Summary of the invention
The portrait that the object of this invention is to provide easy operating, is easy to differentiate is compared the application technology on mobile terminal operating system.
Technical scheme provided by the invention is:
The application technology of a kind of portrait comparison on mobile terminal operating system, comprise the following steps: step 1:: the instant head portrait photo that obtains target person by the camera of mobile terminal, when the distance of pupil center's point of two of the described instant head portrait photo shooting vertical angle that is the head of 30~100 pixels and described camera and target person is 0~20 °, being considered as this instant head portrait photo is satisfactory instant head portrait photo; It should be noted that, it is 0~20 ° that here 0~20 ° is not only forward, is oppositely also 0~20 °, around vertical line angle, is the angular range of 40 °.Preferably, above-mentioned shooting vertical angle is 0~15 °, and the distance of pupil center's point of two is 55~90 pixels; Step 2: extract the face characteristic cluster in above-mentioned satisfactory instant head portrait photo; Step 3: the face characteristic cluster in the existing head portrait photo of extraction target person; Step 4: the face characteristic cluster in the face characteristic cluster in instant head portrait photo and existing head portrait photo is compared, calculate similarity, and whether the described similarity of judgement meet default first threshold, if so, extract and show the identity information that described target person is corresponding.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, in described step 1, also comprise: sub-step 11: by camera testing environment light intensity, when described ambient light intensity is less than default Second Threshold, by the LED lamp light filling carrying on mobile terminal.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, in described step 1, also specifically comprise following sub-step: sub-step 12: the shooting vertical angle of measuring the head of described camera and target person by the gravity sensing device on mobile terminal, described shooting vertical angle data are presented on the display screen of mobile terminal, and whether the described shooting vertical angle data of judgement are 0~20 °, if, carry out sub-step 13, if not, continue to adjust described shooting vertical angle; Sub-step 13: the predeterminated position on the display screen of mobile terminal shows default unit picture element scale, and whether be 30~100 pixels according to the distance of pupil center's point of two of the described instant head portrait photo of described unit picture element scale judgement, if, carry out sub-step 14, if not, continue to adjust the distance of described pupil center's point of two; Sub-step 14: the instant head portrait photo of the camera photographic subjects personage by mobile terminal, and the instant head portrait photo of taking is cut out, remove unnecessary background.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, in described step 1, the shooting level angle of described camera and the head of target person is 0~20 °; Preferably, above-mentioned shooting level angle is 0~15 °, in described step 1, before sub-step 12, also comprise following sub-step: sub-step 15: the shooting level angle of measuring the head of described camera and target person by the gravity sensing device on mobile terminal, and described shooting level angle-data is presented on the display screen of mobile terminal, and whether the described shooting level angle-data of judgement is 0~20 °; If so, carry out sub-step 12, if not, continue to adjust described shooting level angle.It should be noted that, it is 0~20 ° that here 0~20 ° is not only forward, is oppositely also 0~20 °, around horizontal line angle, is the angular range of 40 °.In embodiment, further preferable range is identical therewith.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, in described sub-step 14, in instant head portrait photo after cutting out, people's face of described target person is long-pending be no less than the instant head portrait photo after described cutting out area 3/4.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, described step 2 specifically comprises following sub-step: sub-step 21: people's face location, and human face region is separated from the background of described instant head portrait photo; Sub-step 22: eyes location, the position of definite pupil of both eyes in the human face region obtaining from sub-step 21; Sub-step 23: adjustment of image, proofread and correct the angle of inclination of human face region, and according to the distance at the pupil of both eyes center described in the location positioning of determined pupil of both eyes in sub-step 22, and according to the ratio of the distance at the distance at pupil of both eyes center and default pupil of both eyes center, adjust the pixel of human face region; Sub-step 24: image pre-service, corrected human face region in sub-step 23 is carried out to pre-service, the exposure of color, face, face's light that specifically comprises face evenly, the processing of face's average color, the Gao Guang of face, blur level, brightness average, gray scale dynamic range, unevenness, overexposure ratio, under-exposure ratio, image sharpness, image blur; Sub-step 25: extract face characteristic cluster, the human face region obtaining is extracted to the face characteristic value that is no less than through sub-step 24, the summation of the face characteristic value of described instant head portrait photo is the face characteristic cluster of instant head portrait photo.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, described step 3 specifically comprises following sub-step: sub-step 31: obtain existing head portrait photo, by the card reader of ID card being connected with mobile terminal, obtain existing head portrait photo; Sub-step 32: people's face location, human face region is separated from the background of described existing head portrait photo; Sub-step 33: eyes location, the position of definite pupil of both eyes in the human face region obtaining from sub-step 32; Sub-step 34: adjustment of image, according to the distance at the pupil of both eyes center described in the location positioning of determined pupil of both eyes in sub-step 33, and according to the ratio of the distance at the distance at pupil of both eyes center and default pupil of both eyes center, adjust the pixel of human face region; According to determine the angle of inclination of human face region according to the line of the position of determined pupil of both eyes in sub-step 33, according to described angle of inclination, described human face region is proofreaied and correct to horizontal level; Sub-step 35: image pre-service, corrected human face region in sub-step 34 is carried out to pre-service, the exposure of color, face, face's light that specifically comprises face evenly, the processing of face's average color, the Gao Guang of face, blur level, brightness average, gray scale dynamic range, unevenness, overexposure ratio, under-exposure ratio, image sharpness, image blur; Sub-step 36: extract face characteristic cluster, the human face region obtaining is extracted to the face characteristic value that is no less than through sub-step 35, the summation of the face characteristic value of described existing head portrait photo is the face characteristic cluster of existing head portrait photo.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, described step 4 specifically comprises following sub-step: sub-step 41: comparison face characteristic value, face characteristic value corresponding in the face characteristic cluster of the face characteristic value in the face characteristic cluster of instant head portrait photo and existing head portrait photo is compared, judge whether the difference of corresponding face characteristic value is less than the 3rd default threshold value; Sub-step 42: add up satisfactory face characteristic value quantity, statistics is less than the quantity of the face characteristic value of the 3rd default threshold value; Sub-step 43: calculate similarity, the quantity that calculating is less than the face characteristic value of the 3rd default threshold value accounts for the number percent of everyone face eigenwert, described number percent is similarity, whether the described similarity of judgement is greater than default first threshold, if, extract and export the identity information that described target person is corresponding, if not, failed information is compared in output.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, described step 1 comprises following sub-step: sub-step 110: by camera testing environment light intensity, when described ambient light intensity is less than default Second Threshold, by the LED lamp light filling carrying on mobile terminal; Sub-step 111: the head portrait photo that obtains target person by the camera of mobile terminal; And by the upper gravity sensing device of mobile terminal, measure the shooting vertical angle of the head of camera and target person while taking; Whether the described shooting vertical angle of judgement is less than 20 °, it should be noted that, taking vertical angle, to be less than the meaning of 20 ° be that the angle that the angle with the vertical line of vertical plane is positive negative direction is 20 °, and its angular range is 40 ° altogether.If so, carry out sub-step 112, if not, point out the information of Image Acquisition failure; Sub-step 112: people's face location, human face region is separated from the background of the instant head portrait photo described in sub-step 111; Sub-step 113: eyes location, the position of definite pupil of both eyes in the human face region obtaining from sub-step 112; Sub-step 114: the distance of calculating the center of pupil of both eyes, according to the distance at the center of the position calculation pupil of both eyes of sub-step 113 determined pupil of both eyes, and whether the distance at the center of the pupil of both eyes described in judgement drops in 30~100 pixel coverages, if, carry out step 2, if not, output image obtains the information of failure; Described step 2 comprises following sub-step: sub-step 211: adjustment of image, proofread and correct the angle of inclination of human face region, and according to the ratio of the distance at the distance at resulting pupil of both eyes center in sub-step 114 and default pupil of both eyes center, adjust the pixel of human face region; Sub-step 212: image pre-service, corrected human face region in sub-step 212 is carried out to pre-service, the exposure of color, face, face's light that specifically comprises face evenly, the processing of face's average color, the Gao Guang of face, blur level, brightness average, gray scale dynamic range, unevenness, overexposure ratio, under-exposure ratio, image sharpness, image blur; Sub-step 213: extract face characteristic cluster, the human face region obtaining is extracted to the face characteristic value that is no less than through sub-step 212, the summation of the face characteristic value of described instant head portrait photo is the face characteristic cluster of head portrait photo.
In application technology in above-mentioned portrait comparison on mobile terminal operating system, if similarity does not meet default first threshold in described step 4, also comprise step 5 and step 6: step 5: by obtaining face characteristic cluster in step 2, send in the server of peripheral hardware, the server stores of described peripheral hardware has multiple human face photos, the face characteristic cluster that multiple human face photos are corresponding and corresponding people information, the server of described peripheral hardware detects for the face characteristic cluster sending according to mobile terminal the corresponding human face photo of similarity that meets the first predetermined threshold value, and by the described corresponding human face photo of the similarity that meets the first predetermined threshold value, the people information that this human face photo is corresponding sends in mobile terminal, step 6: human face photo and people information corresponding to this human face photo that the server of reception peripheral hardware sends, and be presented on the display screen of mobile terminal.
The present invention is by setting specific interpupillary distance, shooting angle and taking light, obtain satisfactory instant head portrait photo, and obtain fast result by portrait comparison technology, by above-mentioned processing, can effectively reduce data processing amount, improve comparison accuracy, the used time of reducing comparison, be specially adapted to mini mobile terminal.
Further, in order to eliminate the impact of manual operation on above-mentioned setting data, this programme adopts gravity sensing equipment and setting unit pixel scale assisted user to obtain qualified photo, has improved the success ratio that photo obtains.
The present invention can be for Android system, the IOS system of IPHONE, and the WP of Microsoft,, linux system, pervasive in various mobile terminal devices.
Embodiment
Below in conjunction with embodiment, technical scheme of the present invention is described in further detail, but does not form any limitation of the invention.
One of core idea of the present invention is, by setting specific interpupillary distance, shooting angle and taking light, obtain satisfactory photo, and obtain fast result by portrait comparison technology, by above-mentioned processing, can effectively reduce data processing amount, improve comparison accuracy, reduce the used time of comparison, be specially adapted to mini mobile terminal.
Specific embodiment 1:
The flow diagram of the application technology of a kind of portrait comparison as shown in 1 figure on mobile terminal operating system, specifically comprises the following steps:
Step 1: the instant head portrait photo that obtains target person by the camera of mobile terminal, when the distance of pupil center's point of two of the described instant head portrait photo shooting vertical angle that is the head of 55~90 pixels and described camera and target person is 0~15 °, being considered as this instant head portrait photo is satisfactory instant head portrait photo; As shown in Figure 2, the step that the described camera that passes through mobile terminal obtains the head portrait photo of target person can specifically comprise following sub-step:
Sub-step 11: by camera testing environment light intensity, when described ambient light intensity is less than default Second Threshold, by the LED lamp light filling carrying on mobile terminal.
Sub-step 15: judgement shooting level angle, by the gravity sensing device on mobile terminal, measure the shooting level angle of the head of described camera and target person, and described shooting level angle-data is presented on the display screen of mobile terminal, and whether the described shooting level angle-data of judgement is 0~15 °; If so, carry out sub-step 12, if not, continue to adjust described shooting level angle.
Sub-step 12: vertical angle is taken in judgement, by the gravity sensing device on mobile terminal, measure the shooting vertical angle of the head of described camera and target person, described shooting vertical angle data are presented on the display screen of mobile terminal, and whether the described shooting vertical angle data of judgement are 0~15 °, if, carry out sub-step 13, if not, continue to adjust described shooting vertical angle;
In actual applications, shooting level angle-data and take the position that the user of the screen that vertical angle data are presented at mobile terminal sets, for the ease of user, use, mobile terminal can arrange automatic prompting module, when shooting level angle-data with after taking in the angular range that vertical angle data fall into setting, automatically prompting module sound, the prompting of light or vibrations, user can, in order to avoid stare at above-mentioned data always, improve the convenience of taking.
Sub-step 13: the distance that judges pupil center's point of two, predeterminated position on the display screen of mobile terminal shows default unit picture element scale, and whether be 55~90 pixels according to the distance of pupil center's point of two of the described instant head portrait photo of described unit picture element scale judgement, if, carry out sub-step 14, if not, continue to adjust the distance of described pupil center's point of two;
In practice, unit picture element scale is set in the shown position of the desirable eyes of user on display screen, be generally upside 1/3 place of screen, for the ease of user, identify, unit picture element scale can be set as redness or other colors of easily differentiating, after photograph taking, unit picture element scale can't be stayed on photo, and unit picture element scale only plays a part dimensioning.The specification of unit picture element scale can be according to user's free setting, as is made as 60,70,80,90 pixels etc.
Sub-step 14: take and cut out instant head portrait photo, the instant head portrait photo of the camera photographic subjects personage by mobile terminal, and the instant head portrait photo of taking is cut out, remove unnecessary background.In instant head portrait photo after cutting out, people's face of described target person is long-pending be no less than the instant head portrait photo after described cutting out area 3/4.Concrete implementation method has many prior aries, in the present embodiment, in order to reduce treatment capacity, according to the distance of pupil center's points of locate by unit picture element scale two, determine the approximate location of people's face, the general general people's face scope of location positioning in face specifically according to mankind's eyes, then according to this people's face scope, determine which region is background, which region is portrait, it should be noted that, this method calculated amount is little, can significantly reduce the calculated amount of subsequent step.The method that can certainly adopt user to hand-make is carried out, and this present embodiment is not made too many restrictions.
Need to stress: step 15 in the present embodiment~13 are not fix unalterable, in this programme, sub-step 15,12,13 can sort by any way, then carries out one by one, finally carries out sub-step 14.
Step 2: extract the face characteristic collection in above-mentioned satisfactory instant head portrait photo; As shown in Figure 3, it specifically comprises following sub-step:
Sub-step 21: people's face location, human face region is separated from the background of described instant head portrait photo;
Sub-step 22: eyes location, the position of definite pupil of both eyes in the human face region obtaining from sub-step 21; Adopt the method for above-mentioned adaboost based on face characteristic can effectively determine the position of pupil of both eyes, in the prior art, several method has been developed in the location of human face characteristic point, at this, do not describe in detail one by one, above-mentioned concrete method is not restricted to the method for adaboost based on face characteristic.
In actual applications, because the computing power of existing mobile terminal is limited by the restriction of volume, chip development degree, there is a certain distance in its computing velocity and fixed computing machine, in order further to improve the arithmetic speed of this technology in mobile terminal, above-mentioned people's face location and eyes location adopt following method:
1) utilize to set threshold values automatically by human eye and people's face other parts and background separation or utilize human eye gray-scale value to carry out eyes location;
2) by gray level image being carried out after vertical and horizontal Gray Projection, tentatively fixed to people's face;
3), with the square frame seeker face of pupil size, when falling into black picture element number in frame and reaching maximum, the position of frame is eye position.
Sub-step 23: adjustment of image, proofread and correct the angle of inclination of human face region, and according to the distance at the pupil of both eyes center described in the location positioning of determined pupil of both eyes in sub-step 22, and according to the ratio of the distance at the distance at pupil of both eyes center and default pupil of both eyes center, adjust the pixel of human face region;
In concrete application, the correction at the angle of inclination of human face region comprises two aspects, be photographic plane region angle of inclination correction and perpendicular to the correction at the angle of inclination in photographic plane region, angle of inclination correction for photographic plane region both can have been known the angle that needs correction by the calculating at the line angle of inclination at two through hole centers, the pixel of human face region is mainly adjusted according to the ratio of the distance at the distance at pupil of both eyes center and default pupil of both eyes center, the distance of the pupil of both eyes obtaining when taking pictures is 63 pixels, and the distance at default pupil of both eyes center is 65 pixels, the picture processing module of mobile terminal suitably increases according to the ratio of above-mentioned pixel.The adjustment of above-mentioned angle and pixel is all fine setting, by less calculated amount, both can obtain the desired result of this sub-step 23.
Sub-step 24: image pre-service, corrected human face region in sub-step 23 is carried out to pre-service, the exposure of color, face, face's light that specifically comprises face evenly, the processing of face's average color, the Gao Guang of face, blur level, brightness average, gray scale dynamic range, unevenness, overexposure ratio, under-exposure ratio, image sharpness, image blur;
Sub-step 25: extract face characteristic cluster, the human face region obtaining is extracted to the face characteristic value that is no less than through sub-step 24, the summation of the face characteristic value of described instant head portrait photo is the face characteristic cluster of head portrait photo.After these pre-service of sub-step 24, a plurality of sample people faces of same user just can have certain similarity in some feature, and between different sample people face, just can have certain difference, now extract face characteristic value.Face characteristic value is also referred to as textural characteristics, human face characteristic point proper vector etc. in other documents and materials, existing multiple in prior art, and this is not made too many restrictions.
Step 3: the face characteristic cluster in the existing head portrait photo of extraction target person; As shown in Figure 4, step 3 specifically comprises following sub-step:
Sub-step 31: obtain existing head portrait photo, obtain existing head portrait photo by the card reader of ID card being connected with mobile terminal; Certainly, from I.D., obtaining existing head portrait photo is in the present embodiment a kind of concrete mode, in the storage of this locality of mobile phone, also can store some key persons' photo, such as for police, storage wanted criminal's photo, for supervisor, photo of storage examination hall examinee etc.
Sub-step 32: people's face location, human face region is separated from the background of described existing head portrait photo;
Sub-step 33: eyes location, the position of definite pupil of both eyes in the human face region obtaining from sub-step 32;
Sub-step 34: adjustment of image, according to the distance at the pupil of both eyes center described in the location positioning of determined pupil of both eyes in sub-step 33, and according to the ratio of the distance at the distance at pupil of both eyes center and default pupil of both eyes center, adjust the pixel of human face region; According to determine the angle of inclination of human face region according to the line of the position of determined pupil of both eyes in sub-step 33, according to described angle of inclination, described human face region is proofreaied and correct to horizontal level;
Sub-step 35: image pre-service, corrected human face region in sub-step 34 is carried out to pre-service, the exposure of color, face, face's light that specifically comprises face evenly, the processing of face's average color, the Gao Guang of face, blur level, brightness average, gray scale dynamic range, unevenness, overexposure ratio, under-exposure ratio, image sharpness, image blur;
Sub-step 36: extract face characteristic cluster, the human face region obtaining is extracted to the face characteristic value that is no less than through sub-step 35, the summation of the face characteristic value of described existing head portrait photo is the face characteristic cluster of existing head portrait photo.
Its disposal route is consistent with the method for extracting the face characteristic cluster in head portrait photo in step 2.
Step 4: the face characteristic cluster in the face characteristic cluster in head portrait photo and existing head portrait photo is compared, calculate similarity, and whether the described similarity of judgement meet default first threshold, if so, extract and show the identity information that described target person is corresponding.This first threshold can be according to the free selection of arranging of user, as a plurality of different threshold values such as 50%, 80%, 95% can be set, when high threshold mates when unsuccessful, user can reduce threshold value and improve and be matched to power, as shown in Figure 5, described step 4 specifically comprises following sub-step:
Sub-step 41: comparison face characteristic value, face characteristic value corresponding in the face characteristic cluster of the face characteristic value in the face characteristic cluster of instant head portrait photo and existing head portrait photo is compared, judge whether the difference of corresponding face characteristic value is less than the 3rd default threshold value;
Sub-step 42: add up satisfactory face characteristic value quantity, statistics is less than the quantity of the face characteristic value of the 3rd default threshold value;
Sub-step 43: calculate similarity, the quantity that calculating is less than the face characteristic value of the 3rd default threshold value accounts for the number percent of everyone face eigenwert, described number percent is similarity, whether the described similarity of judgement is greater than default first threshold, if, extract and export the identity information that described target person is corresponding, if not, failed information is compared in output.The data that the identity information that this target person is corresponding can read from card reader of ID card, obtain, also can from the server of the household register Information Statistics of public security organ, obtain by wireless network, also can from this locality storage of mobile terminal, obtain, this is not described in detail one by one.
The standard compliant picture that the present embodiment gathers by user, reduce the difficulty of picture analyzing, can improve significantly the success ratio of portrait comparison, the processing load that reduces processing time and mobile terminal, greatly facilitates the association area personnel's such as residence management personnel, the people's police on duty of public security organ, examination hall monitoring work.
Specific embodiment 2
This specific embodiment 2 is identical with embodiment 1, if when different places is that in sub-step 43, the described number percent of judgement is less than default first threshold, carry out step 5, specifically as shown in Figure 6:
Step 5: send in the server of peripheral hardware obtaining face characteristic cluster in step 2, the server stores of described peripheral hardware has face characteristic cluster and the corresponding people information that multiple human face photos, multiple human face photos are corresponding, the server of described peripheral hardware detects for the face characteristic cluster sending according to mobile terminal the corresponding human face photo of similarity that meets the first predetermined threshold value, and the described corresponding human face photo of the similarity that meets the first predetermined threshold value, the people information that this human face photo is corresponding are sent in mobile terminal;
In actual applications, send to the screening index that also has user to arrange of the server of peripheral hardware together with face characteristic cluster, this screening index comprises the index that sex, the range of age, height etc. can target-marking character features, first server filters out by screening index the person who meets, and then the comparison of pedestrian's face feature cluster, reduce data processing time, improved analysis efficiency.
Step 6: human face photo and people information corresponding to this human face photo that the server of reception peripheral hardware sends, and be presented on the display screen of mobile terminal.
In the present embodiment, adopt the server of peripheral hardware to improve the screening scope of portrait comparison, improved the accuracy of comparison.
Specific embodiment 3:
The flow diagram of the application technology of of the present invention another kind of portrait comparison as shown in Figure 6 on mobile terminal operating system, specifically comprises the following steps:
Step 1: the instant head portrait photo that obtains target person by the camera of mobile terminal, when the distance of pupil center's point of two of the described instant head portrait photo shooting vertical angle that is the head of 55~90 pixels and described camera and target person is 0~20 °, being considered as this instant head portrait photo is satisfactory instant head portrait photo; Described step 1 comprises following sub-step as shown in Figure 7:
Sub-step 110: by camera testing environment light intensity, when described ambient light intensity is less than default Second Threshold, by the LED lamp light filling carrying on mobile terminal.
Sub-step 111: the head portrait photo that obtains target person by the camera of mobile terminal; And by the upper gravity sensing device of mobile terminal, measure the shooting vertical angle of the head of camera and target person while taking; Whether the described shooting vertical angle of judgement is less than 15 °, if so, carries out sub-step 112, if not, points out the information of Image Acquisition failure;
Sub-step 112: people's face location, human face region is separated from the background of the instant head portrait photo described in sub-step 111;
Sub-step 113: eyes location, the position of definite pupil of both eyes in the human face region obtaining from sub-step 112;
In actual applications, because the computing power of existing mobile terminal is limited by the restriction of volume, chip development degree, there is a certain distance in its computing velocity and fixed computing machine, in order further to improve the arithmetic speed of this technology in mobile terminal, above-mentioned people's face location and eyes location adopt following method:
1) utilize to set threshold values automatically by human eye and people's face other parts and background separation or utilize human eye gray-scale value to carry out eyes location;
2) by gray level image being carried out after vertical and horizontal Gray Projection, tentatively fixed to people's face;
3), with the square frame seeker face of pupil size, when falling into black picture element number in frame and reaching maximum, the position of frame is eye position.
Sub-step 114: the distance of calculating the center of pupil of both eyes, according to the distance at the center of the position calculation pupil of both eyes of sub-step 113 determined pupil of both eyes, and whether the distance at the center of the pupil of both eyes described in judgement drops in 55~90 pixel coverages, if, carry out step 2, if not, output image obtains the information of failure;
Step 2: extract the face characteristic collection in above-mentioned satisfactory instant head portrait photo; Described step 2 comprises following sub-step as shown in Figure 8:
Sub-step 211: adjustment of image, the angle of inclination of proofreading and correct human face region, and according to the distance at resulting pupil of both eyes center in sub-step 114 ratio with the distance at default pupil of both eyes center, the pixel of adjustment human face region;
Sub-step 212: image pre-service, corrected human face region in sub-step 212 is carried out to pre-service, the exposure of color, face, face's light that specifically comprises face evenly, the processing of face's average color, the Gao Guang of face, blur level, brightness average, gray scale dynamic range, unevenness, overexposure ratio, under-exposure ratio, image sharpness, image blur;
Sub-step 213: extract face characteristic cluster, the human face region obtaining is extracted to the face characteristic value that is no less than through sub-step 212, the summation of the face characteristic value of described instant head portrait photo is the face characteristic cluster of head portrait photo.
Step 3: the face characteristic cluster in the existing head portrait photo of extraction target person; Consistent with the method for face characteristic cluster in the existing head portrait photo of extraction target person in specific embodiment 1.
Step 4: the face characteristic cluster in the face characteristic cluster in head portrait photo and existing head portrait photo is compared, calculate similarity, and whether the described similarity of judgement meets default first threshold, if, extract and show the identity information that described target person is corresponding, if not, carry out step 5;
Step 5: send in the server of peripheral hardware obtaining face characteristic cluster in step 2, the server stores of described peripheral hardware has face characteristic cluster and the corresponding people information that multiple human face photos, multiple human face photos are corresponding, the server of described peripheral hardware detects for the face characteristic cluster sending according to mobile terminal the corresponding human face photo of similarity that meets the first predetermined threshold value, and the described corresponding human face photo of the similarity that meets the first predetermined threshold value, the people information that this human face photo is corresponding are sent in mobile terminal; In actual applications, send to the screening index that also has user to arrange of the server of peripheral hardware together with face characteristic cluster, this screening index comprises the index that sex, the range of age, height etc. can target-marking character features, first server filters out by screening index the person who meets, and then the comparison of pedestrian's face feature cluster, reduce data processing time, improved analysis efficiency.
Step 6: human face photo and people information corresponding to this human face photo that the server of reception peripheral hardware sends, and be presented on the display screen of mobile terminal.
This specific embodiment adopts first measures shooting angle and light filling in the case of necessary, and then it is separated to carry out people's face, and by whether the photo of eyes location determination shooting is met to standard.This is also a kind of selection scheme of the present invention, has reduced the technical requirement of personnel's operations.
Above-described is only preferred embodiment of the present invention, all any modifications of doing within the scope of the spirit and principles in the present invention, is equal to and replaces and improvement etc., within all should being included in protection scope of the present invention.