CN102129332A - Detection method and device of touch points for image recognition - Google Patents
Detection method and device of touch points for image recognition Download PDFInfo
- Publication number
- CN102129332A CN102129332A CN 201110054191 CN201110054191A CN102129332A CN 102129332 A CN102129332 A CN 102129332A CN 201110054191 CN201110054191 CN 201110054191 CN 201110054191 A CN201110054191 A CN 201110054191A CN 102129332 A CN102129332 A CN 102129332A
- Authority
- CN
- China
- Prior art keywords
- touch point
- pixel
- image
- unit
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a detection method of touch points for image recognition. The method comprises the following steps of: converting a grayscale image into a binary image and acquiring the number of touch points in the binary image and/or the position information of each touch point, wherein the position information is the coordinate of a pixel in the center position of the touch points; calculating the length-width ratio of the minimum enclosing rectangle of each touch point; and correcting the number of the touch points according to a preset range which the length-width ratio belongs to. The invention also discloses a detection device of the touch points for image recognition. By adopting the invention, the detection and the correction of multi-point touch are realized to acquire accurate number and position of the touch points, and the touch points with closer touch positions or even the touch points adhered mutually can be detected through the detection method and device disclosed by the invention so that the human-computer interactive operation experience of users is further improved.
Description
Technical field
The present invention relates to a kind of image recognition technology, relate in particular to a kind of detection method and device of touch point of image recognition.
Background technology
Screen touch technology and people work the life closely bound up, for example, the cash dispenser of bank has touch screen function mostly, and the operating terminal of this touch controllable function is all arranged in a lot of government affairs halls, the library, also has a lot of touch-screen mobile phones, MP4, liquid crystal display digital camera or the like.Touch screen functionality development from the simple to the complex, only support single-point touches from initial product, can realize multiple point touching control till now, its technology has a qualitative leap.So-called single-point touches, a bit realizing on finger touch screen controlled exactly.Such as our every days of the POS terminating machine in supermarket, the perhaps operation of on the check-in on airport terminal, carrying out nearby.Although single-point touches is shielded and resistive touch screen technology is very astonishing and have the revolution meaning, can but why the interaction of user and equipment be confined to a finger? when just defining man-machine interaction, once can only do single action in certain zone of screen with a finger.To this, Apple has made immeasurable contribution for the user interface revolution, and the iPhone of its release has adopted inductance capacitance formula touch-screen, thereby makes multiple point touching become possibility.Want to give full play to the function of application and operating system, just need a plurality of fingers on screen, make the availability that action can realize that it is best.Because this technology has been arranged, the user can carry out single-point touches by finger, also can realize indiscriminately ad. as one wishes controlling, thereby carry out various operations better and more comprehensively to click, double-click, translation, to push, roll and different gestures such as rotation are carried out multiple point touching.
Multiple point touching refers to and allows the computer user to come a kind of operator scheme of control computer simultaneously by a plurality of fingers or contact, continuous exploration and innovation by global fan, up to the present, five kinds of technology that can help the fan to build stable multiple point touching platform have been arranged, and they are respectively: the frustrated total internal reflection multiple point touching technology of being started by Jeff professor Han (FTIR); The back scatter light multiple point touching technology (Rear-DI) that the Surface of Microsoft adopts; Laser plane multiple point touching technology (LLP) by Alex Popovich proposition; By Nima Motamedi LED plane multiple point touching technology (LED-LP) is proposed; Scattered light plane multiple point touching technology (DSI) by Tim Roth proposition.
In the middle of above-mentioned five kinds of technology, be based on as FTIR and Rear-DI technology that reflection takes place in the touch place light principle realizes, but these technology have a drawback, be exactly that existing touch point detection method can lose efficacy even mistake is surveyed when the touch location distance of touch objects on screen is very near even inter-adhesive.For example, with two finger while touch screen, if place, two touch points distance is very near, existing technology just can only detect a touch point on touch-screen, be that it can be judged as a big touch point with two touch points mistakenly, thereby the program that has just misled background system is carried out.And then influence user's touch operation, make the user well not realize man-machine interaction by touch-screen.
Summary of the invention
Technical matters to be solved by this invention is, a kind of detection method and device of touch point of image recognition is provided.
In order to solve the problems of the technologies described above, the invention provides a kind of detection method of touch point of image recognition: with greyscale image transitions is bianry image.Obtain quantity and/or each described touch point position information of touch point in the described bianry image, described positional information is the coordinate of pixel that is positioned at the center of touch point.Calculate the length breadth ratio of boundary rectangle of the area minimum of each described touch point.Proofread and correct the quantity of described touch point according to the preset range under the described length breadth ratio.
Correspondingly, the present invention also provides a kind of pick-up unit of touch point of image recognition, comprising:
The two-value converting unit, being used for greyscale image transitions is bianry image;
With the acquiring unit that described two-value converting unit links to each other, be used for obtaining quantity and/or each described touch point position information of the described bianry image touch point of described two-value converting unit conversion;
With the computing unit that described acquiring unit links to each other, be used to calculate the length breadth ratio of the boundary rectangle of each described touch point minimum;
With the correcting unit that described computing unit links to each other, the preset range under the described length breadth ratio that is used for calculating according to described computing unit is proofreaied and correct the quantity of described touch point.
Implement the present invention, have following beneficial effect:
Adopt the present invention, can realize detection, analyze the quantity and the positional information of touch point, and correct the quantity and the positional information of each touch point multiple point touching.In addition, the present invention can also become bianry image with greyscale image transitions, distinguishes bonding touch point, removes isolated point, and then obtains accurately touch point image clearly, even touch location is very near even bonding mutually, also can detects by the present invention and obtain.And then the operating experience of raising user man-machine interaction.
Description of drawings
Fig. 1 is the process flow diagram of detection method of the touch point of a kind of image recognition of the present invention;
Fig. 2 is the process flow diagram of detection method one embodiment of the touch point of a kind of image recognition of the present invention;
Fig. 3 is the sub-piece synoptic diagram of detection of detection method of the touch point of a kind of image recognition of the present invention;
Fig. 4 is the structural representation of pick-up unit of the touch point of a kind of image recognition of the present invention;
Fig. 5 is the structural representation of an embodiment of pick-up unit of the touch point of a kind of image recognition of the present invention;
Fig. 6 is the original optical imagery that prior art is obtained;
Fig. 7 handles the preferred bianry image that the back obtains through one embodiment of the invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with accompanying drawing.
Fig. 1 is the process flow diagram of detection method of the touch point of a kind of image recognition of the present invention, comprising:
S101 is a bianry image with greyscale image transitions.
Need to prove that on the full ultrared screen of a piece of cloth, when with finger or other object touch screen, infrared ray will reflect in the touch point location, utilization is the reflection of light principle here.After above-mentioned infrared ray reflects, can obtain optical imagery by catching described infrared ray, need to filter with only catching infrared ray the pick-up lens of visible light here, infrared ray just can form optical imagery by pick-up lens on photosensitive material so.The pixel value information that comprises pixel in the information of detected described touch point, the pixel value of gathering each pixel carries out layout, and optical imagery just converts gray level image to.In gray level image, the pixel value in zone, touch point can be than the pixel value height in zone, non-touch point, thereby is shown as brighter whiter effect in gray level image.It is multiple to become the method for bianry image to have greyscale image transitions, and existing technology is the method for fixed threshold conversion, and assignment was 1 when promptly gray-scale value was greater than a certain fixed threshold, otherwise assignment is 0.
S102 obtains quantity and/or each described touch point position information of touch point in the described bianry image, and described positional information is the coordinate of pixel that is positioned at the center of touch point.
Need to prove, in above-mentioned bianry image, having only touch point location ability assignment is 1, so can be that 1 zone obtains the information of corresponding described touch point by detecting assignment, the information of described touch point can comprise quantity and/or each described touch point position information of touch point.The quantity of described touch point is that assignment is the number of 1 simply connected region, and described touch point position information is that assignment is the coordinate of pixel of the center in 1 zone.
S103 calculates the length breadth ratio of boundary rectangle of the area minimum of each described touch point.
Need to prove, be 1 zone by above-mentioned assignment, can obtain its minimum boundary rectangle.Described minimum boundary rectangle is meant and comprises the boundary rectangle that assignment is 1 simply connected region minimum.By described rectangle, can try to achieve the length of described rectangle and wide, wherein, the longer sides of definition rectangle is the length of rectangle, the definition rectangle be the wide of rectangle than minor face, obtain length and width ratio.
S104 proofreaies and correct the quantity of described touch point according to the preset range under the described length breadth ratio.
Need to prove that according to the preset range under the described length breadth ratio of each touch point, we can determine that described touch point is that a touch point or several touch point are mutually bonding.For example, we can belong to first preset range according to the length breadth ratio of touch point and determine that described touch point is a touch point; Belong to second preset range according to the length breadth ratio of touch point and determine that described touch point is that two touch points are mutually bonding; Belong to the 3rd preset range according to the length breadth ratio of touch point and determine that described touch point is that three touch points are mutually bonding.So analogize
Preferably, can choose described preset range and be 1≤q＜1.8 o'clock, be judged to be a touch point; When q 〉=1.8, be judged to be the phase adhesion of two or more touch points, wherein q is described length breadth ratio.Here distinguishing described touch point and be a touch point and still be the mutually bonding critical point in a plurality of touch points is described length breadth ratio q=1.8.Described length breadth ratio q=1.8 is an empirical value, obtains through repeated touch experiment.When q 〉=1.8, decidable is the phase adhesion of a plurality of touch points, at this moment, can redefine the quantity of described touch point, and the center of each described touch point.Can obtain by experiment as for second preset range, the 3rd preset range and follow-up preset range.
In obtaining described bianry image, after the step of the quantity of touch point and/or each described touch point position information, also comprise:, calculate the average area of described touch point according to the described quantity of described touch point and the area of each described touch point.
When proofreading and correct the quantity of described touch point, also proofread and correct in conjunction with described average area according to the preset range under the described length breadth ratio.
Described mode of proofreading and correct the quantity of described touch point according to the preset range under the described length breadth ratio specifically comprises: when described aspect ratio bit during smaller or equal to described average area, judges that described touch point is a touch point in the area of first preset range or described touch point; When described aspect ratio bit during greater than average area, judges that described touch point is the phase adhesion of two or more touch points in the area of second preset range and described touch point.
After the step of the quantity of the described touch point of described correction, also comprise:, proofread and correct the described positional information of described each touch point according to the center of proofreading and correct each described touch point of determining the back.
Preferably, when definite described touch point is mutually bonding by three little touch points, can determine each described little touch point position information according to the coordinate of the pixel of the center of each described little touch point.
Fig. 2 is the process flow diagram of detection method one embodiment of the touch point of a kind of image recognition of the present invention.
S201 judges respectively whether each pixel satisfies threshold condition in the described gray level image, if do not satisfy threshold condition, is 0 with this pixel assignment; If satisfy threshold condition, continue to judge whether this pixel satisfies the homogeneity condition, if satisfy the homogeneity condition, be 1 with this pixel assignment, if do not satisfy the homogeneity condition, be 0 with this pixel assignment.
Judge that the step whether pixel satisfies the homogeneity condition specifically comprises:
The mean value S of the pixel value of all pixels in the sub-piece is detected at the calculating pixel place
Mean, the sub-piece of described detection is by this pixel and the picture element matrix that is listed as with the capable m of n that the contiguous pixel of this pixel is formed.
Calculate the similarity of the sub-piece of described detection, when the similarity of the sub-piece of described detection equals n * m, judge that this pixel satisfies the homogeneity condition, otherwise judge that this pixel does not satisfy the homogeneity condition, wherein, the similarity that detects sub-piece is defined as: the pixel value I that has N pixel in detecting sub-piece
iSatisfy "
" time, the similarity that then detects sub-piece is N, wherein R is a preset value.
Described threshold value is the interior max pixel value of described gray level image and the mean value of minimum pixel value.
As previously mentioned, be that bianry image can be changed by the method for existing fixed threshold value with greyscale image transitions.And the technology of the present invention provides a kind of method of preferred bianry image conversion: described threshold value is the interior max pixel value of described gray level image and the mean value of minimum pixel value.
Need to prove that it is the object pixel for the treatment of assignment that each pixel of described gray level image all should be used as, each pixel is carried out independent two-value conversion.The pixel ability assignment that satisfies threshold condition and homogeneity condition simultaneously is 1, otherwise assignment is 0.
The effect of described threshold condition is that ambiguous spots is removed.Maximum and/or minimum pixel value according to described pixel in the described gray level image are carried out dynamic threshold, and non-stationary threshold values.Threshold value
In, Imax is the max pixel value of described gray level image, Imin is the minimum pixel value of described gray level image.For the pixel value less than threshold value T, its corresponding point is unsharp spots or background, is 0 with this assignment.For the pixel value more than or equal to threshold value T, its value should keep, and gives over to the judgement of carrying out the homogeneity condition.
Preferably, the sub-piece of described detection by this pixel and with the picture element matrix that the capable m of n that the contiguous pixel of this pixel is formed is listed as, get 3 * 3 the sub-piece of detection.
Need to prove, each described pixel pixel adjacent thereto is formed a sub-piece of detection in the described gray level image, the sub-piece of described detection can be made up of 3 * 3 pixels, also can be by being that 4 * 4 pixels are formed ... so analogize, but comprise a pixel to be detected in the sub-piece of described detection, be referred to as object pixel.Each pixel in the described gray level image all can be used as object pixel, can detect sub-piece by the difference that comprises described object pixel described object pixel is carried out the two-value conversion.
Fig. 3 is the sub-piece synoptic diagram of detection of detection method of the touch point of a kind of image recognition of the present invention, and wherein pixel 301 is to treat the object pixel of assignment.
Pixel 301 pixels adjacent thereto are formed one 3 * 3 the sub-piece of detection as shown in Figure 3, and the pixel value of pixel 301 is 194 pixel, specifies calculation step:
(a) try to achieve the mean value of the pixel value of 3 * 3 pixels in the sub-piece of described detection
, can get S according to the calculated for pixel values of nine pixels of Fig. 3 (a)
Mean=195;
(b) to described nine pixel values and S
MeanMake comparisons, can get Fig. 3 (b), be i.e. the pixel value I of each pixel
iSatisfy
R is the empirical value that is obtained by test, preferably, gets R=10.Owing to detect the pixel value I that has 9 pixels in the sub-piece
iSatisfy "
", be 9 so this detects sub-piece similarity.
(c) the homogeneity condition is satisfied in similarity 9=3 * 3 of the sub-piece of detection of Fig. 3, treats that to described pixel 301 assignment of assignment are 1, shown in Fig. 3 (c).
Need to prove, satisfy the homogeneity condition after, from the visual observation of human eye, be exactly that pixel gray-scale value around this pixel and its all compares evenly.The pixel that satisfies this homogeneity condition has two kinds: the one, and background pixel, the 2nd, touch point pixel.Not satisfying the inhomogeneity pixel of this gray-scale value then is the boundary pixel of touch point and background.Satisfying under the threshold condition, fuzzy spots and background pixel all are assigned 0.With the gradation uniformity condition judgment of sub-piece, just can well cut apart touch point and screen background again, and can also well remove noise in the gray level image.So the homogeneity condition of sub-piece just has a function, the very near touch point of touch location distance is separated.
S202 is a bianry image with greyscale image transitions.
It is multiple to become the method for bianry image to have greyscale image transitions, and existing technology is the method for fixed threshold conversion, gray-scale value during greater than a certain fixed threshold assignment be 1, otherwise assignment is 0.Can adopt technical solution of the present invention dynamic threshold method, described dynamic threshold method does not repeat them here as described in the step S201 yet.
S203 carries out morphology to described bianry image to be handled, and obtains preferred bianry image, and described morphology processing is meant carries out closed operation earlier to described bianry image, carries out opening operation again.
Can some isolated points occur on every side in the touch point through the bianry image after step S201 and/or the step S202 processing, at this moment, we carry out morphology to described bianry image again and handle.It is the current techique of Flame Image Process that morphology is handled, and herein, bianry image is carried out earlier carrying out opening operation again after the closed operation, just can well remove the isolated point because of the above-mentioned steps generation, and the level and smooth edge in zone, touch point.So just eliminated the influence of the isolated point after the binaryzation to the final touch point statistics.
S204 obtains quantity and/or each described touch point position information of touch point in the described bianry image, and described positional information is the coordinate of pixel that is positioned at the center of touch point.
Need to prove, in above-mentioned bianry image, having only touch point location ability assignment is 1, so can be that 1 zone obtains the information of corresponding described touch point by detecting assignment, the information of described touch point can comprise quantity and/or each described touch point position information of touch point.The quantity of described touch point is that assignment is the number in 1 zone, and described touch point position information is that assignment is the coordinate of pixel of the center in 1 zone.
S205 calculates the length breadth ratio of the boundary rectangle of each described touch point minimum.
Need to prove, be 1 zone by above-mentioned assignment, can obtain its minimum boundary rectangle.Described minimum boundary rectangle is meant and comprises the boundary rectangle that assignment is 1 simply connected region minimum.By described rectangle, can try to achieve the length of described rectangle and wide, wherein, the longer sides of definition rectangle is the length of rectangle, the definition rectangle be the wide of rectangle than minor face, obtain length and width ratio.
S206 proofreaies and correct the quantity of described touch point according to the preset range under the described length breadth ratio.
Need to prove that according to the preset range under the described length breadth ratio of each touch point, we can determine that described touch point is that a touch point or several touch point are mutually bonding.
In obtaining described bianry image, after the step of the quantity of touch point and/or each described touch point position information, also comprise:, calculate the average area of described touch point according to the described quantity of described touch point and the area of each described touch point.
When proofreading and correct the quantity of described touch point, also proofread and correct in conjunction with described average area according to the preset range under the described length breadth ratio.
Described mode of proofreading and correct the quantity of described touch point according to the preset range under the described length breadth ratio specifically comprises: when described aspect ratio bit during smaller or equal to described average area, judges that described touch point is a touch point in the area of first preset range or described touch point; When described aspect ratio bit during greater than average area, judges that described touch point is the phase adhesion of two or more touch points in the area of second preset range and described touch point.
After the step of the quantity of the described touch point of described correction, also comprise:, proofread and correct the described positional information of described each touch point according to the center of proofreading and correct each described touch point of determining the back.
Fig. 4 is the structural representation of pick-up unit of the touch point of a kind of image recognition of the present invention, comprising:
The two-value converting unit, being used for greyscale image transitions is bianry image.
With the acquiring unit that described two-value converting unit links to each other, be used for obtaining quantity and/or each described touch point position information of the described bianry image touch point of described two-value converting unit conversion.
With the computing unit that described acquiring unit links to each other, be used to calculate the length breadth ratio of boundary rectangle of the area minimum of each described touch point.
With the correcting unit that described computing unit links to each other, the preset range under the described length breadth ratio that is used for calculating according to described computing unit is proofreaied and correct the quantity of described touch point.
Fig. 5 is the structural representation of an embodiment of pick-up unit of the touch point of a kind of image recognition of the present invention, comprises above-mentioned two-value converting unit, acquiring unit, computing unit and correcting unit.
Described computing unit also comprises the area computing module, is used for calculating the average area of described touch point according to the described quantity of described touch point and the area of each described touch point.
When described correcting unit is proofreaied and correct the described touch point position information of the quantity of described touch point and/or each in the length breadth ratio of calculating according to described computing unit, also proofread and correct in conjunction with described area computing module.
Described correcting unit also comprises the position correction module, is used for proofreading and correct the described positional information of described each touch point according to the center of proofreading and correct each described touch point of determining the back.
Described two-value converting unit also comprises one-level identifying unit, secondary identifying unit and assignment unit.
The one-level identifying unit is used for each pixel in the described touch point is satisfied the judgement of threshold condition.
With the secondary identifying unit that described one-level identifying unit links to each other, be used for the described pixel of assignment for the treatment of being satisfied the judgement of homogeneity condition to behind the described threshold condition that satisfies described one-level identifying unit.
With the assignment unit that described secondary identifying unit links to each other, be used for described homogeneity condition to the meeting tier 2 identifying unit after, be 1 to described pixel assignment; The described pixel assignment that does not satisfy described threshold condition or homogeneity condition is 0.
Described one-level identifying unit comprises the threshold setting module, is used to set described threshold value, is the max pixel value in the described gray level image and the mean value of minimum pixel value with described threshold setting.
As shown in Figure 5, the pick-up unit of the touch point of described a kind of image recognition also comprises the optimization process unit, the optimization process unit links to each other with described two-value converting unit, being used for that described bianry image is carried out morphology handles, obtain preferred bianry image, described morphology processing is meant carries out closed operation earlier to described bianry image, carries out opening operation again.
Through the processing of said units, can obtain bianry image clearly, and the information of touch point accurately.Fig. 6 is the original optical imagery that prior art is obtained.Fig. 7 handles the preferred bianry image that the back obtains through the present invention.By Fig. 6 and Fig. 7 contrast, the brown free time spot on Fig. 6 left side is removed in Fig. 7.In addition, separated a bonding touch point among Fig. 6, repaired the border of touch point, made five touch points high-visible in Fig. 7.
Above disclosed is a kind of preferred embodiment of the present invention only, can not limit the present invention's interest field certainly with this, and therefore the equivalent variations of doing according to claim of the present invention still belongs to the scope that the present invention is contained.
Claims (10)
1. the detection method of the touch point of an image recognition is characterized in that, comprising:
With greyscale image transitions is bianry image;
Obtain quantity and/or each described touch point position information of touch point in the described bianry image, described positional information is the coordinate of pixel that is positioned at the center of touch point;
Calculate the length breadth ratio of boundary rectangle of the minimum of each described touch point;
Proofread and correct the quantity of described touch point according to the preset range under the described length breadth ratio.
2. the detection method of the touch point of image recognition according to claim 1 is characterized in that:
In obtaining described bianry image, after the step of the quantity of touch point and/or each described touch point position information, also comprise:, calculate the average area of described touch point according to the described quantity of described touch point and the area of each described touch point;
When proofreading and correct the quantity of described touch point, also proofread and correct in conjunction with described average area according to the preset range under the described length breadth ratio;
Described mode of proofreading and correct the quantity of described touch point according to the preset range under the described length breadth ratio specifically comprises: when described aspect ratio bit during smaller or equal to described average area, judges that described touch point is a touch point in the area of first preset range or described touch point; When described aspect ratio bit during greater than average area, judges that described touch point is the phase adhesion of two or more touch points in the area of second preset range and described touch point;
After the step of the quantity of the described touch point of described correction, also comprise:, proofread and correct the described positional information of described each touch point according to the center of proofreading and correct each described touch point of determining the back.
3. the detection method of the touch point of image recognition according to claim 1 and 2 is characterized in that, described is also to comprise after the step of bianry image with greyscale image transitions:
Described bianry image is carried out morphology handle, obtain preferred bianry image;
Described morphology processing is meant carries out closed operation earlier to described bianry image, carries out opening operation again.
4. the detection method of the touch point of image recognition according to claim 2 is characterized in that, described is that the step of bianry image specifically comprises with greyscale image transitions:
Judging respectively whether each pixel satisfies threshold condition in the described gray level image, if do not satisfy threshold condition, is 0 with this pixel assignment; If satisfy threshold condition, continue to judge whether this pixel satisfies the homogeneity condition, if satisfy the homogeneity condition, be 1 with this pixel assignment, if do not satisfy the homogeneity condition, be 0 with this pixel assignment;
When judging whether pixel satisfies threshold condition, if the pixel value of pixel is judged that then this pixel satisfies threshold condition, otherwise judged that this pixel does not satisfy threshold condition more than or equal to threshold value;
Judge that the step whether pixel satisfies the homogeneity condition specifically comprises,
The mean value S of the pixel value of all pixels in the sub-piece is detected at the calculating pixel place
Mean, the sub-piece of described detection is by this pixel and the picture element matrix that is listed as with the capable m of n that the contiguous pixel of this pixel is formed,
Calculate the similarity of the sub-piece of described detection, when the similarity of the sub-piece of described detection equals n * m, judge that this pixel satisfies the homogeneity condition, otherwise judge that this pixel does not satisfy the homogeneity condition, wherein, the similarity that detects sub-piece is defined as: the pixel value I that has N pixel in detecting sub-piece
iSatisfy "
" time, the similarity that then detects sub-piece is N, wherein R is a preset value.
5. the detection method of the touch point of image recognition according to claim 4 is characterized in that, described threshold value is the interior max pixel value of described gray level image and the mean value of minimum pixel value.
6. the pick-up unit of the touch point of an image recognition is characterized in that, comprising:
The two-value converting unit, being used for greyscale image transitions is bianry image;
With the acquiring unit that described two-value converting unit links to each other, be used for obtaining quantity and/or each described touch point position information of the described bianry image touch point of described two-value converting unit conversion;
With the computing unit that described acquiring unit links to each other, be used to calculate the length breadth ratio of the boundary rectangle of each described touch point minimum;
With the correcting unit that described computing unit links to each other, the preset range under the described length breadth ratio that is used for calculating according to described computing unit is proofreaied and correct the quantity of described touch point.
7. the pick-up unit of the touch point of image recognition according to claim 6 is characterized in that:
Described computing unit comprises the area computing module, is used for calculating the average area of described touch point according to the described quantity of described touch point and the area of each described touch point,
When described correcting unit is proofreaied and correct the quantity of described touch point in the length breadth ratio of calculating according to described computing unit, also proofread and correct in conjunction with described area computing module;
Described correcting unit comprises the position correction module, is used for proofreading and correct the described positional information of described each touch point according to the center of proofreading and correct each described touch point of determining the back.
8. according to the pick-up unit of the touch point of claim 6 or 7 described image recognitions, it is characterized in that, also comprise:
The optimization process unit that links to each other with described two-value converting unit, being used for that the described bianry image that described two-value converting unit is obtained is carried out morphology handles, obtain preferred bianry image, described morphology processing is meant carries out closed operation earlier to described bianry image, carries out opening operation again.
9. the pick-up unit of the touch point of image recognition according to claim 7 is characterized in that, described two-value converting unit comprises:
The one-level identifying unit is used for each pixel in the described touch point is satisfied the judgement of threshold condition;
With the secondary identifying unit that described one-level identifying unit links to each other, be used for the described pixel of assignment for the treatment of being satisfied the judgement of homogeneity condition to behind the described threshold condition that satisfies described one-level identifying unit;
With the assignment unit that described secondary identifying unit links to each other, be used for described homogeneity condition to the meeting tier 2 identifying unit after, be 1 to described pixel assignment; The described pixel assignment that does not satisfy described threshold condition or homogeneity condition is 0.
10. the pick-up unit of the touch point of image recognition according to claim 9 is characterized in that, described one-level identifying unit comprises:
The threshold setting module is used to set described threshold value, is the max pixel value in the described gray level image and the mean value of minimum pixel value with described threshold setting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110054191 CN102129332A (en) | 2011-03-07 | 2011-03-07 | Detection method and device of touch points for image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110054191 CN102129332A (en) | 2011-03-07 | 2011-03-07 | Detection method and device of touch points for image recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102129332A true CN102129332A (en) | 2011-07-20 |
Family
ID=44267435
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110054191 Pending CN102129332A (en) | 2011-03-07 | 2011-03-07 | Detection method and device of touch points for image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102129332A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103019596A (en) * | 2012-12-07 | 2013-04-03 | Tcl通讯（宁波）有限公司 | Method for realizing operation of virtual key based on touch screen and mobile terminal |
| CN103186329A (en) * | 2011-12-27 | 2013-07-03 | 富泰华工业（深圳）有限公司 | Electronic equipment and touch input control method thereof |
| CN103186290A (en) * | 2011-12-28 | 2013-07-03 | 原相科技股份有限公司 | Handwriting system and operating method thereof |
| CN104699327A (en) * | 2013-12-05 | 2015-06-10 | 原相科技股份有限公司 | Optical touch system and suspension judgment method thereof |
| US9519380B2 (en) | 2011-12-22 | 2016-12-13 | Pixart Imaging Inc. | Handwriting systems and operation methods thereof |
| CN112529786A (en) * | 2019-09-19 | 2021-03-19 | 佳能株式会社 | Image processing apparatus and method, and non-transitory computer-readable storage medium |
| CN114253417A (en) * | 2021-12-02 | 2022-03-29 | Tcl华星光电技术有限公司 | Multi-touch-point identification method and device, computer readable medium and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1738426A (en) * | 2005-09-09 | 2006-02-22 | 南京大学 | Video motion goal division and track method |
| CN101145091A (en) * | 2007-11-01 | 2008-03-19 | 复旦大学 | Touch panel based on infrared pick-up and its positioning detection method |
| CN101692282A (en) * | 2009-10-16 | 2010-04-07 | 浙江工业大学 | Morphology based method for separating cells |
| US20100097349A1 (en) * | 2008-10-17 | 2010-04-22 | Kim Jae-Shin | Method and apparatus for detecting touch point |
-
2011
- 2011-03-07 CN CN 201110054191 patent/CN102129332A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1738426A (en) * | 2005-09-09 | 2006-02-22 | 南京大学 | Video motion goal division and track method |
| CN101145091A (en) * | 2007-11-01 | 2008-03-19 | 复旦大学 | Touch panel based on infrared pick-up and its positioning detection method |
| US20100097349A1 (en) * | 2008-10-17 | 2010-04-22 | Kim Jae-Shin | Method and apparatus for detecting touch point |
| CN101692282A (en) * | 2009-10-16 | 2010-04-07 | 浙江工业大学 | Morphology based method for separating cells |
Non-Patent Citations (1)
| Title |
|---|
| 《计算机工程与应用》 20081231 马洋涛等 一种用于手写体汉字切分的优化模型 227-229 1-10 第44卷, 第2期 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9519380B2 (en) | 2011-12-22 | 2016-12-13 | Pixart Imaging Inc. | Handwriting systems and operation methods thereof |
| CN103186329A (en) * | 2011-12-27 | 2013-07-03 | 富泰华工业（深圳）有限公司 | Electronic equipment and touch input control method thereof |
| CN103186329B (en) * | 2011-12-27 | 2017-08-18 | 富泰华工业（深圳）有限公司 | Electronic equipment and its touch input control method |
| CN103186290A (en) * | 2011-12-28 | 2013-07-03 | 原相科技股份有限公司 | Handwriting system and operating method thereof |
| CN103019596A (en) * | 2012-12-07 | 2013-04-03 | Tcl通讯（宁波）有限公司 | Method for realizing operation of virtual key based on touch screen and mobile terminal |
| CN104699327A (en) * | 2013-12-05 | 2015-06-10 | 原相科技股份有限公司 | Optical touch system and suspension judgment method thereof |
| CN104699327B (en) * | 2013-12-05 | 2017-10-27 | 原相科技股份有限公司 | Optical touch control system and its suspension determination methods |
| CN112529786A (en) * | 2019-09-19 | 2021-03-19 | 佳能株式会社 | Image processing apparatus and method, and non-transitory computer-readable storage medium |
| CN114253417A (en) * | 2021-12-02 | 2022-03-29 | Tcl华星光电技术有限公司 | Multi-touch-point identification method and device, computer readable medium and electronic equipment |
| CN114253417B (en) * | 2021-12-02 | 2024-02-02 | Tcl华星光电技术有限公司 | Multi-touch point identification method and device, computer readable medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11650659B2 (en) | User input processing with eye tracking | |
| JP7287959B2 (en) | SYSTEM AND METHOD FOR BEHAVIORAL AUTHENTICATION USING TOUCH SENSOR DEVICE | |
| CN102129332A (en) | Detection method and device of touch points for image recognition | |
| KR101097309B1 (en) | Method and apparatus for recognizing touch operation | |
| CN104641316B (en) | Cursor movement devices | |
| US20150309663A1 (en) | Flexible air and surface multi-touch detection in mobile platform | |
| JP7454743B2 (en) | Barcode scanning based on gesture detection and analysis | |
| JP5213831B2 (en) | Touch point detection method and apparatus | |
| US20200073509A1 (en) | Direction determination system and direction determination method | |
| US20130241888A1 (en) | Detecting Wave Gestures Near an Illuminated Surface | |
| KR20100072207A (en) | Detecting finger orientation on a touch-sensitive device | |
| CN103279225A (en) | Projection type man-machine interactive system and touch control identification method | |
| CN104991684A (en) | Touch control device and working method therefor | |
| KR102339839B1 (en) | Method and apparatus for processing gesture input | |
| US9389702B2 (en) | Input association | |
| CN106406684B (en) | Projection process method, apparatus and projector | |
| US9778796B2 (en) | Apparatus and method for sensing object, and method of identifying calibration pattern in object sensing apparatus | |
| US9430094B2 (en) | Optical touch system, method of touch detection, and computer program product | |
| CN102799344B (en) | Virtual touch screen system and method | |
| CN116301551A (en) | Touch identification method, touch identification device, electronic equipment and medium | |
| WO2018165148A1 (en) | Foreground object detection | |
| US9927523B2 (en) | Event filtering device and motion recognition device thereof | |
| CN109857316A (en) | The adjustment based on prediction at tangible interface | |
| Irri et al. | A study of ambient light-independent multi-touch acquisition and interaction methods for in-cell optical touchscreens | |
| KR100887093B1 (en) | Interface method for tabletop computing environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110720 |