing
What the volume-based model of object was trained, the neural network parameter is used to characterize the volume-based model and true body of laundry item
Corresponding relationship between product；
The volume of the laundry item is determined according to the laundry item volume-based model and the neural network parameter, and according to institute
The volume for stating laundry item generates washing parameter, and controls the washing machine according to the washing parameter and washed.
2. control method of washing of the washing machine according to claim 1 based on image recognition clothing volume, which is characterized in that
Obtain the neural network parameter of description clothing volume characteristic, comprising:
Obtain the image of multiple sample clothings；
The image of the multiple sample clothing is handled, to establish sample clothing volume-based model；
The sample clothing volume-based model is trained by neural network, to obtain the mind of the description clothing volume characteristic
Through network parameter.
3. control method of washing of the washing machine according to claim 1 or 2 based on image recognition clothing volume, feature exist
In obtaining the image of multiple sample clothings or the image of laundry item, comprising:
After the sample clothing or the laundry item are put into washing machine inner tub, alternately to the sample clothing or described
Laundry item taken pictures, controls the interior bucket rotation to shake the sample clothing or the laundry item, until taking pictures time
Number reaches predetermined amount, to obtain the image of the sample clothing of the predetermined amount or the image of the laundry item.
4. control method of washing of the washing machine according to claim 3 based on image recognition clothing volume, which is characterized in that
The image of image or the laundry item to multiple sample clothings is handled, to establish sample clothing or laundry item
Volume-based model, comprising:
Remove the clothing background of the image of each sample clothing or the image of laundry item, and to removal clothing background after
The image of each sample clothing or the image of laundry item carry out binary conversion treatment, are only retained the multiple of clothing region
Sample clothing binary image or laundry item binary image；
Multiple sample clothing binary images or the laundry item binary image are overlapped, to obtain a secondary conjunction
At image；
The gray value of pixel different in the composograph is obtained, to establish the sample clothing or the laundry item body
Product module type.
5. control method of washing of the washing machine described in any one of -4 based on image recognition clothing volume according to claim 1,
It is characterized in that, determining the volume of the laundry item, and washing parameter is generated according to the volume of the laundry item, comprising:
Determine volume grade locating for the volume of the laundry item, and volume according to locating for the volume of the laundry item etc.
Grade generates corresponding washing parameter.
6. a kind of controlling arrangement in washing of washing machine based on image recognition clothing volume characterized by comprising
Photographing module, the photographing module are used to obtain the image of multiple laundry items；
Modeling module, the modeling module is for handling the image of the multiple laundry item, to establish laundry item
Volume-based model；
Obtain module, the neural network parameter for obtaining module and being used to obtain description clothing volume characteristic, the neural network
Parameter is trained by volume-based model of the neural network to sample clothing, and the neural network parameter is for characterizing
Corresponding relationship between the volume-based model and true volume of laundry item；
Main control module, the main control module are used to determine institute according to the laundry item volume-based model and the neural network parameter
The volume of laundry item is stated, and washing parameter is generated according to the volume of the laundry item, and according to the washing parameter control
The washing machine is made to be washed.
7. controlling arrangement in washing of the washing machine according to claim 6 based on image recognition clothing volume, which is characterized in that
The photographing module is also used to obtain the image of multiple sample clothings, and the modeling module is also used to the multiple sample clothing
Image handled, to establish sample clothing volume-based model, the acquisition module is by neural network to the sample clothing
Volume-based model is trained, to obtain the neural network parameter of the description clothing volume characteristic.
8. controlling arrangement in washing of the washing machine according to claim 6 or 7 based on image recognition clothing volume, feature exist
In alternately passing through the photographing module pair after the image of the sample clothing or laundry item are put into washing machine inner tub
The image or laundry item of the sample clothing are taken pictures, control the interior bucket rotation by main control module to shake the sample
This clothing or laundry item, until taking pictures, number reaches predetermined amount, to obtain the image of the sample clothing of the predetermined amount
Or the image of laundry item.
9. controlling arrangement in washing of the washing machine according to claim 8 based on image recognition clothing volume, which is characterized in that
The modeling module is used to remove the clothing background of the image of each sample clothing or the image of laundry item, and to removal
The image of the image of the sample clothing or laundry item carries out binary conversion treatment each of after clothing background, is only retained clothing
The multiple sample clothing binary images or laundry item binary image of object area and multiple sample clothing binaryzations
Image or the laundry item binary image are overlapped, and to obtain a secondary composograph, are obtained in the composograph not
The gray value of same pixel, to establish the sample clothing or the laundry item volume-based model.
10. controlling arrangement in washing of the washing machine based on image recognition clothing volume according to any one of claim 6-9,
It is characterized in that, the main control module is specifically used for determining the volume of the laundry item, and according to the body of the laundry item
Product generates washing parameter, comprising:
The main control module is for determining volume grade locating for the volume of the laundry item, and according to the laundry item
Volume grade locating for volume generates corresponding washing parameter.
11. a kind of washing machine, which is characterized in that be based on image including the washing machine according to any one of claim 6-10
Identify the controlling arrangement in washing of clothing volume.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710047571.4A CN106757976B (en) | 2017-01-22 | 2017-01-22 | Washing machine and its control method of washing and device based on image recognition clothing volume |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710047571.4A CN106757976B (en) | 2017-01-22 | 2017-01-22 | Washing machine and its control method of washing and device based on image recognition clothing volume |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106757976A CN106757976A (en) | 2017-05-31 |
| CN106757976B true CN106757976B (en) | 2019-08-27 |
Family
ID=58941431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710047571.4A Active CN106757976B (en) | 2017-01-22 | 2017-01-22 | Washing machine and its control method of washing and device based on image recognition clothing volume |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106757976B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107729816A (en) * | 2017-09-15 | 2018-02-23 | 珠海格力电器股份有限公司 | Determination method, apparatus, storage medium, processor and the dish-washing machine of washing mode |
| CN108038879A (en) * | 2017-12-12 | 2018-05-15 | 众安信息技术服务有限公司 | A kind of volume of food method of estimation and its device |
| CN108594683A (en) * | 2018-04-17 | 2018-09-28 | 天门市志远信息科技有限公司 | A kind of clothing house keeper intelligent interactive system and method |
| KR102559073B1 (en) * | 2018-08-30 | 2023-07-21 | 엘지전자 주식회사 | Artificial intelligence laundry treating apparatus and method of controlling laundry treating apparatu |
| KR102559072B1 (en) | 2018-08-30 | 2023-07-21 | 엘지전자 주식회사 | Artificial intelligence laundry treating apparatus and method of controlling laundry treating apparatu |
| CN111197226B (en) * | 2018-11-19 | 2023-12-22 | 青岛海尔智能技术研发有限公司 | Washing control method and washing machine |
| CN109610132A (en) * | 2018-11-22 | 2019-04-12 | 无锡小天鹅股份有限公司 | Laundry process, device and controller |
| CN109544542B (en) * | 2018-11-29 | 2020-10-30 | 蚌埠翰邦知识产权服务有限公司 | Cleaning frequency setting platform |
| CN109267292B (en) * | 2018-12-03 | 2020-12-01 | 海宁市利得宝印染有限公司 | Automatic releasing system of laundry detergent |
| CN110331551A (en) * | 2019-05-24 | 2019-10-15 | 珠海格力电器股份有限公司 | Control method of washing, device, computer equipment and the storage medium of washing machine |
| IT201900008847A1 (en) * | 2019-06-13 | 2020-12-13 | Candy Spa | Method and system for the preparation and execution of a laundry washing cycle |
| CN112391793A (en) * | 2019-07-30 | 2021-02-23 | Tcl集团股份有限公司 | Washing machine control method, washing machine and storage medium |
| CN113201905A (en) * | 2020-01-15 | 2021-08-03 | 青岛海尔洗衣机有限公司 | Clothes volume estimation method and control method of clothes treatment equipment and clothes treatment system |
| CN111568207A (en) * | 2020-02-28 | 2020-08-25 | 佛山市云米电器科技有限公司 | Water dispenser control method, water dispenser and computer readable storage medium |
| CN113818191B (en) * | 2020-06-19 | 2023-11-14 | 青岛海尔洗衣机有限公司 | Washing control method, device, equipment and storage medium based on image recognition |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103334265A (en) * | 2013-06-19 | 2013-10-02 | 松下家电研究开发（杭州）有限公司 | Intelligent washing machine, washing control method and remote monitoring method |
| JP2015090629A (en) * | 2013-11-07 | 2015-05-11 | 日立アプライアンス株式会社 | Processing device |
| CN106149286A (en) * | 2016-08-31 | 2016-11-23 | 广东格兰仕集团有限公司 | A kind of control method for washing machine based on image recognition |
| CN106192289A (en) * | 2015-04-30 | 2016-12-07 | 青岛海尔洗衣机有限公司 | Control method for washing machine and washing machine |
| WO2017004787A1 (en) * | 2015-07-07 | 2017-01-12 | 深圳市赛亿科技开发有限公司 | Intelligent washing machine and control method thereof |
-
2017
- 2017-01-22 CN CN201710047571.4A patent/CN106757976B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103334265A (en) * | 2013-06-19 | 2013-10-02 | 松下家电研究开发（杭州）有限公司 | Intelligent washing machine, washing control method and remote monitoring method |
| JP2015090629A (en) * | 2013-11-07 | 2015-05-11 | 日立アプライアンス株式会社 | Processing device |
| CN106192289A (en) * | 2015-04-30 | 2016-12-07 | 青岛海尔洗衣机有限公司 | Control method for washing machine and washing machine |
| WO2017004787A1 (en) * | 2015-07-07 | 2017-01-12 | 深圳市赛亿科技开发有限公司 | Intelligent washing machine and control method thereof |
| CN106149286A (en) * | 2016-08-31 | 2016-11-23 | 广东格兰仕集团有限公司 | A kind of control method for washing machine based on image recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106757976A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106757976B (en) | Washing machine and its control method of washing and device based on image recognition clothing volume | |
| CN106854808A (en) | Washing machine and its control method of washing and device | |
| US11299838B2 (en) | Washing machine and cloud server setting function based on object sensing using artificial intelligence, and method for setting function | |
| CN106676822A (en) | Washing machine, washing control method and device of washing machine based on clothes natures | |
| KR20170138559A (en) | Washing machine control method and washing machine | |
| CN106048977B (en) | The control system and control method of washing machine | |
| CN107574625A (en) | Washing methods, device, storage device, mobile terminal and the washing machine of clothing | |
| WO2017004787A1 (en) | Intelligent washing machine and control method thereof | |
| CN105839355B (en) | The recognition methods of washing machine and washing machine undergarment color and device | |
| CN108866934A (en) | A kind of laundry mode control system and method based on artificial intelligence | |
| CN109064678A (en) | Shared washing machine, control method and medium | |
| CN109112774A (en) | Control method, device, storage medium, program product and the washing machine of washing machine | |
| CN105937124A (en) | Washing machine control method and system and equipment thereof | |
| CN106930055A (en) | Washing machine and for washing machine foam volume detection method and device | |
| CN106782525A (en) | Washing machine and its phonetic controller, method | |
| CN109610132A (en) | Laundry process, device and controller | |
| EP3751454B1 (en) | Method and system for preparing and performing a laundry washing cycle | |
| CN109868613A (en) | A kind of intelligent washing machine control method and system based on Internet of Things | |
| CN113463327B (en) | Clothing display method and device, electronic equipment and computer readable storage medium | |
| CN107558097A (en) | The system of selection of device for clothing processing and its washing procedure and device | |
| CN110894658B (en) | Washing control method and device and clothes treatment device | |
| CN110688953A (en) | Ironing control method and device based on operating system | |
| CN106400388A (en) | Washing equipment and control method thereof | |
| US20200018005A1 (en) | Apparatus and method for treating laundry | |
| CN106400385A (en) | Washing machine and control method and device thereof |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190806 Address after: 214028 No. 18 Changjiang South Road, Wuxi National High-tech Development Zone, Jiangsu Province Applicant after: Wuxi Swan Electrical Appliances Co., Ltd. Address before: 214028 Wuxi Changjiang Road, New District, Jiangsu, No. 18 Applicant before: Wuxi Xiaotianer Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |