CN104541230A - Method and device for performing video verification to prevent mobile phone maloperation - Google Patents

Abstract

The inveiton is suitable for the intelligent terminal field and provides a method for performing video verification to prevent mobile phone maloperation. The method comprises the following steps: obtaining photos of an operator, extracting a facial image of the operator, distinguishing whether the facial image is a human image according to characteristics of the facial image, and switching on intelligent equipment if the facial image is a human image. The technical solution provided by the invention is advantaged by high human identification accuracy and high safety.

Description

Video verification prevents method and the device of mobile phone misoperation

Technical field

The invention belongs to intelligent terminal field, particularly relate to method and device that a kind of video verification prevents mobile phone misoperation.

Background technology

Mobile terminal, be commonly called as mobile phone, existing mobile phone have passed through the simulation epoch, digital Age and intelligence epoch, existing mobile phone major part is smart mobile phone, its function is also more powerful, comprising: the function that online, video calling, app application etc. are numerous, and mobile phone also becomes the instrument of people's a kind of indispensability in daily life gradually.

Mobile phone, due to smaller and more exquisite, generally all can be placed in pocket, though mobile phone meeting screen locking, still there will be maloperation.

Summary of the invention

The object of the embodiment of the present invention is to provide a kind of mobile phone based on recognition of face to verify using method, and it solves the problem of the maloperation of prior art.

The embodiment of the present invention is achieved in that a kind of video verification prevents the method for mobile phone misoperation, and described method comprises the steps: on the one hand

Obtain the photo of operator;

The face-image of operator is extracted from photo;

Whether be character image by the feature resolution of this face-image;

Character image in this way, then open smart machine.

Optionally, whether the described feature resolution by this face-image is that character image is specially:

Extract face contour line and the eye mid point of face-image, determine whether as character graphic according to ratio,

Distance between ratio=eye mid point/eye mid point line mid point is to the vertical range of face contour.

Optionally, whether the described feature resolution by this face-image is that character image is specially:

Extract face contour line and the eye mid point of face-image, with eye mid point for initial point, 360 rays are gone out in 360 ° of uniform emissions, obtain 360 intersecting point coordinates between 360 rays and face contour line, 360 distances between datum point and 360 intersecting point coordinates, calculate the summation of 360 distances, if summation is in character facial threshold range, then confirm as facial image.

Optionally, whether be face according to 360 Distance Judgments, be specially:

$D_n=|{L_n}^2-{\stackrel{\‾}{L}}^2|$

X _n＝|D _n-D _n-1|；

Work as X _nwhen being less than setting threshold value, being defined as face, otherwise being non-face; Wherein, L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances.

Optionally, whether be face according to 360 Distance Judgments, be specially:

$\mathrm{mov}={\mathrm{\γ}}^{*}\frac{2{\stackrel{\‾}{L}}^2}{\sqrt{\frac{\underset{n=1}{\overset{360}{\mathrm{\Σ}}}{(L_n-\stackrel{\‾}{L})}^2}{n-1}}}$

Wherein, γ is experience factor, and mov is side-play amount, and n is the numbering of intersection point; L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances;

When the side-play amount calculated is greater than offset threshold, cumulative amount adds 1, when cumulative amount is more than 100, is defined as face, otherwise is non-face.

On the other hand, provide a kind of intelligent end device, described device comprises:

Acquiring unit, for obtaining the photo of operator;

Extraction unit, for extracting the face-image of operator from photo;

Whether resolution element is character image for the feature resolution by this face-image;

Opening unit, when being character image for telling at resolution element, then opening smart machine.

Optionally, described resolution element specifically comprises:

Middle point module, for extracting face contour line and the eye mid point of face-image;

Judge module, for determining whether as character graphic according to ratio,

Distance between ratio=eye mid point/eye mid point line mid point is to the vertical range of face contour.

Optionally, described resolution element is specifically for extracting face contour line and the eye mid point of face-image, with eye mid point for initial point, 360 rays are gone out in 360 ° of uniform emissions, obtain 360 intersecting point coordinates between 360 rays and face contour line, 360 distances between datum point and 360 intersecting point coordinates, calculate the summation of 360 distances, if summation is in character facial threshold range, then confirm as facial image.

Optionally, whether described resolution element also for being face according to 360 Distance Judgments, is specially:

$D_n=|{L_n}^2-{\stackrel{\‾}{L}}^2|$

X _n＝|D _n-D _n-1|；

Work as X _nwhen being less than setting threshold value, being defined as face, otherwise being non-face; Wherein, L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances.

Optionally, whether described resolution element also for being face according to 360 Distance Judgments, is specially:

$\mathrm{mov}={\mathrm{\γ}}^{*}\frac{2{\stackrel{\‾}{L}}^2}{\sqrt{\frac{\underset{n=1}{\overset{360}{\mathrm{\Σ}}}{(L_n-\stackrel{\‾}{L})}^2}{n-1}}}$

Wherein, γ is experience factor, and mov is side-play amount, and n is the numbering of intersection point; L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances;

When the side-play amount calculated is greater than offset threshold, cumulative amount adds 1, when cumulative amount is more than 100, is defined as face, otherwise is non-face.

In embodiments of the present invention, the mobile phone of technical scheme provided by the invention adopts the mode of recognition of face to monitor checking mobile phone, so it is when cellphone subscriber changes, can the operation of the non-owner of automatic shield, thus its to have security a little high.

Accompanying drawing explanation

Fig. 1 is the process flow diagram that a kind of video verification provided by the invention prevents the method for mobile phone misoperation;

Fig. 2 is the structural drawing of a kind of intelligent end device provided by the invention.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

The specific embodiment of the invention provides a kind of video verification to prevent the method for mobile phone misoperation, and said method is performed by mobile phone or other smart machine (such as ipad, PDA etc.), and the method as shown in Figure 1, comprises the steps:

101, the photo of operator is obtained;

102, from photo, extract the face-image of operator;

103, whether be character image by the feature resolution of this face-image;

104, character image in this way, then open smart machine.

Whether method provided by the invention is obtained by the method for comparison chart picture is character image, if i.e. non-character image, such as namely the image such as dog, cat does not open equipment.

The implementation method of above-mentioned 103 is specifically as follows:

Extract face contour line and the eye mid point of face-image, determine whether as character graphic according to ratio.This ratio is specifically as follows:

Distance between ratio=eye mid point/eye mid point line mid point is to the vertical range of face contour.

Found by data research, dissimilar animal, the ratio of its aforementioned proportion is different, and it is in certain scope, just can very easily tell the image belonging to character image or animal by the ratio of this ratio.

The implementation method of above-mentioned 103 is specifically as follows:

Extract face contour line and the eye mid point of face-image, with eye mid point for initial point, 360 rays are gone out in 360 ° of uniform emissions, obtain 360 intersecting point coordinates between 360 rays and face contour line, 360 distances between datum point and 360 intersecting point coordinates, calculate the summation of 360 distances, if summation is in character facial threshold range, then confirm as facial image.

Found by research, existing face is all larger than animal, the value of its distance obtained of sampling is general all larger like this, so can maloperation be avoided, this kind can not only distinguish humans and animals, child's maloperation can also be avoided because teenage time, the face of child also cannot reach above-mentioned requirements.

The implementation method of above-mentioned 103 is specifically as follows:

$D_n=|{L_n}^2-{\stackrel{\‾}{L}}^2|$

X _n＝|D _n-D _n-1|；

Work as X _nwhen being less than setting threshold value, being defined as face, otherwise being non-face; Wherein, L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances.

The implementation method of above-mentioned 103 is specifically as follows:

$\mathrm{mov}={\mathrm{\γ}}^{*}\frac{2{\stackrel{\‾}{L}}^2}{\sqrt{\frac{\underset{n=1}{\overset{360}{\mathrm{\Σ}}}{(L_n-\stackrel{\‾}{L})}^2}{n-1}}}$

Wherein, γ is experience factor, is specifically as follows 0.15.When the side-play amount calculated (mov) is greater than offset threshold, cumulative amount adds 1, when cumulative amount is more than 100, is defined as face, otherwise is non-face.

The specific embodiment of the invention also provides a kind of intelligent end device, and this device as shown in Figure 2, comprising:

Acquiring unit 21, for obtaining the photo of operator;

Extraction unit 22, for extracting the face-image of operator from photo;

Whether resolution element 23 is character image for the feature resolution by this face-image;

Opening unit 24, when being character image for telling at resolution element, then opening smart machine.

Optionally, resolution element 23 specifically comprises:

Middle point module 231, for extracting face contour line and the eye mid point of face-image;

Judge module 232, for determining whether as character graphic according to ratio,

Distance between ratio=eye mid point/eye mid point line mid point is to the vertical range of face contour.

Optionally, resolution element 23 is specifically for extracting face contour line and the eye mid point of face-image, with eye mid point for initial point, 360 rays are gone out in 360 ° of uniform emissions, obtain 360 intersecting point coordinates between 360 rays and face contour line, 360 distances between datum point and 360 intersecting point coordinates, calculate the summation of 360 distances, if summation is in character facial threshold range, then confirm as facial image.

Found by research, existing face is all larger than animal, the value of its distance obtained of sampling is general all larger like this, so can maloperation be avoided, this kind can not only distinguish humans and animals, child's maloperation can also be avoided because teenage time, the face of child also cannot reach above-mentioned requirements.

Optionally, whether resolution element 23 also for being face according to 360 Distance Judgments, is specially:

$D_n=|{L_n}^2-{\stackrel{\‾}{L}}^2|$

X _n＝|D _n-D _n-1|；

Work as X _nwhen being less than setting threshold value, being defined as face, otherwise being non-face; Wherein, L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances.

Optionally, whether resolution element 23 also for being face according to 360 Distance Judgments, is specially:

$\mathrm{mov}={\mathrm{\γ}}^{*}\frac{2{\stackrel{\‾}{L}}^2}{\sqrt{\frac{\underset{n=1}{\overset{360}{\mathrm{\Σ}}}{(L_n-\stackrel{\‾}{L})}^2}{n-1}}}$

Wherein, γ is experience factor, is specifically as follows 0.15.When the side-play amount calculated (mov) is greater than offset threshold, cumulative amount adds 1, when cumulative amount is more than 100, is defined as face, otherwise is non-face.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1. video verification prevents a method for mobile phone misoperation, it is characterized in that, described method comprises the steps:

Obtain the photo of operator;

The face-image of operator is extracted from photo;

Whether be character image by the feature resolution of this face-image;

Character image in this way, then open smart machine.

2. method according to claim 1, is characterized in that, whether the described feature resolution by this face-image is that character image is specially:

Extract face contour line and the eye mid point of face-image, determine whether as character graphic according to ratio,

Distance between ratio=eye mid point/eye mid point line mid point is to the vertical range of face contour.

3. method according to claim 1, is characterized in that, whether the described feature resolution by this face-image is that character image is specially:

Extract face contour line and the eye mid point of face-image, with eye mid point for initial point, 360 rays are gone out in 360 ° of uniform emissions, obtain 360 intersecting point coordinates between 360 rays and face contour line, 360 distances between datum point and 360 intersecting point coordinates, calculate the summation of 360 distances, if summation is in character facial threshold range, then confirm as facial image.

4. whether method according to claim 3, is characterized in that, be face, be specially according to 360 Distance Judgments:

$D_n=|{L_n}^2-{\stackrel{\‾}{L}}^2|$

X _n＝|D _n-D _n-1|；

Work as X _nwhen being less than setting threshold value, being defined as face, otherwise being non-face; Wherein, L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances.

5. whether method according to claim 3, is characterized in that, be face, be specially according to 360 Distance Judgments:

$\mathrm{mov}=\mathrm{\γ}*\frac{2{\stackrel{\‾}{L}}^2}{\sqrt{\frac{\underset{n=1}{\overset{360}{\mathrm{\Σ}}}{(L_n-\stackrel{\‾}{L})}^2}{n-1}}}$

Wherein, γ is experience factor, and mov is side-play amount, and n is the numbering of intersection point; L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances;

When the side-play amount calculated is greater than offset threshold, cumulative amount adds 1, when cumulative amount is more than 100, is defined as face, otherwise is non-face.

6. an intelligent end device, is characterized in that, described device comprises:

Acquiring unit, for obtaining the photo of operator;

Extraction unit, for extracting the face-image of operator from photo;

Whether resolution element is character image for the feature resolution by this face-image;

Opening unit, when being character image for telling at resolution element, then opening smart machine.

7. device according to claim 6, is characterized in that, described resolution element specifically comprises:

Middle point module, for extracting face contour line and the eye mid point of face-image;

Judge module, for determining whether as character graphic according to ratio,

Distance between ratio=eye mid point/eye mid point line mid point is to the vertical range of face contour.

8. device according to claim 6, it is characterized in that, described resolution element is specifically for extracting face contour line and the eye mid point of face-image, with eye mid point for initial point, go out 360 rays in 360 ° of uniform emissions, obtain 360 intersecting point coordinates between 360 rays and face contour line, 360 distances between datum point and 360 intersecting point coordinates, calculate the summation of 360 distances, if summation is in character facial threshold range, then confirm as facial image.

9. device according to claim 8, is characterized in that, whether described resolution element also for being face according to 360 Distance Judgments, is specially:

$D_n=|{L_n}^2-{\stackrel{\‾}{L}}^2|$

X _n＝|D _n-D _n-1|；

Work as X _nwhen being less than setting threshold value, being defined as face, otherwise being non-face; Wherein, L _nfor the distance between initial point and the n-th intersecting point coordinate, it is the average of 360 distances.

10. device according to claim 8, is characterized in that, whether described resolution element also for being face according to 360 Distance Judgments, is specially:

$\mathrm{mov}=\mathrm{\γ}*\frac{2{\stackrel{\‾}{L}}^2}{\sqrt{\frac{\underset{n=1}{\overset{360}{\mathrm{\Σ}}}{(L_n-\stackrel{\‾}{L})}^2}{n-1}}}$

Wherein, γ is experience factor, and mov is side-play amount, and n is the numbering of intersection point; L _nfor the distance between initial point and the n-th intersecting point coordinate, L is the average of 360 distances;

When the side-play amount calculated is greater than offset threshold, cumulative amount adds 1, when cumulative amount is more than 100, is defined as face, otherwise is non-face.