CN106599771B - A kind of recognition methods and system of images of gestures - Google Patents
A kind of recognition methods and system of images of gestures Download PDFInfo
- Publication number
- CN106599771B CN106599771B CN201610919827.1A CN201610919827A CN106599771B CN 106599771 B CN106599771 B CN 106599771B CN 201610919827 A CN201610919827 A CN 201610919827A CN 106599771 B CN106599771 B CN 106599771B
- Authority
- CN
- China
- Prior art keywords
- images
- gestures
- image
- judged
- result
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000004364 calculation method Methods 0.000 claims abstract description 33
- 230000011218 segmentation Effects 0.000 claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 22
- 239000000284 extract Substances 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 3
- 230000009467 reduction Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 8
- 235000013399 edible fruits Nutrition 0.000 description 5
- 210000003811 finger Anatomy 0.000 description 5
- 206010011878 Deafness Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 210000004247 hand Anatomy 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 210000004932 little finger Anatomy 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000008569 process Effects 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/107—Static hand or arm
- G06V40/113—Recognition of static hand signs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
Abstract
The present embodiments relate to image recognition technologys, more particularly to a kind of recognition methods and system of images of gestures, the recognition methods of one of images of gestures includes, an images of gestures is obtained, and analysis processing is done to obtain the images of gestures histogram to match with the images of gestures to the images of gestures；A segmentation threshold is formed according to the images of gestures histogram calculation；Divide the images of gestures, contours extract according to the segmentation threshold to form an image to be judged；The image to be judged is identified according to predetermined computation method and forms judging result output, it adopts this method, on the one hand the basic data amount of image recognition is reduced, improve the recognition efficiency of images of gestures, on the other hand judgement purpose only can be realized by single images of gestures, no setting is required multiple acquisition units, reduction hardware manufacturing cost.
Description
Technical field
The present embodiments relate to the recognition methods and system of image recognition technology more particularly to a kind of images of gestures.
Background technique
With the development of communication technology, terminal is integrated with more and more functions, so that the system function of terminal arranges
More and more corresponding applications are contained in table.It can be related to some special services in some applications, specially be supplied to special population
It uses, for example, to gesture identification service provided by deaf-mute.These applications are needed using image collecting device, such as infrared, super
The equipment such as sound wave, multi-cam acquire arm posture image, that is, images of gestures of user, in turn, according to hand collected in real time
Gesture image carries out the identifying processing of images of gestures, to export corresponding voice signal.But in the prior art, by taking the photograph more
As the image collecting device that figure equipment obtains, the image data amount identified is relatively large, and then causes recognition speed slow, identifies
Rate is low.
Summary of the invention
In view of the deficiencies of the prior art, the present invention provides the recognition methods and system of a kind of images of gestures, it is intended to improve hand
The recognition speed and recognition accuracy of gesture image.
In a first aspect, the embodiment of the present invention provides a kind of recognition methods of images of gestures, wherein include:
It obtains an images of gestures, and analysis processing is done to the images of gestures to obtain and match with the images of gestures
Images of gestures histogram；
A segmentation threshold is formed according to the images of gestures histogram calculation；
Divide the images of gestures according to the segmentation threshold to form an image to be judged；
The image to be judged is identified according to predetermined computation method and forms judging result output.
Preferably, the recognition methods of above-mentioned images of gestures, wherein obtain the images of gestures, and to the gesture figure
Include: as doing analysis processing to obtain the images of gestures histogram
Obtain the images of gestures；
The images of gestures is divided and extracts hand images, binaryzation obtains binaryzation hand images；
The images of gestures histogram is formed according to the binaryzation hand images.
Preferably, the recognition methods of above-mentioned images of gestures, wherein；It is described wait judge according to the identification of scheduled calculation method
Image simultaneously forms the judging result output, includes:
Obtain the profile perimeter of the image to be judged, the palm width of image to be judged；
According to the profile perimeter, the palm width in conjunction with the calculation method identify described in image to be judged and formed
The judging result.
Preferably, the recognition methods of above-mentioned images of gestures, wherein；The judging result includes the first judging result, the
Two judging results, third judging result；The calculation method includes the first algorithm, wide according to the profile perimeter, the palm
Degree image to be judged in conjunction with described in calculation method identification simultaneously forms the judging result, comprising:
It is preset with first threshold range；
It calculates to form the first calculated result in conjunction with first algorithm according to the profile perimeter, the palm width；
Wherein, first algorithm is
K=L/W；
K is first calculated result；L is the profile perimeter；W is the palm width；
Judge whether first calculated result matches the first interval；In first calculated result matching described the
First judging result is exported in the state of one section.
Preferably, the recognition methods of above-mentioned images of gestures, wherein；It further include second interval, 3rd interval；
It is mismatched under first interval state in first calculated result；Continue to judge first calculated result whether
With the second interval or the 3rd interval,
Second judging result is exported in the state that first calculated result matches the second interval；
The third judging result is exported in the state that first calculated result matches the 3rd interval.
Preferably, the recognition methods of above-mentioned images of gestures, wherein；Further include: it further include the 4th section；The calculating side
Method includes the second algorithm,
In the state that first calculated result mismatches the first threshold；Obtain the pixel of the image to be judged
The profile length of summation, image to be judged；
Second is formed in conjunction with second algorithm according to the pixel summation of image to be judged, the profile length of image to be judged
Calculated result；
Judge whether second calculated result matches the 4th section；
It is matched in second calculated result and exports the second judging result under the four-range state；
It is mismatched in second calculated result and exports third judging result under the four-range state；
Wherein, second algorithm are as follows:
K '=Sp/h
K ' is second calculated result；Sp is the pixel summation；H is the profile length.
Second aspect, the present invention provide a kind of identifying system of images of gestures again, wherein include:
Acquisition unit obtains an images of gestures,
Processing unit divides gesture image graph picture and extracts hand images to the images of gestures, then binary conversion treatment obtains
Binaryzation hand images are obtained, do analysis processing to obtain the images of gestures histogram to match with the binaryzation hand images；
Computing unit forms a segmentation threshold according to the images of gestures histogram calculation；
Cutting unit divides the images of gestures, contours extract according to the segmentation threshold to form an image to be judged；
Judging unit identifies the image to be judged according to predetermined computation method and forms judging result output.
Preferably, the identifying system of above-mentioned images of gestures, wherein the processing unit includes:
First processing unit divides the images of gestures and extracts hand images, and binary conversion treatment obtains binaryzation hand
Portion's image；
Histogram calculation device calculates according to the binaryzation hand images and obtains the images of gestures histogram.
Preferably, the identifying system of above-mentioned images of gestures, wherein the calculation method includes the first algorithm, described to sentence
Disconnected unit includes:
First data acquisition facility obtains the profile perimeter of the image to be judged, the palm width of image to be judged；
Judgment means, according to the profile perimeter, the palm width wait judge in conjunction with described in calculation method judgement
Image simultaneously forms the judging result.
Preferably, the identifying system of above-mentioned images of gestures, wherein the judgment means are preset with first interval, described
Judging result includes the first judging result, the second judging result, third judging result；The calculation method includes the first algorithm,
The judgment means further include,
First computing device calculates to form according to the profile perimeter, the palm width in conjunction with first algorithm
One calculated result；
Wherein, first algorithm is
K=L/W；
K is first calculated result；L is the profile perimeter；W is the palm width；
First determining device, judges whether first calculated result matches first interval；In first calculated result
With exporting first judging result in the state of the first interval.
Preferably, the identifying system of above-mentioned images of gestures, wherein the judgment means are also prefabricated with second interval,
Three sections；
Second determining device, judges whether first calculated result matches second interval or 3rd interval, Yu Suoshu first
Calculated result exports second judging result in the state of matching the second interval；Institute is matched in first calculated result
It states and exports the third judging result in the state of 3rd interval.
Preferably, the identifying system of above-mentioned images of gestures, wherein further include: the calculation method includes the second algorithm,
The judgment means are also prefabricated with the 4th section；
Second data acquisition facility obtains pixel summation, the profile length of image to be judged of the image to be judged；
Second calculator, according to the second algorithm described in the pixel summation of image to be judged, the profile length of image to be judged
Form the second calculated result；
Third determining device, judges whether second calculated result matches the 4th section；
It is matched in second calculated result and exports the second judging result under the four-range state；
It is mismatched in second calculated result and exports third judging result under the four-range state；
Wherein second algorithm are as follows:
K '=Sp/h
K ': for the second calculated result；Sp: for pixel summation；H is profile length.
Compared with prior art, the beneficial effects of the present invention are:
In the present invention, single acquisition unit obtains an images of gestures, forms a segmentation threshold by a gesture image histogram
Non- feature image information in images of gestures is removed by the segmentation threshold, feature image information is only extracted, by spy by value
The identification of sign image information forms judging result, adopts this method, on the one hand reduces the basic data amount of image recognition, mention
The high recognition efficiency of images of gestures, on the other hand only can be realized judgement purpose by single images of gestures, it is more that no setting is required
A acquisition unit reduces hardware manufacturing cost.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of the recognition methods of images of gestures provided by the invention；
Fig. 2 is a kind of flow diagram of the recognition methods of images of gestures provided by the invention；
Fig. 3 is a kind of flow diagram of the recognition methods of images of gestures provided by the invention；
Fig. 4 is a kind of flow diagram of the recognition methods of images of gestures provided by the invention；
Fig. 5 is a kind of structural schematic diagram of the identifying system of images of gestures provided by the invention；
Fig. 6 is a kind of identifying system structural schematic diagram of images of gestures provided by the invention.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
As shown in Figure 1, the present embodiment provides a kind of flow diagrams of the recognition methods of images of gestures, wherein include:
Step S110, as shown in Figure 5 a, obtain an images of gestures, and to the images of gestures do analysis processing with obtain with
The images of gestures histogram that the images of gestures matches；Images of gestures is obtained by single acquisition unit.Acquisition units can be
Camera, camera or other image collecting devices.
Step S120, a segmentation threshold as shown in Figure 5 b, is formed according to the images of gestures histogram calculation；
Step S130, the images of gestures as shown in Figure 5 c, is divided according to the segmentation threshold and forms an image to be judged；
Step S140, the image to be judged is identified according to predetermined computation method and form judging result output.
The working principle of the present embodiment is:
Obtain an images of gestures by single photographic device, included at least in the usual images of gestures feature image information and
Non- feature image information leads to the reduction of image recognition efficiency because containing non-feature image information in images of gestures, therefore in order to
Image recognition efficiency is improved, the present invention forms a segmentation threshold by a gesture image histogram, by described in the segmentation threshold
Images of gestures removes the non-characteristic information in images of gestures, to improve images of gestures recognition efficiency, specifically, first to institute
It states images of gestures and does analysis processing to obtain the images of gestures histogram to match with the images of gestures；Then according to gesture figure
As histogram calculation forms a segmentation threshold；Continue to be schemed according to the segmentation threshold segmentation images of gestures formation one wait judge
Picture, wherein eliminating non-feature image information wait judge to remain with feature image information in images of gestures in image；Finally according to
Predetermined computation method identifies the image to be judged and forms judging result output.
In the present invention, single acquisition unit obtains an images of gestures, forms a segmentation threshold by a gesture image histogram
Non- feature image information in images of gestures is removed by the segmentation threshold, feature image information is only extracted, by spy by value
The identification of sign image information forms judging result, adopts this method, on the one hand reduces the basic data amount of image recognition, mention
The high recognition efficiency of images of gestures, on the other hand only can be realized judgement purpose by single images of gestures, it is more that no setting is required
A acquisition unit reduces hardware manufacturing cost.
As further preferred embodiment, the recognition methods of above-mentioned images of gestures, wherein Yu Zhihang step S110,
An images of gestures is obtained, and straight to obtain the images of gestures to match with the images of gestures to images of gestures analysis processing
In square figure, further comprise:
Step S1101, the images of gestures is obtained；
Step S1102, the images of gestures is divided and extracts hand images, binaryzation obtains binaryzation hand images.
Step S1103, the images of gestures histogram is formed according to the binaryzation hand images.Images of gestures histogram
For binaryzation hand images vertical direction histogram, binaryzation hand images vertical direction histogram has reacted binaryzation hand figure
The distributed intelligence of pixel summation corresponding to vertical direction as in.
The binaryzation hand images histogram calculation that is formed through the above steps forms a segmentation threshold, in the present invention, point
Cutting threshold value is value corresponding to pixel summation maximum value.Divide the images of gestures according to the segmentation threshold to form one wait sentence
Disconnected image.Wait judge to eliminate the non-feature image information such as the colour of skin, illumination and complex scene in image.
As shown in Fig. 2, as further preferred embodiment, the recognition methods of above-mentioned images of gestures, wherein step
S140, the image to be judged is identified according to scheduled calculation method and forms judging result output, include:
Step S1401, the palm width of the profile perimeter of image to be judged described in acquisition, image to be judged, wherein palm
The peak value of width matching histogram, or perhaps the value at cut-off rule；
Step S1402, institute is identified in conjunction with the calculation method according to the judgement image outline perimeter, the palm width
It states image to be judged and forms the judging result.
As further preferred embodiment, the recognition methods of above-mentioned images of gestures, wherein the judging result includes
First judging result, the second judging result, third judging result；The calculation method includes the first algorithm, executes step
S1402, according to the profile perimeter, the palm width in conjunction with the calculation method identify described in image to be judged and form institute
It states in judging result, specifically includes:
Step S140211, it is preset with first threshold, the first threshold virtual value is 2.3~2.8；
Step S140212, it calculates to form first in conjunction with first algorithm according to the profile perimeter, the palm width
Calculated result；
Wherein, first algorithm is
K=L/W；
K is first calculated result；L is the profile perimeter；W is the palm width；
Step S140213, judge whether first calculated result matches first interval；In first calculated result
With exporting first judging result in the state of the first interval.
As further preferred embodiment, on the basis of the recognition methods of the images of gestures of Yu Shangshu, wherein further include
Second threshold virtual value described in two threshold values is 3.2-3.6；
Step S140214, in the state that the first calculated result of Yu Suoshu mismatches first interval, continue to judge described first
Whether calculated result matches second interval or 3rd interval.Wherein, first interval is (0, M1), M1The secondth area ∈ (2.3~2.8)
Between be [M1, M2), 3rd interval is [M2, ∞), M2∈ (3.2~3.6).Step S140215, the first calculated result of Yu Suoshu
With exporting second judging result in the state of second range；
Step S140216, the third is exported in the state of the first calculated result of Yu Suoshu matching third range to sentence
Disconnected result.
Enumerate a kind of specific embodiment:
The first judging result is pre-seted to match gesture of clenching fist, the second judging result is to match scissors gesture, third
Judging result is to match palm hand gesture.
The profile perimeter of the image to be judged, the profile width of image to be judged are obtained first, according to the profile week
Long, the described profile width calculates to form the first calculated result in conjunction with the first algorithm K=L/W；Judge that described first calculates knot
Whether fruit matches first interval；First judgement is exported in the state that first calculated result matches the first interval
As a result, i.e. current image recognition to be judged is gesture of clenching fist.The first interval is mismatched in first calculated result
Under state, continue to judge whether first calculated result matches second interval or 3rd interval, the first calculated result of Yu Suoshu
It matches and exports second judging result in the state of the second interval；I.e. current image recognition to be judged is scissors hand
Gesture.The third judging result is exported in the state that first calculated result matches the 3rd interval；It is i.e. current to
Judge image recognition for palm hand gesture.
Embodiment two:
As shown in figure 3, the recognition methods of above-mentioned images of gestures, wherein the judging result include the first judging result,
Second judging result, third judging result；Step S1402, according to the profile perimeter, the palm width in conjunction with the calculating
Method identifies the image to be judged and forms the judging result, comprising:
Step S140211, it is preset with first interval；Second interval, 3rd interval；Wherein, first interval is (0, M1), M1
∈ (2.3~2.8) second interval is [M1, M2), 3rd interval is [M2, ∞), M2∈ (3.2~3.6).
Step S140222, it calculates to form first in conjunction with first algorithm according to the profile perimeter, the palm width
Calculated result；
Wherein, first algorithm is
K=L/W；
K1For first calculated result；L is the profile perimeter；W is the palm width；
Step S140223, judge whether first calculated result matches first interval or second interval or third area
Between；
Step S140224, described first is exported in the state of the first calculated result of Yu Suoshu matching first interval to sentence
Disconnected result；
Step S140225, described second is exported in the state of the first calculated result of Yu Suoshu matching second interval to sentence
Disconnected result；
Step S140226, the third is exported in the state of the first calculated result of Yu Suoshu matching 3rd interval to sentence
Disconnected result.
A kind of specific embodiment is enumerated again:
The first judging result is pre-seted to match gesture of clenching fist, the second judging result is to match scissors gesture, third
Judging result is to match palm hand gesture.
The profile perimeter of the image to be judged, the palm width of image to be judged are obtained first, according to the profile week
Long, the described palm width calculates to form the first calculated result in conjunction with the first algorithm K=L/W；Judge that described first calculates knot
Whether fruit matches first interval or second interval or 3rd interval；First range is matched in first calculated result
First judging result is exported under state, i.e., current image recognition to be judged is gesture of clenching fist.Knot is calculated in described first
Fruit exports second judging result in the state of matching the second interval；I.e. current image recognition to be judged is scissors hand
Gesture.The third judging result is exported in the state that first calculated result matches the 3rd interval；It is i.e. current to
Judge image recognition for palm hand gesture.
It should be noted that above-mentioned range is only for example, not limitation of the invention further, in reality
In use process, it can increase or increase as the case may be section to increase the quantity of range to match different gestures.Example
Four ranges are such as set, identify four kinds of images of gestures etc. to judge.It different one illustrates herein.
Embodiment three
It is according to the profile perimeter, the palm width in the above embodiments one, embodiment two in conjunction with described
One algorithm calculates to form the first calculated result；There are a defects for such mode, i.e., when profile perimeter, institute between gesture A and gesture B
State palm width gap it is smaller when, the matching accuracy of the above method is relatively low, such as stretches out index finger, middle finger, the third finger
The profile perimeter of the images of gestures of the profile perimeter of images of gestures, the ratio of palm width and stretching index finger, the third finger, little finger of toe,
The ratio of profile width is close or similar, is easy for causing to misidentify using above-mentioned technical solution.
In order to improve the accuracy of images of gestures, as shown in figure 4, the present invention provides a kind of identification side of images of gestures again
Method specifically includes:
Step S310, obtain an images of gestures, and to the images of gestures divide extract hand images, to hand images into
Row binary conversion treatment, then it is straight to obtain the images of gestures to match with the images of gestures with binary image analysis processing
Fang Tu.
Step S320, a segmentation threshold is formed according to the images of gestures histogram calculation.
Step S330, the binaryzation hand images are divided to form an image to be judged according to the segmentation threshold.
Step S340, the image to be judged is identified according to predetermined computation method and form judging result output.Wherein
The calculation method includes the first algorithm.
It specifically includes: step S3401, being preset with first threshold K1, further, the first threshold K1Virtual value is 2.3
~2.8；
Step S3402, it calculates to form the first meter in conjunction with first algorithm according to the profile perimeter, the palm width
Calculate result；
Wherein, first algorithm is
K=L/W；
K is the first calculated result；L is the profile perimeter；W is the palm width；
Step S3403, judge whether first calculated result matches the first interval；In first calculated result
It matches and exports first judging result in the state of the first interval.
Step S3404, in the state of the first calculated result of Yu Suoshu mismatch first interval；It obtains described wait judge
Pixel summation, the profile length of image to be judged of image；
Step S3405, according to the pixel summation of image to be judged, the profile length of image to be judged, in conjunction with described second
Algorithm forms the second calculated result；Wherein, second algorithm are as follows:
K '=Sp/h
K ': for second calculated result；Sp: for the pixel summation；H ' is the profile length.
Step S3406, third threshold k is set3, effective range value 38~45；Judge second calculated result whether
It is (0, M with the 4th range3), wherein M3∈(38、45)。
Step S3407, the second calculated result of Yu Suoshu matches and exports the second judging result under the four-range state；
Step S3408, the second calculated result of Yu Suoshu mismatches output third judgement knot under the four-range state
Fruit.
The present embodiment, above-described embodiment one, implement two basis on, in conjunction with image to be judged pixel summation, to
Judge that the profile length data of image are further to image to be judged, it is intended to improve the accuracy of image recognition.
Enumerate a kind of specific embodiment:
The first judging result is pre-seted to match gesture of clenching fist, the second judging result is to match scissors gesture, third
Judging result is to match palm hand gesture.
The profile perimeter of the image to be judged, the palm width of image to be judged are obtained first, according to the profile week
Long, the described palm width calculates to form the first calculated result in conjunction with the first algorithm K=L/W；Judge that described first calculates knot
Whether fruit matches the first range, and the first calculated result of Yu Suoshu exports first judgement in the state of matching first range
As a result, i.e. current image recognition to be judged is gesture of clenching fist.The first interval is mismatched in first calculated result
Under state, pixel summation, the profile length of image to be judged of the image to be judged are obtained；According to the pixel of image to be judged
The profile length of summation, image to be judged forms the second calculated result in conjunction with second algorithm；Continue to judge second meter
Calculate whether result matches the 4th range；The second judgement of output in the state that second calculated result matches the 4th range
As a result；I.e. current image recognition to be judged is scissors gesture.The 4th range is mismatched in second calculated result
The third judging result is exported under state；I.e. current image recognition to be judged is palm hand gesture.
It should be noted that above-mentioned threshold range is only for example, not limitation of the invention further, In
In actual use, it can increase or increase as the case may be threshold value to increase the quantity of range to match different hands
Gesture.Such as four ranges of setting, four kinds of images of gestures etc. are identified to judge.It different one illustrates herein.
Example IV
As shown in fig. 6, a kind of identifying system structural schematic diagram of images of gestures.
For a kind of recognition methods of above-mentioned images of gestures, the present invention provides a kind of identifying system of images of gestures again,
Wherein, comprising:
Acquisition unit 1 obtains an images of gestures,
Processing unit 2 divides the images of gestures and extracts hand images, then binary conversion treatment to the images of gestures
Binaryzation hand images are obtained, do analysis processing to obtain the images of gestures histogram to match with the binaryzation hand images
Figure；
Computing unit 3 forms a segmentation threshold according to the images of gestures histogram calculation；
Cutting unit 4 is divided the images of gestures according to the segmentation threshold and is extracted to form an image to be judged；
Judging unit 5 identifies the image to be judged according to predetermined computation method and forms judging result output.
As further preferred embodiments, the identifying system of above-mentioned images of gestures, wherein the processing unit 2 wraps
It includes:
First processing unit 21 divides the images of gestures and extracts hand images, and binary conversion treatment obtains binaryzation
Hand images；
Histogram technology device 22 calculates according to the binaryzation hand images and obtains the images of gestures histogram.
As further preferred embodiments, the identifying system of above-mentioned images of gestures, wherein the calculation method includes
First algorithm, the judging unit 5 include:
First data acquisition facility 51 obtains the profile perimeter of the image to be judged, the palm width of image to be judged；
Judgment means 52, according to the profile perimeter, the palm width wait sentence in conjunction with described in first algorithm judgement
Disconnected image simultaneously forms the judging result.
As further preferred embodiments, the identifying system of above-mentioned images of gestures, wherein the judgment means 52 are pre-
Equipped with the first range, the judging result includes the first judging result, the second judging result, third judging result；The judgement
Device 52 further includes,
First computing device 53 is calculated in conjunction with the calculation method and to be formed according to the profile perimeter, the palm width
First calculated result；
Wherein, first algorithm is
K=L/W；
K is first calculated result；L is the profile perimeter；W is the palm width；
First determining device, judges whether first calculated result matches the first range；In first calculated result
With exporting first judging result in the state of first range.
As further preferred embodiments, the identifying system of above-mentioned images of gestures, wherein the judgment means 52 are also
It is prefabricated with the second range, third range；
Second determining device, judges whether first calculated result matches the second range or third range, Yu Suoshu first
Calculated result exports second judging result in the state of matching second range；Institute is matched in first calculated result
It states and exports the third judging result in the state of third range.
As further preferred embodiments, the identifying system of above-mentioned images of gestures, wherein
The first threshold, virtual value are 2.3~2.8, the second threshold, virtual value 3.2-3.4, and first interval is
(0、M1), M1∈ (2.3~2.8) second interval is [M1, M2), 3rd interval is [M2, ∞), M2∈ (3.2~3.6).
As further effective embodiment, the identifying system of above-mentioned images of gestures, wherein further include: the calculating
Method includes the second algorithm, and the judgment means 52 are also prefabricated with the 4th range；
Second data acquisition facility obtains pixel summation, the profile length of image to be judged of the image to be judged；
Second calculator, according to the second algorithm described in the pixel summation of image to be judged, the profile length of image to be judged
Form the second calculated result；
Third determining device, judges whether second calculated result matches the 4th range；
The second judging result is exported in the state that second calculated result matches the 4th range；
Third judging result is exported in the state that second calculated result mismatches the 4th range；
Wherein second algorithm are as follows:
K '=Sp/h
K ': for the second calculated result；Sp: for pixel summation；H ': for profile length.
As further preferred embodiments, the identifying system of above-mentioned images of gestures, wherein the 4th range be (0,
M3), wherein M3∈(38、45)。
The identifying system of above-mentioned images of gestures can realize the recognition methods of above-mentioned images of gestures, and working principle is similar, this
Place does not repeat them here.
Although various aspects of the invention provide in the independent claim, other aspects of the invention include coming from
The combination of the dependent claims of the feature of described embodiment and/or the feature with independent claims, and not only
It is the combination clearly provided in claim.
It is to be noted here that although these descriptions are not the foregoing describe example embodiment of the invention
It should be understood in a limiting sense.It is wanted on the contrary, several change and modification can be carried out without departing from such as appended right
The scope of the present invention defined in asking.
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (8)
1. a kind of recognition methods of images of gestures characterized by comprising
An images of gestures is obtained, and analysis processing is done to obtain the gesture to match with the images of gestures to the images of gestures
Image histogram；
A segmentation threshold is formed according to the images of gestures histogram calculation；
Divide the images of gestures according to the segmentation threshold to form an image to be judged；
The image to be judged is identified according to predetermined computation method and forms judging result output, wherein the judging result packet
Include the first judging result, the second judging result, third judging result；The predetermined computation method includes the first algorithm, wherein pressing
The image to be judged is identified according to scheduled calculation method and forms the judging result output, includes:
Obtain the profile perimeter of the image to be judged, the palm width of image to be judged；
It is preset with first threshold range；
It calculates to form the first calculated result in conjunction with first algorithm according to the profile perimeter, the palm width；
Wherein, first algorithm is
K=L/W；
K is first calculated result；L is the profile perimeter；W is the palm width；
Judge whether first calculated result matches first interval；The first interval is matched in first calculated result
First judging result is exported under state.
2. the recognition methods of images of gestures according to claim 1, which is characterized in that obtain the images of gestures, and right
The images of gestures does analysis processing to obtain the images of gestures histogram
Obtain the images of gestures；
The images of gestures is divided and extracts hand images, binaryzation obtains binaryzation hand images；
The images of gestures histogram is formed according to the binaryzation hand images.
3. the recognition methods of images of gestures according to claim 1, which is characterized in that further include second interval, third area
Between；
It is mismatched under first interval state in first calculated result；Continue to judge whether first calculated result matches institute
Second interval or the 3rd interval are stated,
Second judging result is exported in the state that first calculated result matches the second interval；
The third judging result is exported in the state that first calculated result matches the 3rd interval.
4. the recognition methods of images of gestures according to claim 1, which is characterized in that further include: it further include the 4th section；
The calculation method includes the second algorithm,
In the state that first calculated result mismatches the first threshold；The pixel for obtaining the image to be judged is total
With the profile length of image to be judged；
Second is formed in conjunction with second algorithm and is calculated according to the pixel summation of image to be judged, the profile length of image to be judged
As a result；
Judge whether second calculated result matches the 4th section；
It is matched in second calculated result and exports the second judging result under the four-range state；
It is mismatched in second calculated result and exports third judging result under the four-range state；
Wherein, second algorithm are as follows:
K '=Sp/h
K ' is second calculated result；Sp is the pixel summation；H is the profile length.
5. a kind of identifying system of images of gestures characterized by comprising
Acquisition unit obtains an images of gestures,
Processing unit divides gesture image graph picture and extracts hand images to the images of gestures, then binary conversion treatment obtains two
Value hand images do analysis processing to obtain the images of gestures histogram to match with the binaryzation hand images；
Computing unit forms a segmentation threshold according to the images of gestures histogram calculation；
Cutting unit divides the images of gestures, contours extract according to the segmentation threshold to form an image to be judged；
Judging unit identifies the image to be judged according to predetermined computation method and forms judging result output, wherein is described
Calculation method includes the first algorithm, and the judging unit includes:
First data acquisition facility obtains the profile perimeter of the image to be judged, the palm width of image to be judged；
Judgment means, according to the profile perimeter, palm width image to be judged in conjunction with described in calculation method judgement
And form the judging result；The judgment means are preset with first interval, and the judging result includes the first judging result, the
Two judging results, third judging result；The calculation method includes the first algorithm, and the judgment means further include,
First computing device calculates to form the first meter according to the profile perimeter, the palm width in conjunction with first algorithm
Calculate result；
Wherein, first algorithm is
K=L/W；
K is first calculated result；L is the profile perimeter；W is the palm width；
First determining device, judges whether first calculated result matches first interval；Institute is matched in first calculated result
It states and exports first judging result in the state of first interval.
6. the identifying system of images of gestures according to claim 5, which is characterized in that the processing unit includes:
First processing unit divides the images of gestures and extracts hand images, and binary conversion treatment obtains binaryzation hand figure
Picture；
Histogram calculation device calculates according to the binaryzation hand images and obtains the images of gestures histogram.
7. the identifying system of images of gestures according to claim 6, which is characterized in that the judgment means are also prefabricated with
Two sections, 3rd interval；
Second determining device, judges whether first calculated result matches second interval or 3rd interval, and Yu Suoshu first is calculated
As a result it matches and exports second judging result in the state of the second interval；In first calculated result matching described the
The third judging result is exported in the state of three sections.
8. the identifying system of images of gestures according to claim 6, which is characterized in that further include: the calculation method packet
The second algorithm is included, the judgment means are also prefabricated with the 4th section；
Second data acquisition facility obtains pixel summation, the profile length of image to be judged of the image to be judged；
Second calculator is formed according to the second algorithm described in the pixel summation of image to be judged, the profile length of image to be judged
Second calculated result；
Third determining device, judges whether second calculated result matches the 4th section；
It is matched in second calculated result and exports the second judging result under the four-range state；
It is mismatched in second calculated result and exports third judging result under the four-range state；
Wherein second algorithm are as follows:
K '=Sp/h
K ': for the second calculated result；Sp: for pixel summation；H is profile length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610919827.1A CN106599771B (en) | 2016-10-21 | 2016-10-21 | A kind of recognition methods and system of images of gestures |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610919827.1A CN106599771B (en) | 2016-10-21 | 2016-10-21 | A kind of recognition methods and system of images of gestures |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106599771A CN106599771A (en) | 2017-04-26 |
| CN106599771B true CN106599771B (en) | 2019-11-22 |
Family
ID=58555719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610919827.1A Expired - Fee Related CN106599771B (en) | 2016-10-21 | 2016-10-21 | A kind of recognition methods and system of images of gestures |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106599771B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111382644A (en) * | 2018-12-29 | 2020-07-07 | Tcl集团股份有限公司 | Gesture recognition method and device, terminal equipment and computer readable storage medium |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101661556A (en) * | 2009-09-25 | 2010-03-03 | 哈尔滨工业大学深圳研究生院 | Static gesture identification method based on vision |
| CN102184021B (en) * | 2011-05-27 | 2013-06-12 | 华南理工大学 | Television man-machine interaction method based on handwriting input and fingertip mouse |
| WO2013078657A1 (en) * | 2011-12-01 | 2013-06-06 | Nokia Corporation | A gesture recognition method, an apparatus and a computer program for the same |
| US10839227B2 (en) * | 2012-08-29 | 2020-11-17 | Conduent Business Services, Llc | Queue group leader identification |
| CN103530613B (en) * | 2013-10-15 | 2017-02-01 | 易视腾科技股份有限公司 | Target person hand gesture interaction method based on monocular video sequence |
| CN103598870A (en) * | 2013-11-08 | 2014-02-26 | 北京工业大学 | Optometry method based on depth-image gesture recognition |
| CN103679145A (en) * | 2013-12-06 | 2014-03-26 | 河海大学 | Automatic gesture recognition method |
| CN104299004B (en) * | 2014-10-23 | 2018-05-01 | 浙江大学 | A kind of gesture identification method based on multiple features fusion and finger tip detection |
| CN105893925A (en) * | 2015-12-01 | 2016-08-24 | 乐视致新电子科技（天津）有限公司 | Human hand detection method based on complexion and device |
| CN105787471B (en) * | 2016-03-25 | 2019-05-03 | 南京邮电大学 | It is a kind of applied to help the elderly help the disabled Information Mobile Service robot control gesture identification method |
-
2016
- 2016-10-21 CN CN201610919827.1A patent/CN106599771B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN106599771A (en) | 2017-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Verma et al. | Gesture recognition using kinect for sign language translation | |
| CN107862295B (en) | Method for identifying facial expressions based on WiFi channel state information | |
| Rao et al. | Sign Language Recognition System Simulated for Video Captured with Smart Phone Front Camera. | |
| CN107239735A (en) | A kind of biopsy method and system based on video analysis | |
| CN111652017B (en) | Dynamic gesture recognition method and system | |
| Raheja et al. | Android based portable hand sign recognition system | |
| JP2007257088A (en) | Robot device and its communication method | |
| CN110942037A (en) | Action recognition method for video analysis | |
| Nandwana et al. | A survey paper on hand gesture recognition | |
| Padhi et al. | Hand Gesture Recognition using DenseNet201-Mediapipe Hybrid Modelling | |
| CN106599771B (en) | A kind of recognition methods and system of images of gestures | |
| CN109145861B (en) | Emotion recognition device and method, head-mounted display equipment and storage medium | |
| CN113128368B (en) | Method, device and system for detecting character interaction relationship | |
| CN111126250A (en) | Pedestrian re-identification method and device based on PTGAN | |
| Shi et al. | An FPGA-based smart camera for gesture recognition in HCI applications | |
| CN110110631B (en) | Method and equipment for recognizing and making call | |
| CN111783677A (en) | Face recognition method, face recognition device, server and computer readable medium | |
| CN115205943A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| Ye et al. | Human motion analysis based on extraction of skeleton and dynamic time warping algorithm using RGBD camera | |
| KR101158016B1 (en) | Apparatus and method for detecting upper body pose and hand shape | |
| CN113894779A (en) | Multi-mode data processing method applied to robot interaction | |
| Lekhashri et al. | Use of motion-print in sign language recognition | |
| Monica et al. | Recognition of medicine using cnn for visually impaired | |
| CN106503650B (en) | A kind of recognition methods and system of images of gestures | |
| CN112800816A (en) | Video motion recognition detection method based on multiple models |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200429 Address after: Room 612, Zone D, building 20, No. 1-42, Lane 83, Hongxiang North Road, Wanxiang Town, Pudong New Area, Shanghai Patentee after: Shanghai Yingchen Information Technology Co., Ltd Address before: 200233, 1122 North Qinzhou Road, Shanghai, 90, building 8, Xuhui District Patentee before: SHANGHAI PARTNERX ROBOTICS Co.,Ltd. |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20210113 Address after: No. 10, Lane 255, Xiaotang Road, Fengxian District, Shanghai, 2010 Patentee after: Shanghai Wujin robot Co.,Ltd. Address before: Room 612, area D, building 20, no.1-42, Lane 83, Hongxiang North Road, Wanxiang Town, Pudong New Area, Shanghai, 201313 Patentee before: Shanghai Yingchen Information Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210331 Address after: 200233 8th floor, building 90, 1122 Qinzhou North Road, Xuhui District, Shanghai Patentee after: SHANGHAI PARTNERX ROBOTICS Co.,Ltd. Address before: No. 10, Lane 255, Xiaotang Road, Fengxian District, Shanghai, 2010 Patentee before: Shanghai Wujin robot Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191122 Termination date: 20201021 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |