CN109447005A - Gesture recognition method and device, storage medium and electric appliance - Google Patents
Gesture recognition method and device, storage medium and electric appliance Download PDFInfo
- Publication number
- CN109447005A CN109447005A CN201811296517.4A CN201811296517A CN109447005A CN 109447005 A CN109447005 A CN 109447005A CN 201811296517 A CN201811296517 A CN 201811296517A CN 109447005 A CN109447005 A CN 109447005A
- Authority
- CN
- China
- Prior art keywords
- image
- gesture
- identified
- area
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000012545 processing Methods 0.000 claims abstract description 35
- 238000004590 computer program Methods 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 3
- 108010001267 Protein Subunits Proteins 0.000 claims description 2
- 230000005611 electricity Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 10
- 230000006870 function Effects 0.000 description 7
- 238000012549 training Methods 0.000 description 5
- 230000007797 corrosion Effects 0.000 description 4
- 238000005260 corrosion Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000004888 barrier function Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000003595 mist Substances 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/107—Static hand or arm
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a gesture recognition method, a gesture recognition device, a storage medium and an electric appliance, wherein the method comprises the following steps: acquiring a first image of a gesture to be recognized; judging whether the acquired first image meets a preset condition or not so as to determine whether the first image is fuzzy or not; if the first image is determined to be blurred, performing preset deblurring processing on the first image to obtain a processed second image; and performing gesture recognition on the second image to obtain corresponding gesture information of the gesture to be recognized. The scheme provided by the invention can be used for deblurring the image, so that the gesture area is accurately captured and the gesture information is recognized.
Description
Technical field
The present invention relates to field of image recognition more particularly to a kind of gesture identification method, device, storage medium and electric appliances.
Background technique
Currently, Gesture Recognition is largely used in smart home, hand can be used when speech comparison is noisy
Gesture removes control household electrical appliances.However, when light than it is darker when, have smog or user's electrical ionizer farther out when, pass through gesture identification
Technology identifies gesture, and the probability is relatively small, the experience sense of user can be reduced in this way, it is, therefore, desirable to provide a kind of raising smart home
The scheme of gesture identification probability in control.
Summary of the invention
It is a primary object of the present invention to overcome the defect of the above-mentioned prior art, provide a kind of gesture identification method, device,
Storage medium and electric appliance, with solve in the prior art when light than it is darker, have smog or user's electrical ionizer farther out when, pass through
The smaller problem of the identification probability of the gesture of Gesture Recognition identification control household electrical appliances.
One aspect of the present invention provides a kind of gesture identification method, comprising: acquires the first image of gesture to be identified;Judgement
Whether the first image of acquisition meets preset condition, to determine whether the first image obscures;If it is determined that described first
Image is fuzzy, then carries out default deblurring processing to the first image, second image that obtains that treated;To second figure
As carrying out gesture identification, to obtain the corresponding gesture information of the gesture to be identified.
Optionally, the preset condition, comprising: transmissivity is less than default transmissivity threshold value in the first image;With/
Or, the value of R, G of the first image, channel B is less than corresponding preset value;And/or the gesture area in the first image
Far from image capture device.
Optionally, preset deblurring processing is carried out to the first image, comprising: if the transmissivity of the first image
Less than default transmissivity threshold value, then by dark channel prior algorithm and/or the image defogging algorithm based on deep learning to described
First image carries out defogging processing;If the value of R, G of the first image, channel B is less than corresponding preset value, to described the
R, G of one image, the value of channel B are adjusted, to enhance the light of the first image;If the gesture in the first image
Region then carries out corrosion and micronization processes to the first image far from image capture device.
Optionally, further includes: if it is determined that the first image does not obscure, then gesture identification is carried out to the first image,
Obtain the corresponding gesture information of the gesture to be identified.
Optionally, gesture identification is carried out to the first image and/or gesture identification is carried out to second image, with
To the corresponding gesture information of the gesture to be identified, comprising: extracted in the first image and/or the second image to be identified
The third image of the gesture area of gesture;Gesture knowledge is carried out to the third image based on gesture identification model trained in advance
Not, to obtain the corresponding gesture information of the gesture to be identified.
Optionally, further includes: whether the area of the gesture area judged is greater than given threshold;In the gesture
In the case that the area in region is greater than the given threshold, based on gesture identification model trained in advance to the third image into
Row gesture identification.
Optionally, the gesture to be identified, comprising: for controlling the gesture of electric appliance;The method, further includes: will identify
The obtained corresponding gesture information is sent to corresponding electric appliance, to control the operation of the corresponding electric appliance.
Another aspect of the present invention provides a kind of gesture identifying device, comprising: acquisition unit, for acquiring gesture to be identified
The first image;Judging unit, for judging whether the first image of acquisition meets preset condition, with determination described first
Whether image obscures;Processing unit is used for if it is determined that the first image is fuzzy, then carries out default removing mould to the first image
Paste processing, second image that obtains that treated;Recognition unit, for carrying out gesture identification to second image, to obtain
State the corresponding gesture information of gesture to be identified.
Optionally, the preset condition, comprising: transmissivity is less than default transmissivity threshold value in the first image;With/
Or, the value of R, G of the first image, channel B is less than corresponding preset value;And/or the gesture area in the first image
Far from image capture device.
Optionally, the processing unit carries out preset deblurring processing to the first image, comprising: if described the
The transmissivity of one image is less than default transmissivity threshold value, then passes through dark channel prior algorithm and/or the image based on deep learning
Defogging algorithm carries out defogging processing to the first image;If the value of R, G of the first image, channel B is less than corresponding pre-
If value, then be adjusted the value of R, G of the first image, channel B, to enhance the light of the first image;If described
Gesture area in first image then carries out corrosion and micronization processes to the first image far from image capture device.
Optionally, the recognition unit, is also used to: if it is determined that the first image does not obscure, then to the first image
Gesture identification is carried out, the corresponding gesture information of the gesture to be identified is obtained.
Optionally, the recognition unit, comprising: subelement is extracted, in the first image and/or the second image
Extract the third image of the gesture area of gesture to be identified;Subelement is identified, for based on gesture identification mould trained in advance
Type carries out gesture identification to the third image, to obtain the corresponding gesture information of the gesture to be identified.
Optionally, the recognition unit, further includes: judgment sub-unit, the face of the gesture area for judging to obtain
Whether product is greater than given threshold;The identification subelement judges that the area of the gesture area is greater than described set in judging unit
In the case where determining threshold value, gesture identification is carried out to the third image based on gesture identification model trained in advance.
Optionally, the gesture to be identified, comprising: for controlling the gesture of electric appliance;Described device, further includes: send single
Member, the corresponding gesture information for obtaining identification is sent to corresponding electric appliance, to control the operation of the corresponding electric appliance.
Another aspect of the invention provides a kind of storage medium, is stored thereon with computer program, and described program is processed
The step of device realizes aforementioned any the method when executing.
Further aspect of the present invention provides a kind of electric appliance, including processor, memory and storage on a memory can be
The step of computer program run on processor, the processor realizes aforementioned any the method when executing described program.
Further aspect of the present invention provides a kind of electric appliance, including aforementioned any gesture identifying device.
According to the technique and scheme of the present invention, when the image of gesture to be identified is fuzzy, deblurring processing, energy are carried out to image
Enough remove the light of the smog and enhancing image in image, to accurately capture gesture area and identify gesture information;In turn
There can be the case where smog, dark or gesture motion leave home electrical appliance farther out, gesture, and root can also be accurately identified
Control household electrical appliance are removed according to the gesture information of identification, improve the experience sense that user uses household electrical appliance, and by improving gesture
Identification model improves recognition efficiency.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes a part of the invention, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is the method schematic diagram of an embodiment of gesture identification method provided by the invention;
Fig. 2 be it is according to an embodiment of the present invention to second image carry out gesture identification to obtain the gesture to be identified
Corresponding gesture information the step of a kind of specific embodiment flow diagram;
Fig. 3 is trained gesture identification model according to an embodiment of the present invention and carries out gesture based on the gesture identification model
The flow chart of identification;
Fig. 4 is the method schematic diagram of another embodiment of gesture identification method provided by the invention;
Fig. 5 is the structural schematic diagram of an embodiment of gesture identifying device provided by the invention;
Fig. 6 is a kind of structural schematic diagram of specific embodiment of recognition unit according to an embodiment of the present invention;
Fig. 7 is the structural schematic diagram of another embodiment of gesture identifying device provided by the invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with the specific embodiment of the invention and
Technical solution of the present invention is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the present invention one
Section Example, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to
Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product
Or other step or units that equipment is intrinsic.
Fig. 1 is the method schematic diagram of an embodiment of gesture identification method provided by the invention.The gesture identification method
It can be used in electric appliance, the gesture for controlling electric appliance identified.
As shown in Figure 1, according to one embodiment of present invention, the gesture identification method includes at least step S110, step
Rapid S120, step S130 and step S140.
Step S110 acquires the first image of gesture to be identified.
The first image of gesture to be identified is acquired by image capture device.For example, camera is configured on household appliances,
The first image of gesture to be identified is acquired by camera.
Step S120, judges whether the first image of acquisition meets preset condition, to determine that the first image is
It is no fuzzy.
Specifically, it may include situation, gesture area dark that gesture area is in smog that image is fuzzy
Situation, gesture area far from image capture device in the case where at least one of.
The case where for being in smog, the preset condition be specifically as follows in the first image transmissivity be less than it is default
Transmissivity threshold value.That is, judge whether the transmissivity in the first image is less than default transmissivity threshold value, if so,
Show that gesture area is in smog, then can determine that the first image is fuzzy.
The case where for gesture area dark, the preset condition is specifically as follows R, G, B of the first image
The value in channel is less than corresponding preset value.The value that tri- channels R, G, B of the first image can be read respectively, judges each channel
Value whether be less than corresponding preset value, if the value in tri- channels R, G, B is respectively less than corresponding preset value, show gesture area
Dark, then can determine that the first image is fuzzy.
The case where for gesture area far from image capture device, can judge gesture by judging the size of gesture area
Whether whether region is far from acquisition equipment, i.e., far from household electrical appliance.If judging, gesture area, can be with far from image capture device
Determine that the first image is fuzzy.
Step S130, however, it is determined that the first image is fuzzy, then carries out default deblurring processing to the first image, obtain
To treated the second image.
If it is determined that the first image is fuzzy, then corresponding preset is carried out according to the vague category identifier of the first image and remove mould
Paste processing.The vague category identifier is that gesture area corresponding with the image ambiguity in abovementioned steps S120 is in smog, hand
Gesture zonal ray is darker, gesture area is far from image capture device.
Specifically, for being in smog it the case where, is gone by dark channel prior algorithm and/or the image based on deep learning
Mist algorithm carries out defogging processing to the first image.The case where for gesture area dark, to the first image
R, G, channel B value be adjusted, to enhance the light of the first image, that is to say, that R, G, B of the first image
The value in channel increases analog value respectively, to enhance the light intensity of the first image.For gesture area far from Image Acquisition
The case where equipment, can carry out corrosion and micronization processes, so that the gesture area is more clear to the first image.
Step S140 carries out gesture identification to second image, to obtain the corresponding gesture letter of the gesture to be identified
Breath.
Fig. 2 be it is according to an embodiment of the present invention to second image carry out gesture identification to obtain the gesture to be identified
Corresponding gesture information the step of a kind of specific embodiment flow diagram.
As described in Figure 2, in a specific embodiment, step S140 includes step S141 and step S142.
Step S141 extracts the third image of the gesture area of gesture to be identified in second image.
Step S142 carries out gesture identification to the third image based on gesture identification model trained in advance, to obtain
The corresponding gesture information of the gesture to be identified.
Specifically, the of the gesture area of gesture to be identified can be extracted in second image by YOLO algorithm
Three images, that is, the image of the area-of-interest (that is, ROI) of original image.After obtaining the third image of gesture area, then pass through
Trained gesture identification model identifies the corresponding gesture information of gesture to be identified in advance.
Fig. 3 shows the flow chart of trained gesture identification model according to an embodiment of the present invention.
As shown in figure 3, training gesture identification model can specifically include: being specifically by training sample by image preprocessing
(image preprocessing process is identical as the deblurring processing in step S130) and image segmentation, identify corresponding gesture area, obtain
Obtain the gesture area (interesting image regions) in each frame image.It identifies gesture area relevant information, then utilizes improvement
FasterR_CNN model is trained, which is specifically to obtain a large amount of training sample data to be input to model, by several layers
(convolution mainly carries out sliding with input picture with each layer of weight W of model and is multiplied, and then biasing sets letter again for image convolution processing
Number) and pond layer (pond layer main purpose is to carry out dimensionality reduction to image) acquisition gesture area (interesting image regions ROI) spy
Reference breath finally passes through subsequently into full articulamentum (full articulamentum main purpose is that ROI feature information is generated feature vector)
Softmax layers obtain gesture probability, and further carry out calculating loss function with label, calculate loss function and reversely pass
Optimized model weight and bias function are broadcast, generates gesture identification model by the above process using mass data;It recognizes corresponding
Information be sent to electric appliance, for control household electrical appliance when, using the first image of the gesture to be identified of acquisition as test specimens
This, obtains the classification results of gesture to be identified by process same as training sample and is transferred to household electrical appliance.
Optionally, when the gesture area of gesture to be identified has barrier, the area of the gesture area judged is
It is no to be greater than given threshold;In the case where the area of the gesture area is greater than the given threshold, based on hand trained in advance
Gesture identification model carries out gesture identification to the third image.
Further, however, it is determined that the first image does not obscure, then carries out gesture identification to the first image, obtain institute
State the corresponding gesture information of gesture to be identified.
Wherein, gesture identification is carried out to the first image, obtains the tool of the corresponding gesture information of the gesture to be identified
Body embodiment can carry out gesture identification to second image with reference to aforementioned to obtain the corresponding hand of the gesture to be identified
It the step of gesture information, is not added and repeats herein.
Fig. 4 is the method schematic diagram of another embodiment of gesture identification method provided by the invention.The gesture identification side
Method can be used in electric appliance, identify to the gesture for controlling electric appliance.
As shown in figure 4, can also include step S150 when the gesture identification method is used for electric appliance.
The corresponding gesture information that identification obtains is sent to corresponding electric appliance, to control the corresponding electricity by step S150
The operation of device.
Wherein it is possible to which corresponding gesture information is sent to the corresponding electric appliance by wired or wireless way.Institute
State wired mode such as network interface, serial ports etc..Described wireless mode such as WiFi, bluetooth etc..
Fig. 5 is the structural schematic diagram of an embodiment of gesture identifying device provided by the invention.The gesture identification method
It can be used in electric appliance, the gesture for controlling electric appliance identified.
As shown in figure 5, the gesture identifying device 100 includes: acquisition unit 110, judging unit 120, processing unit 130
With recognition unit 140.
Acquisition unit 110 is used to acquire the first image of gesture to be identified;Judging unit 120 is used to judge the described of acquisition
Whether the first image meets preset condition, to determine whether the first image obscures;Processing unit 130 is used for if it is determined that described
First image is fuzzy, then carries out default deblurring processing to the first image, second image that obtains that treated;Recognition unit
140 for carrying out gesture identification to second image, to obtain the corresponding gesture information of the gesture to be identified.
Acquisition unit 110 acquires the first image of gesture to be identified.Specifically, can by image capture device acquire to
Identify the first image of gesture.For example, configuring camera on household appliances, the first of gesture to be identified is acquired by camera
Image.
Judging unit 120 judges whether the first image of acquisition meets preset condition, to determine the first image
Whether obscure.Specifically, it may include situation, gesture area dark that gesture area is in smog that image is fuzzy
Situation, gesture area far from image capture device in the case where at least one of.
The case where for being in smog, the preset condition be specifically as follows in the first image transmissivity be less than it is default
Transmissivity threshold value.That is, judge whether the transmissivity in the first image is less than default transmissivity threshold value, if so,
Show that gesture area is in smog, then can determine that the first image is fuzzy.
The case where for gesture area dark, the preset condition is specifically as follows R, G, B of the first image
The value in channel is less than corresponding preset value.The value that tri- channels R, G, B of the first image can be read respectively, judges each channel
Value whether be less than corresponding preset value, if the value in tri- channels R, G, B is respectively less than corresponding preset value, show gesture area
Dark, then can determine that the first image is fuzzy.
The case where for gesture area far from image capture device, can judge gesture by judging the size of gesture area
Whether whether region is far from acquisition equipment, i.e., far from household electrical appliance.If judging, gesture area, can be with far from image capture device
Determine that the first image is fuzzy.
If it is determined that the first image is fuzzy, then processing unit 130 carries out default deblurring processing to the first image,
Second image that obtains that treated.Specifically, however, it is determined that the first image is fuzzy, then processing unit 130 is according to described first
The vague category identifier of image carries out default deblurring processing accordingly.The vague category identifier, that is, with image ambiguity pair above-mentioned
The gesture area answered is in smog, gesture area dark, gesture area far from image capture device.
Specifically, for being in smog it the case where, is gone by dark channel prior algorithm and/or the image based on deep learning
Mist algorithm carries out defogging processing to the first image.The case where for gesture area dark, to the first image
R, G, channel B value be adjusted, to enhance the light of the first image, that is to say, that R, G, B of the first image
The value in channel increases analog value respectively, to enhance the light intensity of the first image.For gesture area far from Image Acquisition
The case where equipment, can carry out corrosion and micronization processes, so that the gesture area is more clear to the first image.
Gesture identification is carried out to second image, to obtain the corresponding gesture information of the gesture to be identified.
Fig. 6 is a kind of structural schematic diagram of specific embodiment of recognition unit according to an embodiment of the present invention.
As described in Figure 6, in a specific embodiment, recognition unit 140 includes extracting subelement 141 and identifying that son is single
Member 142.
Extract the third figure that subelement 141 is used to extract the gesture area of gesture to be identified in second image
Picture.Identify that subelement 142 is used to carry out gesture identification to the third image based on gesture identification model trained in advance, with
To the corresponding gesture information of the gesture to be identified.
The concrete mode of training gesture identification model can refer to the process of trained gesture identification model shown in Fig. 3 in advance
Corresponding description in figure and preceding method embodiment.
Specifically, the of the gesture area of gesture to be identified can be extracted in second image by YOLO algorithm
Three images.Obtain the third image of gesture area, then by advance trained gesture identification model identify it is to be identified
The corresponding gesture information of gesture.
Optionally, the recognition unit 140 further includes judgment sub-unit (not shown), for judging the obtained gesture
Whether the area in region is greater than given threshold;The identification subelement 142 judges the face of the gesture area in judgment sub-unit
In the case that product is greater than the given threshold, gesture knowledge is carried out to the third image based on gesture identification model trained in advance
Not.
Specifically, when the gesture area of gesture to be identified has barrier, the area of the gesture area judged is
It is no to be greater than given threshold;In the case where the area of the gesture area is greater than the given threshold, based on hand trained in advance
Gesture identification model carries out gesture identification to the third image.
Optionally, the recognition unit 140 is also used to: if the judging unit 120 determines that the first image does not obscure,
Gesture identification then is carried out to the first image, obtains the corresponding gesture information of the gesture to be identified.Wherein, the identification is single
140 pairs of the first images of member carry out gesture identification, obtain the specific embodiment party of the corresponding gesture information of the gesture to be identified
Formula can carry out gesture identification to second image with reference to aforementioned to obtain the corresponding gesture information of the gesture to be identified
Step is not added herein and repeats.
Fig. 7 is the structural schematic diagram of another embodiment of gesture identifying device provided by the invention.The gesture to be identified,
It include: the gesture for controlling electric appliance.As shown in fig. 7, described device 100 further include: transmission unit 150.
Transmission unit 150 is used to the corresponding gesture information that identification obtains being sent to corresponding electric appliance, described in control
The operation of corresponding electric appliance.
Wherein it is possible to which corresponding gesture information is sent to the corresponding electric appliance by wired or wireless way.Institute
State wired mode such as network interface, serial ports etc..Described wireless mode such as WiFi, bluetooth etc..
The present invention also provides a kind of storage mediums for corresponding to the gesture identification method, are stored thereon with computer journey
Sequence, the step of aforementioned any the method is realized when described program is executed by processor.
The present invention also provides a kind of electric appliances for corresponding to the gesture identification method, including processor, memory and deposit
The computer program that can be run on a processor on a memory is stored up, the processor is realized aforementioned any when executing described program
The step of the method.
The present invention also provides a kind of electric appliances for corresponding to the gesture identifying device, know including aforementioned any gesture
Other device.
Accordingly, scheme provided by the invention carries out deblurring processing to image when the image of gesture to be identified is fuzzy,
The light of the smog and enhancing image in image can be removed, to accurately capture gesture area and identify gesture information;Into
And can have the case where smog, dark or gesture motion leave home electrical appliance farther out, gesture can also be accurately identified,
And control household electrical appliance are gone according to the gesture information of identification, improve the experience sense that user uses household electrical appliance.
Function described herein can be implemented in hardware, the software executed by processor, firmware or any combination thereof.
If implemented in the software executed by processor, computer can be stored in using function as one or more instructions or codes
It is transmitted on readable media or via computer-readable media.Other examples and embodiment are wanted in the present invention and appended right
It asks in the scope and spirit of book.For example, due to the property of software, function described above can be used by processor,
Hardware, firmware, hardwired or the software implementation for appointing the combination of whichever to execute in these.In addition, each functional unit can integrate
In one processing unit, it is also possible to each unit to physically exist alone, can also be integrated in two or more units
In one unit.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, Ke Yiwei
A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or module
It connects, can be electrical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, and fill as control
The component set may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
On unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code
Medium.
The above description is only an embodiment of the present invention, is not intended to restrict the invention, for those skilled in the art
For member, the invention may be variously modified and varied.All within the spirits and principles of the present invention, it is made it is any modification,
Equivalent replacement, improvement etc., should be included within scope of the presently claimed invention.
Claims (17)
1. a kind of gesture identification method characterized by comprising
Acquire the first image of gesture to be identified;
Judge whether the first image of acquisition meets preset condition, to determine whether the first image obscures;
If it is determined that the first image is fuzzy, then default deblurring processing is carried out to the first image, obtains that treated the
Two images;
Gesture identification is carried out to second image, to obtain the corresponding gesture information of the gesture to be identified.
2. the method according to claim 1, wherein the preset condition, comprising:
Transmissivity is less than default transmissivity threshold value in the first image;And/or
R, G of the first image, the value of channel B are less than corresponding preset value;And/or
Gesture area in the first image is far from image capture device.
3. according to the method described in claim 2, it is characterized in that, carry out preset deblurring processing to the first image,
Include:
If the transmissivity of the first image is less than default transmissivity threshold value, by dark channel prior algorithm and/or based on deep
The image defogging algorithm for spending study carries out defogging processing to the first image;
If the value of R, G of the first image, channel B is less than corresponding preset value, R, G, channel B to the first image
Value be adjusted, to enhance the light of the first image;
If the gesture area in the first image corrodes the first image and is refined far from image capture device
Processing.
4. method according to claim 1-3, which is characterized in that further include:
If it is determined that the first image does not obscure, then gesture identification is carried out to the first image, obtain the gesture to be identified
Corresponding gesture information.
5. method according to claim 1-4, which is characterized in that carry out gesture identification to the first image
And/or gesture identification is carried out to second image, to obtain the corresponding gesture information of the gesture to be identified, comprising:
The third image of the gesture area of gesture to be identified is extracted in the first image and/or the second image;
Gesture identification is carried out to the third image based on gesture identification model trained in advance, to obtain the gesture to be identified
Corresponding gesture information.
6. according to the method described in claim 5, it is characterized by further comprising:
Judge whether the area of the obtained gesture area is greater than given threshold;
In the case where the area of the gesture area is greater than the given threshold, based on gesture identification model pair trained in advance
The third image carries out gesture identification.
7. method according to claim 1-6, which is characterized in that the gesture to be identified, comprising: for controlling
The gesture of electric appliance;The method, further includes:
The corresponding gesture information that identification obtains is sent to corresponding electric appliance, to control the operation of the corresponding electric appliance.
8. a kind of gesture identifying device characterized by comprising
Acquisition unit, for acquiring the first image of gesture to be identified;
Judging unit, for judging whether the first image of acquisition meets preset condition, to determine that the first image is
It is no fuzzy;
Processing unit is used to then carry out default deblurring processing to the first image, obtain if it is determined that the first image is fuzzy
To treated the second image;
Recognition unit, for carrying out gesture identification to second image, to obtain the corresponding gesture letter of the gesture to be identified
Breath.
9. device according to claim 8, which is characterized in that the preset condition, comprising:
Transmissivity is less than default transmissivity threshold value in the first image;And/or
R, G of the first image, the value of channel B are less than corresponding preset value;And/or
Gesture area in the first image is far from image capture device.
10. device according to claim 9, which is characterized in that the processing unit presets the first image
Deblurring processing, comprising:
If the transmissivity of the first image is less than default transmissivity threshold value, by dark channel prior algorithm and/or based on deep
The image defogging algorithm for spending study carries out defogging processing to the first image;
If the value of R, G of the first image, channel B is less than corresponding preset value, R, G, channel B to the first image
Value be adjusted, to enhance the light of the first image;
If the gesture area in the first image corrodes the first image and is refined far from image capture device
Processing.
11. according to the described in any item devices of claim 8-10, which is characterized in that the recognition unit is also used to:
If it is determined that the first image does not obscure, then gesture identification is carried out to the first image, obtain the gesture to be identified
Corresponding gesture information.
12. according to the described in any item devices of claim 8-11, which is characterized in that the recognition unit, comprising:
Subelement is extracted, for extracting in the first image and/or the second image the gesture area of gesture to be identified
Third image;
Identify subelement, for carrying out gesture identification to the third image based on gesture identification model trained in advance, with
To the corresponding gesture information of the gesture to be identified.
13. device according to claim 12, which is characterized in that the recognition unit, further includes:
Judgment sub-unit, for judging whether the area of the obtained gesture area is greater than given threshold;
The identification subelement, in the case where judging unit judges the area of the gesture area greater than the given threshold,
Gesture identification is carried out to the third image based on gesture identification model trained in advance.
14. according to the described in any item devices of claim 8-13, which is characterized in that the gesture to be identified, comprising: for controlling
The gesture of electric appliance processed;Described device, further includes:
Transmission unit, the corresponding gesture information for obtaining identification is sent to corresponding electric appliance, to control the corresponding electricity
The operation of device.
15. a kind of storage medium, which is characterized in that it is stored thereon with computer program, it is real when described program is executed by processor
The step of existing claim 1-7 any the method.
16. a kind of electric appliance, which is characterized in that can be transported on a processor on a memory including processor, memory and storage
The step of capable computer program, the processor realizes claim 1-7 any the method when executing described program.
17. a kind of electric appliance, which is characterized in that including the gesture identifying device as described in claim 8-14 is any.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811296517.4A CN109447005A (en) | 2018-11-01 | 2018-11-01 | Gesture recognition method and device, storage medium and electric appliance |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811296517.4A CN109447005A (en) | 2018-11-01 | 2018-11-01 | Gesture recognition method and device, storage medium and electric appliance |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109447005A true CN109447005A (en) | 2019-03-08 |
Family
ID=65550532
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811296517.4A Pending CN109447005A (en) | 2018-11-01 | 2018-11-01 | Gesture recognition method and device, storage medium and electric appliance |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109447005A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110955243A (en) * | 2019-11-28 | 2020-04-03 | 新石器慧通(北京)科技有限公司 | Travel control method, travel control device, travel control apparatus, readable storage medium, and mobile device |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102385439A (en) * | 2011-10-21 | 2012-03-21 | 华中师范大学 | Man-machine gesture interactive system based on electronic whiteboard |
CN105141939A (en) * | 2015-08-18 | 2015-12-09 | 宁波盈芯信息科技有限公司 | Three-dimensional depth perception method and three-dimensional depth perception device based on adjustable working range |
CN106293387A (en) * | 2016-07-27 | 2017-01-04 | 上海与德通讯技术有限公司 | The control method of application program and system |
CN106648423A (en) * | 2016-11-24 | 2017-05-10 | 深圳奥比中光科技有限公司 | Mobile terminal and interactive control method thereof |
CN106851937A (en) * | 2017-01-25 | 2017-06-13 | 触景无限科技(北京)有限公司 | A kind of method and device of gesture control desk lamp |
CN108021880A (en) * | 2017-11-30 | 2018-05-11 | 宁波高新区锦众信息科技有限公司 | A kind of intelligent home control system based on gesture identification |
US20180260735A1 (en) * | 2017-03-08 | 2018-09-13 | International Business Machines Corporation | Training a hidden markov model |
-
2018
- 2018-11-01 CN CN201811296517.4A patent/CN109447005A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102385439A (en) * | 2011-10-21 | 2012-03-21 | 华中师范大学 | Man-machine gesture interactive system based on electronic whiteboard |
CN105141939A (en) * | 2015-08-18 | 2015-12-09 | 宁波盈芯信息科技有限公司 | Three-dimensional depth perception method and three-dimensional depth perception device based on adjustable working range |
CN106293387A (en) * | 2016-07-27 | 2017-01-04 | 上海与德通讯技术有限公司 | The control method of application program and system |
CN106648423A (en) * | 2016-11-24 | 2017-05-10 | 深圳奥比中光科技有限公司 | Mobile terminal and interactive control method thereof |
CN106851937A (en) * | 2017-01-25 | 2017-06-13 | 触景无限科技(北京)有限公司 | A kind of method and device of gesture control desk lamp |
US20180260735A1 (en) * | 2017-03-08 | 2018-09-13 | International Business Machines Corporation | Training a hidden markov model |
CN108021880A (en) * | 2017-11-30 | 2018-05-11 | 宁波高新区锦众信息科技有限公司 | A kind of intelligent home control system based on gesture identification |
Non-Patent Citations (2)
Title |
---|
北方工业大学教务处: "《青春向前进 大学生科学研究与创业行动计划研究报告论文集2014》", 31 October 2015 * |
张秀彬等: "《发明解析论》", 30 June 2014, 上海交通大学出版社 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110955243A (en) * | 2019-11-28 | 2020-04-03 | 新石器慧通(北京)科技有限公司 | Travel control method, travel control device, travel control apparatus, readable storage medium, and mobile device |
CN110955243B (en) * | 2019-11-28 | 2023-10-20 | 新石器慧通(北京)科技有限公司 | Travel control method, apparatus, device, readable storage medium, and mobile apparatus |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108229277B (en) | Gesture recognition method, gesture control method, multilayer neural network training method, device and electronic equipment | |
CN110738101B (en) | Behavior recognition method, behavior recognition device and computer-readable storage medium | |
CN107067006B (en) | Verification code identification method and system serving for data acquisition | |
CN105095882B (en) | Gesture recognition method and device | |
US20190122072A1 (en) | Reverse neural network for object re-identification | |
CN107832684B (en) | Intelligent vein authentication method and system with autonomous learning capability | |
CN110070029B (en) | Gait recognition method and device | |
CN109359666A (en) | A kind of model recognizing method and processing terminal based on multiple features fusion neural network | |
CN109002755B (en) | Age estimation model construction method and estimation method based on face image | |
CN103903006A (en) | Crop pest identification method and system based on Android platform | |
CN111160186B (en) | Intelligent garbage classification processing method and related products | |
CN107067022B (en) | Method, device and equipment for establishing image classification model | |
CN109886070A (en) | Equipment control method and device, storage medium and equipment | |
CN108107743A (en) | Distribution method and device of control authority, storage medium and processor | |
CN109343701A (en) | A kind of intelligent human-machine interaction method based on dynamic hand gesture recognition | |
CN117392733B (en) | Acne grading detection method and device, electronic equipment and storage medium | |
CN116740384B (en) | Intelligent control method and system of floor washing machine | |
CN104021384A (en) | Face recognition method and device | |
CN111027534A (en) | Compact double-license-plate detection method and device | |
CN116385430A (en) | Machine vision flaw detection method, device, medium and equipment | |
CN111178221A (en) | Identity recognition method and device | |
CN109447005A (en) | Gesture recognition method and device, storage medium and electric appliance | |
CN111091122A (en) | Training and detecting method and device for multi-scale feature convolutional neural network | |
CN115116111B (en) | Anti-disturbance human face living body detection model training method and device and electronic equipment | |
CN106707789A (en) | Smart home control system based on fingerprint recognition |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190308 |
|
RJ01 | Rejection of invention patent application after publication |