CN111695475B - NMI-based intelligent household appliance control method - Google Patents
NMI-based intelligent household appliance control method Download PDFInfo
- Publication number
- CN111695475B CN111695475B CN202010497497.8A CN202010497497A CN111695475B CN 111695475 B CN111695475 B CN 111695475B CN 202010497497 A CN202010497497 A CN 202010497497A CN 111695475 B CN111695475 B CN 111695475B
- Authority
- CN
- China
- Prior art keywords
- gesture
- image
- nmi
- gestures
- binarized
- 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
Images
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
-
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/2803—Home automation networks
- H04L12/2816—Controlling appliance services of a home automation network by calling their functionalities
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Automation & Control Theory (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
A method for intelligently controlling a home appliance based on NMI, comprising: firstly, constructing a database containing NMI characteristic values of various gestures, and recording the NMI characteristic values of gesture images for controlling the on/off of a household appliance and adjusting the temperature/brightness of an air conditioner/the volume of a television into the database; step two, a gesture acquisition system is built, and a camera is used for acquiring gestures made by a user; step three, a gesture recognition system is established, the NMI is utilized to conduct feature extraction on the preprocessed gesture image, and gesture recognition on a user is completed by comparing NMI feature values of the gesture image to be recognized with NMI feature values of various gestures in a database, so that gesture information is obtained; and step four, inputting the obtained gesture information into a control system, inputting control commands corresponding to various gesture information into the control system in advance, and then applying the control commands to the household appliances to make corresponding operation changes. The gesture can be used for controlling various household appliances, and is simple and easy to memorize.
Description
Technical Field
The invention relates to a method for intelligently controlling household appliances.
Technical Field
Under the development of computer technology, gesture recognition technology is becoming more popular as the next emerging man-machine interaction technology because of its non-contact control mode. People can directly use the hand to act, so as to realize the purposes of controlling the on/off of the household appliances, adjusting the temperature of the air conditioner, the brightness of the electric lamp, the volume of the television and the like. However, at present, different household appliances need to be controlled by different controllers, so that the problems of bulkiness of a household equipment control system, excessive occupation of household space by the controllers and the like are caused. In addition, the traditional household equipment with the gesture recognition function needs to collect and input various gestures made by each user in advance, and the workload of collecting and inputting the early-stage data is large.
Disclosure of Invention
Aiming at the problems, the invention provides a method for intelligently controlling household appliances based on NMI.
According to the method, translation transformation, rotation transformation, brightness transformation and proportional transformation invariance of NMI features are utilized, a database containing NMI feature values of various gestures is firstly constructed, then a gesture acquisition system is constructed to acquire and record gestures made by a user, gesture features of gesture images subjected to gray scale processing, comprehensive filtering processing and binarization processing are extracted by utilizing NMI, then the NMI feature values of the gesture images to be identified are compared with NMI feature values in the database, and then differences of the NMI feature values and the NMI feature values are compared with a judgment threshold value set by the system to obtain correct gesture information. And finally, inputting the obtained gesture information into a control system to obtain a control command corresponding to the gesture, and then applying the control command to the household appliance to enable the household appliance to make corresponding operation changes.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a method for intelligently controlling household appliances based on NMI, comprising the following steps:
step one, constructing a database containing NMI characteristic values of various gestures;
inputting NMI characteristic values of gesture images for controlling the on/off of household appliances and adjusting the temperature of an air conditioner, the brightness of an electric lamp and the volume of a television into a database;
because the difference between the left hand and the right hand of a normal person is very small, the NMI characteristic value difference of gesture images of the same gesture made by the left hand and the right hand is very small, namely when the gesture image recorded into a database is the right hand, and the household appliance is controlled by the left hand in practice, the control requirement can still be met, and error reporting can not occur; in addition, because NMI features have good translational conversion, rotational conversion, brightness conversion and scaling conversion invariance, gestures and standard gestures when the household appliances are controlled in practice are not influenced; meanwhile, if the standard gestures in the database are respectively composed of actions made by hands of children, teenagers and adults, when other people in the same age range make the same gestures on the household appliance, the household appliance can be accurately controlled, namely, only a certain gesture for controlling the household appliance by a person in a certain age range is recorded in the database, the correct control on the household appliance can be realized when other people in the same age range make corresponding gestures on the household appliance, and the processing work such as data acquisition and input in the early stage is reduced virtually;
step two, building a gesture acquisition system;
acquiring gestures made by a user by adopting a camera;
step three, establishing a gesture recognition system;
and (3) carrying out feature extraction on the preprocessed gesture image by utilizing NMI, and completing the recognition of gestures made by a user by comparing NMI feature values of the gesture image to be recognized with NMI feature values of various gestures in a database to obtain gesture information, wherein the implementation flow is as follows:
preprocessing a gesture image;
preprocessing a gesture image mainly comprises image graying, image comprehensive filtering processing and image binarization;
(1) Graying of the image;
carrying out graying treatment on the gesture color image by using a weighted average method; because the induction intensity of human eyes on the RGB three colors is different, the invention carries out weighted average treatment on the RGB three colors of the image, namely:
Gray(i,j)＝0.3R(i,j)+0.59G(i,j)+0.11B(i,j) (1)
(2) And (5) comprehensive filtering processing of the image.
The graying gesture image is affected by Gaussian noise and spiced salt noise to a certain extent, and the existence of the noise can increase the difficulty of subsequent research; processing the gray level map of the gesture image by adopting comprehensive filtering, namely removing Gaussian noise of the image by adopting average filtering firstly, and removing salt and pepper noise of the image by adopting median filtering secondly so as to improve the quality of the gray level map of the gesture image;
(3) Binarizing the image;
after the gray level image of the gesture image is subjected to comprehensive filtering treatment, the contrast ratio between the gesture lines of the image and the background area is increased, and the degree of distinction between the gesture lines and the background area is further improved; because the Niblack dynamic threshold segmentation algorithm has outstanding performance advantages in the aspects of image segmentation effect, running efficiency, actual operability and the like, the algorithm is utilized to carry out binarization processing on the gesture image after comprehensive filtering processing;
the Niblack dynamic threshold segmentation algorithm calculates the average value of the gray values of pixels in the neighborhood by using all elements in the neighborhood with the size of M multiplied by N:
standard deviation:
then, carrying out weight addition processing on the gray value mean value and the standard variance, and finally, taking the obtained gray value as a judgment threshold value to carry out binarization to obtain a binary feature map of the gesture image; the method comprises the following steps:
H(i,j)＝α×k(i,j)+β×s(i,j) (4)
wherein t (i, j) is a pixel gray value in an MxN neighborhood after the comprehensive filtering process, k (i, j) is a gray value average value, s (i, j) is a standard deviation, H (i, j) is a threshold value, and alpha and beta are correction weights;
(II) extracting gesture features based on NMI;
calculating NMI characteristic values of the binary image around the center of gravity of the image by using the sum of gray values of pixel points of the binary image, the center of gravity of the binary image and the moment of inertia of the binary image around the center of gravity of the image, wherein the NMI characteristic values of the binary image around the center of gravity of the image are specifically realized by the following steps:
(1) Regarding a binarized gesture image M multiplied by N as a plane slice, regarding pixels of the binarized image as particles on an XOY plane, and regarding gray values f (i, j) of the pixels of the binarized image as mass of the particles, wherein i and j respectively represent rows and columns of the pixels in a pixel matrix;
(2) Calculating the quality m (f (i, j)) of the binarized gesture image:
(3) Calculating center of gravity of binarized gesture image
(4) Calculating the center of gravity of a binarized gesture image around the image center
Moment of inertia->
(5) Calculating the center of gravity of a binarized gesture image around the image center
To be abbreviated as Normalized Moment of Inertia (NMI):
thirdly, carrying out gesture recognition;
on the basis of obtaining NMI characteristic value X of gesture image to be identified, it is matched with NMI characteristic value Y containing various gestures in database i One-to-one comparison is carried out to obtain a difference A between the two i Then the difference isComparing and analyzing the value with a judgment threshold T set by a system; if A i If the difference is less than or equal to T, the gesture to be recognized is recognized to correspond to the gesture which generates the NMI characteristic value difference in the database, otherwise, the gesture does not correspond to the gesture;
and step four, inputting the obtained gesture information into a control system, inputting control commands corresponding to various gesture information into the control system in advance, and then applying the control commands to the household appliances to make corresponding operation changes.
Preferably, the gesture image for controlling on/off of the household appliance and adjusting the temperature of the air conditioner/brightness of the electric lamp/volume of the television set in the step one includes: the control of the household appliance is realized by utilizing the hand action, after the five fingers are fully extended to realize the starting operation of the household appliance, any finger is extended to four fingers to realize the upward adjustment of the temperature/brightness/volume and the like of the appliance, and the upper limit is 5, namely the appliance is in 1 after being started; on the basis of adjusting the temperature/brightness/volume of the electric appliance upwards, the lower limit of the electric appliance is 1 grade, and the temperature/brightness/volume of the electric appliance can be adjusted downwards when the number of fingers is reduced to control the household electric appliance.
Preferably, in the gesture collection system in the second step, in order to avoid gesture ambiguity caused by gesture self-shielding under the condition of single viewpoint, a binocular camera with a camera angle of 90 ° is adopted to collect gestures made by a user.
The invention has the advantages that:
the invention provides a method for intelligently controlling household appliances by gestures based on NMI, which is used for intelligently controlling household appliances used in daily life of people. The outstanding characteristics are as follows: firstly, one gesture can control various household appliances, the control gesture is simple and easy to memorize, the problems that the control systems of different household appliances are provided with different controllers at present, and the like are large, the controllers excessively occupy the household space, and the like can be solved, people are liberated from the controllers of the household appliances, the control freedom is realized, and meanwhile, the life perception capability of people with handicapped actions, vision and hearing can be improved; secondly, because NMI features have very good translation transformation, rotation transformation, brightness transformation and proportion transformation invariance, even if a gesture of a person is not recorded into a database in advance, the gesture is prevented from deviating from a standard gesture when the user controls the household appliance, and when the user makes a corresponding control gesture to the household appliance, the user can still accurately control the household appliance, so that the problem that the traditional household equipment with the gesture recognition function needs to collect and record various gestures made by the user in advance for each user can be solved, and the workload of collecting and recording the data in the earlier stage is reduced.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic view of the binocular camera of the present invention at 90;
FIG. 3 is a technical roadmap of the gesture recognition system of the invention;
fig. 4 is a flowchart of the gesture image synthesis filtering process of the present invention.
Detailed Description
In order to verify the feasibility and superiority of the method provided by the invention, the invention is further described with reference to application scenes:
a method for intelligently controlling household appliances based on NMI, comprising the following steps:
step one, constructing a database containing NMI characteristic values of various gestures;
NMI characteristic values of gesture images for controlling the on/off of the household appliance (five fingers are fully extended/five fingers are folded towards the palm), adjusting the temperature of an air conditioner, the brightness of an electric lamp, the volume of a television and the like (any finger is extended to four fingers) are recorded into a database.
The invention realizes the control of the household appliance by utilizing the hand action, when the five fingers are fully extended to realize the starting operation of the household appliance, any finger is extended to four fingers immediately, so that the temperature/brightness/volume and the like of the appliance can be adjusted upwards, and the upper limit is 5, namely the appliance is in 1 grade after being started. On the basis of adjusting the temperature/brightness/volume of the electric appliance upwards, the lower limit of the electric appliance is 1 grade, and the temperature/brightness/volume of the electric appliance can be adjusted downwards when the number of fingers is reduced to control the household electric appliance.
Because the difference between the left hand and the right hand of a normal person is very small, the NMI characteristic value difference of gesture images of the same gesture made by the left hand and the right hand is very small, namely when the gesture image recorded into a database is the right hand, and the household appliance is controlled by the left hand in practice, the control requirement can still be met, and error reporting can not occur; in addition, because NMI features have good translational conversion, rotational conversion, brightness conversion and scaling conversion invariance, gestures and standard gestures when the household appliances are controlled in practice are not influenced; meanwhile, if the standard gestures in the database are respectively composed of actions made by hands of children, teenagers and adults, when other people in the same age range make the same gestures on the household appliance, the household appliance can be accurately controlled, namely, only a certain gesture for controlling the household appliance by a person in a certain age range is recorded in the database, the correct control on the household appliance can be realized when other people in the same age range make corresponding gestures on the household appliance, and the processing work such as data acquisition and input in the early stage is reduced virtually.
Step two, building a gesture acquisition system;
in order to avoid gesture ambiguity caused by gesture self-shielding under the condition of single viewpoint, the invention adopts a binocular camera with a camera angle of 90 degrees to collect gestures made by a user.
Step three, establishing a gesture recognition system;
according to the gesture recognition method, the NMI is utilized to conduct feature extraction on the preprocessed gesture image, recognition of gestures made by a user is completed by comparing NMI feature values of the gesture image to be recognized with NMI feature values of various gestures in a database, gesture information is obtained, and the implementation flow is as follows:
preprocessing a gesture image;
the preprocessing of the gesture image mainly comprises image graying, image comprehensive filtering processing and image binarization.
(1) Graying of the image;
the invention uses weighted average method to gray the hand color image. Because the induction intensity of human eyes on the RGB three colors is different, the invention carries out weighted average treatment on the RGB three colors of the image, namely:
Gray(i,j)＝0.3R(i,j)+0.59G(i,j)+0.11B(i,j) (1)
(2) Comprehensive filtering treatment of the image;
the grayed gesture image is affected by Gaussian noise and spiced salt noise to a certain extent, and the existence of the noise can increase the difficulty of subsequent research. According to the method, the gesture image gray level map is processed by adopting comprehensive filtering, namely Gaussian noise of the image is removed by adopting average filtering, and then spiced salt noise of the image is removed by adopting median filtering, so that the quality of the gesture image gray level map is improved.
(3) Binarizing the image;
after the gray level map of the gesture image is subjected to comprehensive filtering treatment, the contrast ratio between the gesture lines of the image and the background area is increased, and the degree of distinction between the gesture lines and the background area is further improved. Because the Niblack dynamic threshold segmentation algorithm has outstanding performance advantages in the aspects of image segmentation effect, operation efficiency, actual operability and the like, the method is used for carrying out binarization processing on the gesture image after comprehensive filtering processing.
The Niblack dynamic threshold segmentation algorithm calculates the average value of the gray values of pixels in the neighborhood by using all elements in the neighborhood with the size of M multiplied by N:
standard deviation:
and then carrying out weight addition processing on the gray value mean value and the standard variance, and finally taking the obtained gray value as a judgment threshold value to carry out binarization to obtain a binary feature map of the gesture image. The method comprises the following steps:
H(i,j)＝α×k(i,j)+β×s(i,j) (4)
wherein t (i, j) is a pixel gray value in an m×n neighborhood after the synthesis filtering process, k (i, j) is a gray value average value, s (i, j) is a standard deviation, H (i, j) is a threshold value, and α and β are correction weights.
(II) extracting gesture features based on NMI;
the invention calculates NMI characteristic value of the binary image around the center of gravity of the image by using the sum of gray values of pixel points of the binary image, the center of gravity of the binary image and the moment of inertia of the binary image around the center of gravity of the image, and the specific implementation steps are as follows:
(1) The binarized gesture image M x N is regarded as a planar sheet, the pixels of the binarized image are regarded as particles on the XOY plane, and the gray values f (i, j) of the pixels of the binarized image are regarded as mass of the particles, wherein i and j represent the rows and columns of the pixels in the pixel matrix, respectively.
(2) Calculating the quality m (f (i, j)) of the binarized gesture image:
(3) Calculating center of gravity of binarized gesture image
(4) Calculating the center of gravity of a binarized gesture image around the image center
Moment of inertia->
(5) Calculating the center of gravity of a binarized gesture image around the image center
To be abbreviated as Normalized Moment of Inertia (NMI):
thirdly, carrying out gesture recognition;
on the basis of obtaining NMI characteristic value X of gesture image to be identified, it is matched with NMI characteristic value Y containing various gestures in database i One-to-one comparison is carried out to obtain a difference A between the two i The difference is then compared with a decision threshold T set by the system. If A i And if the difference is less than or equal to T, the gesture to be recognized corresponds to the gesture which generates the NMI characteristic value difference in the database, otherwise, the gesture does not correspond to the gesture.
And step four, inputting the obtained gesture information into a control system, inputting control commands corresponding to various gesture information into the control system in advance, and then applying the control commands to the household appliances to make corresponding operation changes.
The invention has the advantages that:
the invention provides a method for intelligently controlling household appliances by gestures based on NMI, which is used for intelligently controlling household appliances used in daily life of people. The outstanding characteristics are as follows: firstly, one gesture can control various household appliances, the control gesture is simple and easy to memorize, the problems that the control systems of different household appliances are provided with different controllers at present, and the like are large, the controllers excessively occupy the household space, and the like can be solved, people are liberated from the controllers of the household appliances, the control freedom is realized, and meanwhile, the life perception capability of people with handicapped actions, vision and hearing can be improved; secondly, because NMI features have very good translation transformation, rotation transformation, brightness transformation and proportion transformation invariance, even if a gesture of a person is not recorded into a database in advance, the gesture is prevented from deviating from a standard gesture when the user controls the household appliance, and when the user makes a corresponding control gesture to the household appliance, the user can still accurately control the household appliance, so that the problem that the traditional household equipment with the gesture recognition function needs to collect and record various gestures made by the user in advance for each user can be solved, and the workload of collecting and recording the data in the earlier stage is reduced.
The embodiments described in the present specification are merely examples of implementation forms of the inventive concept, and the scope of protection of the present invention should not be construed as being limited to the specific forms set forth in the embodiments, and the scope of protection of the present invention and equivalent technical means that can be conceived by those skilled in the art based on the inventive concept.
Claims (3)
1. A method for intelligently controlling household appliances based on NMI, comprising the following steps:
step one, constructing a database containing NMI characteristic values of various gestures;
inputting NMI characteristic values of gesture images for controlling the on/off of household appliances and adjusting the temperature of an air conditioner, the brightness of an electric lamp and the volume of a television into a database;
because the difference between the left hand and the right hand of a normal person is very small, the NMI characteristic value difference of gesture images of the same gesture made by the left hand and the right hand is very small, namely when the gesture image recorded into a database is the right hand, and the household appliance is controlled by the left hand in practice, the control requirement can still be met, and error reporting can not occur; in addition, because NMI features have good translational conversion, rotational conversion, brightness conversion and scaling conversion invariance, gestures and standard gestures when the household appliances are controlled in practice are not influenced; meanwhile, if the standard gestures in the database are respectively composed of actions made by hands of children, teenagers and adults, when other people in the same age range make the same gestures on the household appliance, the household appliance can be accurately controlled, namely, only a certain gesture for controlling the household appliance by a person in a certain age range is recorded in the database, the correct control on the household appliance can be realized when other people in the same age range make corresponding gestures on the household appliance, and the early-stage data acquisition and recording processing work is reduced virtually;
step two, building a gesture acquisition system;
acquiring gestures made by a user by adopting a camera;
step three, establishing a gesture recognition system;
and (3) carrying out feature extraction on the preprocessed gesture image by utilizing NMI, and completing the recognition of gestures made by a user by comparing NMI feature values of the gesture image to be recognized with NMI feature values of various gestures in a database to obtain gesture information, wherein the implementation flow is as follows:
preprocessing a gesture image;
preprocessing a gesture image comprises image graying, image comprehensive filtering processing and image binarization;
(1) Graying of the image;
carrying out graying treatment on the gesture color image by using a weighted average method; because the induction strengths of human eyes to the RGB three colors are different, the weighted average processing is carried out to the RGB three colors of the image, namely:
Gray(i,j)＝0.3R(i,j)+0.59G(i,j)+0.11B(i,j) (1)
(2) Comprehensive filtering treatment of the image;
the graying gesture image is affected by Gaussian noise and spiced salt noise to a certain extent, and the existence of the noise can increase the difficulty of subsequent research; processing the gray level map of the gesture image by adopting comprehensive filtering, namely removing Gaussian noise of the image by adopting average filtering firstly, and removing salt and pepper noise of the image by adopting median filtering secondly so as to improve the quality of the gray level map of the gesture image;
(3) Binarizing the image;
after the gray level image of the gesture image is subjected to comprehensive filtering treatment, the contrast ratio between the gesture lines of the image and the background area is increased, and the degree of distinction between the gesture lines and the background area is further improved; because the Niblack dynamic threshold segmentation algorithm has outstanding performance advantages in the aspects of image segmentation effect, running efficiency and actual operability, the algorithm is utilized to carry out binarization processing on the gesture image after comprehensive filtering processing;
the Niblack dynamic threshold segmentation algorithm calculates the average value of the gray values of pixels in the neighborhood by using all elements in the neighborhood with the size of M multiplied by N:
standard deviation:
then, carrying out weight addition processing on the gray value mean value and the standard variance, and finally, taking the obtained gray value as a judgment threshold value to carry out binarization to obtain a binary feature map of the gesture image; the method comprises the following steps:
H(i,j)＝α×k(i,j)+β×s(i,j) (4)
wherein t (i, j) is a pixel gray value in an MxN neighborhood after the comprehensive filtering process, k (i, j) is a gray value average value, s (i, j) is a standard deviation, H (i, j) is a threshold value, and alpha and beta are correction weights;
(II) extracting gesture features based on NMI;
calculating NMI characteristic values of the binarized gesture image around the center of gravity of the image by using the sum of gray values of pixel points of the binarized image and the center of gravity of the binarized image and the moment of inertia of the binarized image around the center of gravity of the image, wherein the NMI characteristic values of the binarized gesture image around the center of gravity of the image are specifically realized by the following steps:
(1) Regarding a binarized gesture image M multiplied by N as a plane slice, regarding pixels of the binarized image as particles on an XOY plane, and regarding gray values f (i, j) of the pixels of the binarized image as mass of the particles, wherein i and j respectively represent rows and columns of the pixels in a pixel matrix;
(2) Calculating the quality m (f (i, j)) of the binarized gesture image:
(3) Calculating center of gravity of binarized gesture image
(4) Calculating the center of gravity of a binarized gesture image around the image center
Moment of inertia->
(5) Calculating the center of gravity of a binarized gesture image around the image center
Normalized moment of inertia NMI for short:
thirdly, carrying out gesture recognition;
on the basis of obtaining NMI characteristic value X of gesture image to be identified, it is matched with NMI characteristic value Y containing various gestures in database i One-to-one comparison is carried out to obtain a difference A between the two i Then comparing and analyzing the difference value with a judgment threshold value T set by a system; if A i If the difference is less than or equal to T, the gesture to be recognized is recognized to correspond to the gesture which generates the NMI characteristic value difference in the database, otherwise, the gesture does not correspond to the gesture;
and step four, inputting the obtained gesture information into a control system, inputting control commands corresponding to various gesture information into the control system in advance, and then applying the control commands to the household appliances to make corresponding operation changes.
2. The method for intelligently controlling the home appliances based on the NMI as claimed in claim 1, wherein: the first step of controlling the on/off state of the household appliance, the gesture image for adjusting the temperature of the air conditioner/the brightness of the electric lamp/the volume of the television comprises: the control of the household appliance is realized by utilizing the hand action, after the five fingers are fully extended to realize the starting operation of the household appliance, any finger is extended to four fingers to realize the upward adjustment of the temperature/brightness/volume of the appliance, and the upper limit is 5 grades, namely the appliance is in 1 grade after being started; on the basis of adjusting the temperature/brightness/volume of the electric appliance upwards, the downward adjustment of the temperature/brightness/volume of the electric appliance can be realized when the number of fingers is reduced to control the household electric appliance, and the lower limit is 1 grade.
3. The method for intelligently controlling the home appliances based on the NMI as claimed in claim 1, wherein: in order to avoid gesture ambiguity caused by gesture self-shielding under the condition of single viewpoint, the gesture acquisition system adopts a binocular camera with a camera angle of 90 degrees to acquire gestures made by a user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010497497.8A CN111695475B (en) | 2020-06-03 | 2020-06-03 | NMI-based intelligent household appliance control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010497497.8A CN111695475B (en) | 2020-06-03 | 2020-06-03 | NMI-based intelligent household appliance control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111695475A CN111695475A (en) | 2020-09-22 |
| CN111695475B true CN111695475B (en) | 2023-06-27 |
Family
ID=72479459
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010497497.8A Active CN111695475B (en) | 2020-06-03 | 2020-06-03 | NMI-based intelligent household appliance control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111695475B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112379776A (en) * | 2020-11-23 | 2021-02-19 | 南京科盈信息科技有限公司 | Control method for improving recognition accuracy of gesture recognition system in exhibition hall |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853071A (en) * | 2010-05-13 | 2010-10-06 | 重庆大学 | Gesture identification method and system based on visual sense |
| CN103679145A (en) * | 2013-12-06 | 2014-03-26 | 河海大学 | Automatic gesture recognition method |
| WO2015184760A1 (en) * | 2014-06-03 | 2015-12-10 | 深圳Tcl新技术有限公司 | Air gesture input method and apparatus |
| CN110309806A (en) * | 2019-07-08 | 2019-10-08 | 哈尔滨理工大学 | A kind of gesture recognition system and its method based on video image processing |
-
2020
- 2020-06-03 CN CN202010497497.8A patent/CN111695475B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853071A (en) * | 2010-05-13 | 2010-10-06 | 重庆大学 | Gesture identification method and system based on visual sense |
| CN103679145A (en) * | 2013-12-06 | 2014-03-26 | 河海大学 | Automatic gesture recognition method |
| WO2015184760A1 (en) * | 2014-06-03 | 2015-12-10 | 深圳Tcl新技术有限公司 | Air gesture input method and apparatus |
| CN110309806A (en) * | 2019-07-08 | 2019-10-08 | 哈尔滨理工大学 | A kind of gesture recognition system and its method based on video image processing |
Non-Patent Citations (1)
| Title |
|---|
| 余超.基于视觉的手势识别研究.2015,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111695475A (en) | 2020-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108334814B (en) | Gesture recognition method of AR system | |
| CN107390573B (en) | Intelligent wheelchair system based on gesture control and control method | |
| CN109614922A (en) | A kind of dynamic static gesture identification method and system | |
| CN109451634B (en) | Gesture-based electric lamp control method and intelligent electric lamp system thereof | |
| CN103353935A (en) | 3D dynamic gesture identification method for intelligent home system | |
| CN105739702A (en) | Multi-posture fingertip tracking method for natural man-machine interaction | |
| Premaratne et al. | Hand gesture tracking and recognition system using Lucas–Kanade algorithms for control of consumer electronics | |
| CN103279191A (en) | 3D (three dimensional) virtual interaction method and system based on gesture recognition technology | |
| CN106547356B (en) | Intelligent interaction method and device | |
| CN109542233B (en) | Lamp control system based on dynamic gestures and face recognition | |
| CN108197534A (en) | A kind of head part's attitude detecting method, electronic equipment and storage medium | |
| CN110032932B (en) | Human body posture identification method based on video processing and decision tree set threshold | |
| CN105718885B (en) | A kind of Facial features tracking method | |
| CN102402289A (en) | Mouse recognition method for gesture based on machine vision | |
| CN110046544A (en) | Digital gesture identification method based on convolutional neural networks | |
| CN106503619B (en) | Gesture recognition method based on BP neural network | |
| CN112487981A (en) | MA-YOLO dynamic gesture rapid recognition method based on two-way segmentation | |
| Vishwakarma et al. | Simple and intelligent system to recognize the expression of speech-disabled person | |
| CN110956099A (en) | Dynamic gesture instruction identification method | |
| CN112906550A (en) | Static gesture recognition method based on watershed transformation | |
| CN111695475B (en) | NMI-based intelligent household appliance control method | |
| CN114938556B (en) | Automatic adjusting method and device for light of desk lamp, electronic equipment and storage medium | |
| CN110334631B (en) | Sitting posture detection method based on face detection and binary operation | |
| CN116449947B (en) | Automobile cabin domain gesture recognition system and method based on TOF camera | |
| CN112199015A (en) | Intelligent interaction all-in-one machine and writing 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |