Summary of the invention
The present invention, just based on the problems referred to above, proposes a kind of new technical scheme, can promote the Consumer's Experience performance of face recognition result further.
In view of this, an aspect of of the present present invention proposes a kind of face recognition display packing, comprising: the face-image obtaining destination object; Identified by the carrying out of predetermined facial recognition modules to described face-image, determine age information and the gender information of described destination object; Detect the age information of described destination object and the accuracy of gender information; According to testing result, determine the age information of described destination object and attribute information display mode corresponding to gender information, wherein, described attribute information display mode comprises accurate model and fuzzy pattern.
In this technical scheme, can over there after portion's image recognition age information of going out the destination object described in it and gender information, the age information identified and the accuracy of gender information, thus determine further according to accurate model or the attribute information such as appellation according to this destination object of fuzzy pattern.Predetermined face recognition algorithms described here is existing in the related to be embodied, and the face attribute released as Omron group infers technology, namely reads the information such as sex, age bracket by face-image.And determine the age information of destination object and gender information whether accurate, specifically can determine according to predetermined reliability algorithm such as, whether reach predetermined value by the scanning frame number detecting facial scan-data or judged by the reliability threshold test preset.When the age information and gender information that judge described destination object are accurate, then show according to accurate model; Judging that the age information of destination object and gender information are as time inaccurate, show according to fuzzy pattern.
Wherein, accurate model specifically refers to: detect the scanning frame number of facial scan data reach predetermined value or reliability reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are accurate, the age bracket accurately identified according to facial recognition modules and sex, show according to the division of predetermined age and sex.And fuzzy display specifically refers to: detect the scanning frame number of facial scan data do not reach predetermined value or reliability do not reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are inaccurate, and when showing, the attribute detecting inaccurate information is shown according to predetermined setting.
In addition, facial recognition modules detect gender information and age information all inaccurate time, can according to detect age information, carry out fuzzy display to gender information; Or after fuzzy display is carried out to age information, display in age is subtracted to age information, and show neutral appellation corresponding to the age information after subtracting age.
By this technical scheme, by the attribute information determined with fuzzy pattern to make up the Technique deviation of face recognition technology, make the result hommization more of face recognition, improve Consumer's Experience.
In technique scheme, preferably, described according to testing result, determine the age information of described destination object and attribute information display mode corresponding to gender information, specifically comprise: judge whether described accuracy reaches reliability detection threshold; When judged result is for being, show the age information of described destination object and attribute information corresponding to gender information according to described accurate model; When judged result is no, show the age information of described destination object and attribute information corresponding to gender information according to described fuzzy pattern.
In this technical scheme, can be whether credible according to the accuracy of the age information of reliability detection threshold determination destination object and gender information, when determining that this accuracy reaches reliability detection threshold, explanation testing result is credible, therefore can according to accurate model display attribute information, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 90%, reach reliability detection threshold 80%, then direct according to attribute information be female user, the age 50 shows.And if this accuracy does not reach reliability detection threshold, illustrate that testing result may have deviation, need to carry out display attribute information according to fuzzy pattern, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 50%, do not reach reliability detection threshold 80%, then according to attribute information be female user, working clan carries out fuzzy display.Fuzzy pattern described here, can comprise sex Fuzzy processing and age dimmedization process two parts.Such as, for age Fuzzy processing, the appellation that different age brackets is corresponding different can be set, and when detecting that destination object is female user, automatically the appellation that a upper age bracket (less age bracket) of the age bracket described in destination object is corresponding is shown as the appellation information to destination object.
In addition, about the fog-level of fuzzy pattern, can arrange according to the actual demand of user.Such as, under can being arranged on fuzzy pattern, the age information detected is reduced three years old to show, or also can arrange and the appellation of destination object is shown according to appellation corresponding after subtracting five years old age, thus make the result hommization more of face recognition, improve Consumer's Experience.
In above-mentioned arbitrary technical scheme, preferably, described judge whether described accuracy reaches reliability detection threshold before, also comprise: according to the threshold value setting command received, described reliability detection threshold be set.
In this technical scheme, reliability detection threshold can be preset, to adapt to the actual demand of user.
In above-mentioned arbitrary technical scheme, preferably, also comprising: the image according to receiving beautifies order, predetermined beautify select target in grade multiple and beautify grade landscaping treatment is carried out to described face-image; After landscaping treatment is carried out to described face-image, determine the age information of the described destination object after landscaping treatment and attribute information corresponding to gender information according to described accurate model.
In this technical scheme, can also carry out classification beautify face-image, wherein, beautifying higher grade, its correspondence to beautify rear age information lower.After completing landscaping treatment, subtract effect in age owing to having, therefore can directly according to the attribute information of accurate model display-object object.
In above-mentioned arbitrary technical scheme, preferably, after the age information determining the described destination object after landscaping treatment according to described accurate model described and attribute information corresponding to gender information, also comprise: according to the display setting command received, when being arranged on the described face-image of display after landscaping treatment, not show or according to the age information of the described destination object of described accurate model simultaneous display after landscaping treatment and attribute information corresponding to gender information.
In this technical scheme, after image is beautified, directly can be beautified the rear age information determined and attribute information corresponding to gender information is presented on the image after beautifying, also can according to the actual demand of user, do not show it and beautify the rear age information determined and attribute information corresponding to gender information, thus adapt to user's needs.
Another aspect of the present invention proposes a kind of face recognition display device, comprising: face-image acquiring unit, obtains the face-image of destination object; Face-image recognition unit, is identified by the carrying out of predetermined facial recognition modules to described face-image, determines age information and the gender information of described destination object; Accuracy detecting unit, detects the age information of described destination object and the accuracy of gender information; Display mode determining unit, according to testing result, determine the age information of described destination object and attribute information display mode corresponding to gender information, wherein, described attribute information display mode comprises accurate model and fuzzy pattern.
In this technical scheme, can over there after portion's image recognition age information of going out the destination object described in it and gender information, the age information identified and the accuracy of gender information, thus determine further according to accurate model or the attribute information such as appellation according to this destination object of fuzzy pattern.Predetermined face recognition algorithms described here is existing in the related to be embodied, and the face attribute released as Omron group infers technology, namely reads the information such as sex, age bracket by face-image.And determine the age information of destination object and gender information whether accurate, specifically can determine according to predetermined reliability algorithm such as, whether reach predetermined value by the scanning frame number detecting facial scan-data or judged by the reliability threshold test preset.When the age information and gender information that judge described destination object are accurate, then show according to accurate model; Judging that the age information of destination object and gender information are as time inaccurate, show according to fuzzy pattern.
Wherein, accurate model specifically refers to: detect the scanning frame number of facial scan data reach predetermined value or reliability reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are accurate, the age bracket accurately identified according to facial recognition modules and sex, show according to the division of predetermined age and sex.And fuzzy display specifically refers to: detect the scanning frame number of facial scan data do not reach predetermined value or reliability do not reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are inaccurate, and when showing, the attribute detecting inaccurate information is shown according to predetermined setting.
In addition, facial recognition modules detect gender information and age information all inaccurate time, can according to detect age information, carry out fuzzy display to gender information; Or after fuzzy display is carried out to age information, display in age is subtracted to age information, and show neutral appellation corresponding to the age information after subtracting age.
By this technical scheme, by the attribute information determined with fuzzy pattern to make up the Technique deviation of face recognition technology, make the result hommization more of face recognition, improve Consumer's Experience.
In technique scheme, preferably, described display mode determining unit comprises: judging unit, judges whether described accuracy reaches reliability detection threshold; Display unit, when judged result is for being, the age information of described destination object and attribute information corresponding to gender information is shown according to described accurate model, and when judged result is no, show the age information of described destination object and attribute information corresponding to gender information according to described fuzzy pattern.
In this technical scheme, can be whether credible according to the accuracy of the age information of reliability detection threshold determination destination object and gender information, when determining that this accuracy reaches reliability detection threshold, explanation testing result is credible, therefore can according to accurate model display attribute information, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 90%, reach reliability detection threshold 80%, then direct according to attribute information be female user, the age 50 shows.And if this accuracy does not reach reliability detection threshold, illustrate that testing result may have deviation, need to carry out display attribute information according to fuzzy pattern, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 50%, do not reach reliability detection threshold 80%, then according to attribute information be female user, working clan carries out fuzzy display.Fuzzy pattern described here, can comprise sex Fuzzy processing and age dimmedization process two parts.Such as, for age dimmedization process, the appellation that different age brackets is corresponding different can be set, and when detecting that destination object is female user, automatically the appellation that a upper age bracket (less age bracket) of the age bracket described in destination object is corresponding is shown as the appellation information to destination object.
In addition, about the fog-level of fuzzy pattern, can arrange according to the actual demand of user.Such as, under can being arranged on fuzzy pattern, the age information detected is reduced three years old to show, or also can arrange and the appellation of destination object is shown according to appellation corresponding after subtracting five years old age, thus make the result hommization more of face recognition, improve Consumer's Experience.
In above-mentioned arbitrary technical scheme, preferably, also comprise: threshold setting unit, described judge whether described accuracy reaches reliability detection threshold before, according to the threshold value setting command received, described reliability detection threshold is set.
In this technical scheme, reliability detection threshold can be preset, to adapt to the actual demand of user.
In above-mentioned arbitrary technical scheme, preferably, also comprise: image beautifies unit, the image according to receiving beautifies order, predetermined beautify select target in grade multiple and beautifies grade and carry out landscaping treatment to described face-image; And described display mode determining unit also for: after landscaping treatment is carried out to described face-image, determine the age information of the described destination object after landscaping treatment and attribute information corresponding to gender information according to described accurate model.
In this technical scheme, can also carry out classification beautify face-image, wherein, beautifying higher grade, its correspondence to beautify rear age information lower.After completing landscaping treatment, subtract effect in age owing to having, therefore can directly according to the attribute information of accurate model display-object object.
In above-mentioned arbitrary technical scheme, preferably, also comprise: display setting device, after the age information determining the described destination object after landscaping treatment according to described accurate model described and attribute information corresponding to gender information, according to the display setting command received, when being arranged on the described face-image of display after landscaping treatment, not show or according to the age information of the described destination object of described accurate model simultaneous display after landscaping treatment and attribute information corresponding to gender information.
In this technical scheme, after image is beautified, directly can be beautified the rear age information determined and attribute information corresponding to gender information is presented on the image after beautifying, also can according to the actual demand of user, do not show it and beautify the rear age information determined and attribute information corresponding to gender information, thus adapt to user's needs.
Another aspect of the invention proposes a kind of terminal, comprises the face recognition display device according to any one of technique scheme.Therefore, this terminal has the technique effect identical with the face recognition display device according to any one of technique scheme, does not repeat them here.
By above technical scheme, by the attribute information determined with accurate model or fuzzy pattern to make up the Technique deviation of face recognition technology, make the result hommization more of face recognition, improve Consumer's Experience.
Embodiment
In order to more clearly understand above-mentioned purpose of the present invention, feature and advantage, below in conjunction with the drawings and specific embodiments, the present invention is further described in detail.It should be noted that, when not conflicting, the feature in the embodiment of the application and embodiment can combine mutually.
Set forth a lot of detail in the following description so that fully understand the present invention; but; the present invention can also adopt other to be different from other modes described here and implement, and therefore, protection scope of the present invention is not by the restriction of following public specific embodiment.
Fig. 1 shows the process flow diagram of face recognition display packing according to an embodiment of the invention.
As shown in Figure 1, face recognition display packing according to an embodiment of the invention, comprising:
Step 102, obtains the face-image of destination object;
Step 104, is identified by the carrying out of predetermined facial recognition modules to described face-image, determines age information and the gender information of described destination object;
Step 106, detects the age information of described destination object and the accuracy of gender information;
Step 108, according to testing result, determine the age information of described destination object and attribute information display mode corresponding to gender information, wherein, described attribute information display mode comprises accurate model and fuzzy pattern.
In this technical scheme, can over there after portion's image recognition age information of going out the destination object described in it and gender information, the age information identified and the accuracy of gender information, thus determine further according to accurate model or the attribute information such as appellation according to this destination object of fuzzy pattern.Predetermined face recognition algorithms described here is existing in the related to be embodied, and the face attribute released as Omron group infers technology, namely reads the information such as sex, age bracket by face-image.And determine the age information of destination object and gender information whether accurate, specifically can determine according to predetermined reliability algorithm such as, whether reach predetermined value by the scanning frame number detecting facial scan-data or judged by the reliability threshold test preset.When the age information and gender information that judge described destination object are accurate, then show according to accurate model; Judging that the age information of destination object and gender information are as time inaccurate, show according to fuzzy pattern.
Wherein, accurate model specifically refers to: detect the scanning frame number of facial scan data reach predetermined value or reliability reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are accurate, the age bracket accurately identified according to facial recognition modules and sex, show according to the division of predetermined age and sex.And fuzzy display specifically refers to: detect the scanning frame number of facial scan data do not reach predetermined value or reliability do not reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are inaccurate, and when showing, the attribute detecting inaccurate information is shown according to predetermined setting.
In addition, facial recognition modules detect gender information and age information all inaccurate time, can according to detect age information, carry out fuzzy display to gender information; Or after fuzzy display is carried out to age information, display in age is subtracted to age information, and show neutral appellation corresponding to the age information after subtracting age.
By this technical scheme, by the attribute information determined with fuzzy pattern to make up the Technique deviation of face recognition technology, make the result hommization more of face recognition, improve Consumer's Experience.
In technique scheme, preferably, step 108 specifically comprises: judge whether described accuracy reaches reliability detection threshold; When judged result is for being, show the age information of described destination object and attribute information corresponding to gender information according to described accurate model; When judged result is no, show the age information of described destination object and attribute information corresponding to gender information according to described fuzzy pattern.
In this technical scheme, can be whether credible according to the accuracy of the age information of reliability detection threshold determination destination object and gender information, when determining that this accuracy reaches reliability detection threshold, explanation testing result is credible, therefore can according to accurate model display attribute information, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 90%, reach reliability detection threshold 80%, then direct according to attribute information be female user, the age 50 shows.And if this accuracy does not reach reliability detection threshold, illustrate that testing result may have deviation, need to carry out display attribute information according to fuzzy pattern, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 50%, do not reach reliability detection threshold 80%, then according to attribute information be female user, working clan carries out fuzzy display.Fuzzy pattern described here, can comprise sex Fuzzy processing and age dimmedization process two parts.Such as, for age dimmedization process, the appellation that different age brackets is corresponding different can be set, and when detecting that destination object is female user, automatically the appellation that a upper age bracket (less age bracket) of the age bracket described in destination object is corresponding is shown as the appellation information to destination object.
In addition, about the fog-level of fuzzy pattern, can arrange according to the actual demand of user.Such as, under can being arranged on fuzzy pattern, the age information detected is reduced three years old to show, or also can arrange and the appellation of destination object is shown according to appellation corresponding after subtracting five years old age, thus make the result hommization more of face recognition, improve Consumer's Experience.
In above-mentioned arbitrary technical scheme, preferably, described judge whether described accuracy reaches reliability detection threshold before, also comprise: according to the threshold value setting command received, described reliability detection threshold be set.
In this technical scheme, reliability detection threshold can be preset, to adapt to the actual demand of user.
In above-mentioned arbitrary technical scheme, preferably, also comprising: the image according to receiving beautifies order, predetermined beautify select target in grade multiple and beautify grade landscaping treatment is carried out to described face-image; After landscaping treatment is carried out to described face-image, determine the age information of the described destination object after landscaping treatment and attribute information corresponding to gender information according to described accurate model.
In this technical scheme, can also carry out classification beautify face-image, wherein, beautifying higher grade, its correspondence to beautify rear age information lower.After completing landscaping treatment, subtract effect in age owing to having, therefore can directly according to the attribute information of accurate model display-object object.
In above-mentioned arbitrary technical scheme, preferably, after the age information determining the described destination object after landscaping treatment according to described accurate model described and attribute information corresponding to gender information, also comprise: according to the display setting command received, when being arranged on the described face-image of display after landscaping treatment, not show or according to the age information of the described destination object of described accurate model simultaneous display after landscaping treatment and attribute information corresponding to gender information.
In this technical scheme, after image is beautified, directly can be beautified the rear age information determined and attribute information corresponding to gender information is presented on the image after beautifying, also can according to the actual demand of user, do not show it and beautify the rear age information determined and attribute information corresponding to gender information, thus adapt to user's needs.
Fig. 2 shows the block diagram of face recognition display device according to an embodiment of the invention.
As shown in Figure 2, face recognition display device 200 according to an embodiment of the invention, comprising: face-image acquiring unit 202, obtains the face-image of destination object; Face-image recognition unit 204, is identified the carrying out of face-image by predetermined facial recognition modules, determines age information and the gender information of destination object; Accuracy detecting unit 206, the age information of detected target object and the accuracy of gender information; Display mode determining unit 208, according to testing result, determine the age information of destination object and attribute information display mode corresponding to gender information, wherein, attribute information display mode comprises accurate model and fuzzy pattern.
In this technical scheme, can over there after portion's image recognition age information of going out the destination object described in it and gender information, the age information identified and the accuracy of gender information, thus determine further according to accurate model or the attribute information such as appellation according to this destination object of fuzzy pattern.Predetermined face recognition algorithms described here is existing in the related to be embodied, and the face attribute released as Omron group infers technology, namely reads the information such as sex, age bracket by face-image.And determine the age information of destination object and gender information whether accurate, specifically can determine according to predetermined reliability algorithm such as, whether reach predetermined value by the scanning frame number detecting facial scan-data or judged by the reliability threshold test preset.When the age information and gender information that judge described destination object are accurate, then show according to accurate model; Judging that the age information of destination object and gender information are as time inaccurate, show according to fuzzy pattern.
Wherein, accurate model specifically refers to: detect the scanning frame number of facial scan data reach predetermined value or reliability reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are accurate, the age bracket accurately identified according to facial recognition modules and sex, show according to the division of predetermined age and sex.And fuzzy display specifically refers to: detect the scanning frame number of facial scan data do not reach predetermined value or reliability do not reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are inaccurate, and when showing, the attribute detecting inaccurate information is shown according to predetermined setting.
In addition, facial recognition modules detect gender information and age information all inaccurate time, can according to detect age information, carry out fuzzy display to gender information; Or after fuzzy display is carried out to age information, display in age is subtracted to age information, and show neutral appellation corresponding to the age information after subtracting age.
By this technical scheme, by the attribute information determined with fuzzy pattern to make up the Technique deviation of face recognition technology, make the result hommization more of face recognition, improve Consumer's Experience.
In technique scheme, preferably, display mode determining unit 208 comprises: judging unit 2082, and whether accuracy of judgement degree reaches reliability detection threshold; Display unit 2084, when judged result is for being, according to the age information of accurate model display-object object and attribute information corresponding to gender information, and when judged result is no, according to the age information of fuzzy pattern display-object object and attribute information corresponding to gender information.
In this technical scheme, can be whether credible according to the accuracy of the age information of reliability detection threshold determination destination object and gender information, when determining that this accuracy reaches reliability detection threshold, explanation testing result is credible, therefore can according to accurate model display attribute information, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 90%, reach reliability detection threshold 80%, then direct according to attribute information be female user, the age 50 shows.And if this accuracy does not reach reliability detection threshold, illustrate that testing result may have deviation, need to carry out display attribute information according to fuzzy pattern, such as, if face recognition testing result be female user, the age 50 time, if its accuracy is 50%, do not reach reliability detection threshold 80%, then according to attribute information be female user, working clan carries out fuzzy display.Here fuzzy pattern, can comprise sex Fuzzy processing and age dimmedization process two parts.Such as, for age dimmedization process, the appellation that different age brackets is corresponding different can be set, and when detecting that destination object is female user, automatically the appellation that a upper age bracket (less age bracket) of the age bracket of destination object is corresponding is shown as the appellation information to destination object.
In addition, about the fog-level of fuzzy pattern, can arrange according to the actual demand of user.Such as, under can being arranged on fuzzy pattern, the age information detected is reduced three years old to show, or also can arrange and the appellation of destination object is shown according to appellation corresponding after subtracting five years old age, thus make the result hommization more of face recognition, improve Consumer's Experience.
In above-mentioned arbitrary technical scheme, preferably, also comprise: threshold setting unit 210, before whether accuracy of judgement degree reaches reliability detection threshold, according to the threshold value setting command received, reliability detection threshold is set.
In this technical scheme, reliability detection threshold can be preset, to adapt to the actual demand of user.
In above-mentioned arbitrary technical scheme, preferably, also comprise: image beautifies unit 212, the image according to receiving beautifies order, predetermined beautify select target in grade multiple and beautifies grade and carry out landscaping treatment to face-image; And display mode determining unit 208 also for: after portion's image carries out landscaping treatment over there, determine the age information of the destination object after landscaping treatment and attribute information corresponding to gender information according to accurate model.
In this technical scheme, can also carry out classification beautify face-image, wherein, beautifying higher grade, its correspondence to beautify rear age information lower.After completing landscaping treatment, subtract effect in age owing to having, therefore can directly according to the attribute information of accurate model display-object object.
In above-mentioned arbitrary technical scheme, preferably, also comprise: display setting device 214, after the attribute information that the age information and gender information of determining the destination object after landscaping treatment according to accurate model are corresponding, according to the display setting command received, when being arranged on the face-image of display after landscaping treatment, not show or according to the age information of the destination object of accurate model simultaneous display after landscaping treatment and attribute information corresponding to gender information.
In this technical scheme, after image is beautified, directly can be beautified the rear age information determined and attribute information corresponding to gender information is presented on the image after beautifying, also can according to the actual demand of user, do not show it and beautify the rear age information determined and attribute information corresponding to gender information, thus adapt to user's needs.
Fig. 3 shows the block diagram of terminal according to an embodiment of the invention.
As shown in Figure 3, terminal 300 according to an embodiment of the invention, comprises face recognition display device 200 as shown in Figure 2, and therefore, terminal 300 has the technique effect identical with the face recognition display device 200 shown in Fig. 2, does not repeat them here.
Fig. 4 shows the process flow diagram of face recognition display packing according to another embodiment of the invention.
As shown in Figure 4, face recognition display packing according to another embodiment of the invention, comprising:
Step 402, after the face-image obtaining destination object, by age information and the gender information of facial recognition modules determination destination object.
Step 404, determine the age information of destination object and gender information whether accurate, when judged result is for being, enter step 406, otherwise, enter step 408.
Step 406, carries out accurate model and shows.
Step 408, carries out fuzzy pattern and shows.
Wherein, in step 404, determine the age information of destination object and gender information whether accurate, specifically can determine according to predetermined reliability algorithm, such as, whether the scanning frame number by detecting facial scan-data is reached predetermined value or is judged by the reliability threshold test preset.When the age information and gender information that judge described destination object are accurate, then show according to accurate model; Judging that the age information of destination object and gender information are as time inaccurate, show according to fuzzy pattern.Wherein, fuzzy pattern can comprise sex obfuscation display and age dimmedization display two parts.Such as, for age dimmedization display, can appellation corresponding to preset different age bracket, and when detecting that destination object is female user, automatically the appellation that a upper age bracket (less age bracket) of the age bracket of destination object is corresponding is shown as the appellation information to destination object.
Accurate model specifically refers to: detect the scanning frame number of facial scan data reach predetermined value or reliability reach default reliability threshold value time, assert that age information and the gender information of facial recognition modules identification are accurate, the age bracket accurately identified according to facial recognition modules and sex, show according to the division of predetermined age and sex.Predetermined age and the division of sex can define by form as shown in table 1 below:
Table 1
Age bracket |
The male sex |
Women |
Age bracket 1 |
Appellation A |
Appellation D |
Age bracket 2 |
Appellation B |
Appellation E |
Age bracket 3 |
Appellation C |
Appellation F |
In addition, sex appellation includes but not limited to: networking appellation (as little fresh meat), ancient customs appellation (as sixteen-year-old beauty), traditional colloquial style appellation (as girl), dialect appellation (as thin child), foreign language appellation (as Cinderella) etc., in concrete application scenarios, can be as shown in table 2 below:
Table 2
Age bracket |
The male sex |
Women |
0-6 |
Well-behaved treasured |
Well-behaved treasured |
7-13 |
Just too |
Loli |
14-21 |
Little fresh meat |
Beautiful young girl |
22-30 |
Potentiality man god |
Vigor goddess |
31-39 |
Glamour man god |
Makings goddess |
40-59 |
Sprout uncle |
Sprout mother |
>60 |
Uncle |
Aunt |
Fuzzy display specifically refers to: detect the scanning frame number of facial scan data do not reach predetermined value or reliability do not reach default reliability threshold value time, assert that age information and/or the gender information of facial recognition modules identification are inaccurate, and when showing, the attribute detecting inaccurate information is shown according to predetermined setting, such as, predetermined set is as following table 3:
Table 3
Age bracket |
Neutral appellation |
Age bracket 1 |
Appellation X |
Age bracket 2 |
Appellation Y |
Age bracket 3 |
Appellation Z |
In concrete application scenarios, when judging that gender information is inaccurate, can export neutral appellation according to Different age group, sex obfuscation reliability algorithm is in a particular embodiment as following table 4:
Table 4
Age bracket |
Neutral appellation |
0-6 |
Well-behaved treasured |
7-13 |
Young Pioneer |
14-21 |
Student party |
22-30 |
Overtime work race |
31-39 |
Potential stocks |
40-59 |
Good performance stocks |
>60 |
Life winner |
When age information judges inaccurate, according to sex and age testing result, can show, such as, shown in following table 5 according to non-linearly deducting the corresponding real gender information of certain age on current age:
Table 5
Age bracket |
The male sex |
Women |
Age bracket 1 |
Deduct N1 year |
Deduct n1 year |
Age bracket 2 |
Deduct N2 year |
Deduct n2 year |
Age bracket 3 |
Deduct N3 year |
Deduct n3 year |
In concrete application scenarios, age obfuscation in a particular embodiment can according to as shown in table 6 below:
Table 6
Age bracket |
The male sex |
Women |
0-21 |
Deduct 0 years old |
Deduct 1 years old |
22-30 |
Deduct 1 years old |
Deduct 3 years old |
31-39 |
Deduct 3 years old |
Deduct 5 years old |
40-59 |
Deduct 5 years old |
Deduct 7 years old |
>60 |
Deduct 7 years old |
Deduct 9 years old |
In addition, facial recognition modules detect gender information and age information all inaccurate time, can according to detect age information, carry out fuzzy display to gender information; Or after fuzzy display is carried out to age information, display in age is subtracted to age information, and show neutral appellation corresponding to the age information after subtracting age.
Fig. 5 A and Fig. 5 B shows the schematic diagram that setting according to an embodiment of the invention beautifies grade.
As shown in Figure 5A, when portion's image beautifies over there, can select to make a reservation for beautify grade is secondary.
As shown in Figure 5 B, after completing and beautifying, by facial recognition modules, the image after beautifying is identified, determine that the age information of its correspondence and appellation are respectively 33 years old and makings goddess, and be directly presented on image.
As shown in Figure 5 C, when portion's image beautifies over there, also can select to make a reservation for beautify grade is Pyatyi.
As shown in Figure 5 D, after completing and beautifying, by facial recognition modules, the image after beautifying is identified, determine that the age information of its correspondence and appellation are respectively 19 years old and beautiful young girl, and be directly presented on image.
Fig. 6 shows the process flow diagram of display of taking pictures according to an embodiment of the invention.
As shown in Figure 6, according to one embodiment of present invention, comprise when taking pictures:
Step 602, detecting and whether opens " band face information is taken pictures " function, when judged result is for being, entering step 604, otherwise, enter step 606.
Step 604, preserves the photo of normal photo and band face information when taking pictures, as shown in Figure 7 A, for being with the photo of face information.Particularly, identified by facial recognition modules comparison film upon taking a picture, and according to the accuracy identified, determine to use accurate model or use fuzzy pattern to show its face information.
Step 606, only preserves normal photo when taking pictures, and as shown in Figure 7 B, is the normal photo not with face information.
More than be described with reference to the accompanying drawings technical scheme of the present invention, by technical scheme of the present invention, by the attribute information determined with accurate model or fuzzy pattern to make up the Technique deviation of face recognition technology, make the result hommization more of face recognition, improve Consumer's Experience.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.