Face recognition method and device, storage medium and electronic equipment
阅读说明：本技术 人脸识别方法、装置、存储介质及电子设备 (Face recognition method and device, storage medium and electronic equipment ) 是由 梁昆 于 2017-06-30 设计创作，主要内容包括：一种人脸识别方法、装置、存储介质以及电子设备,所述方法包括：获取待测人脸数据(S101)；根据待测人脸数据,生成与待测人脸数据相关的第一衍生人脸数据集(S102)；将第一衍生人脸数据集中的衍生人脸数据分别进行年龄判别,生成与所述第一衍生人脸数据集对应的年龄分布区间(S103)；判断年龄分布区间是否与第一参考年龄区间匹配(S104)；若是,则根据年龄分布区间,获得待测人脸数据对应的年龄数据(S105)。所述方法可提高人脸年龄识别的准确率。(face recognition method, device, storage medium and electronic equipment, wherein the method comprises the steps of obtaining face data to be detected (S101), generating a derived face data set related to the face data to be detected (S102) according to the face data to be detected, respectively carrying out age discrimination on derived face data in a derived face data set to generate an age distribution interval corresponding to the derived face data set (S103), judging whether the age distribution interval is matched with a reference age interval (S104), and if so, obtaining age data corresponding to the face data to be detected (S105) according to the age distribution interval.)
A face recognition method, comprising the following steps:
acquiring data of a face to be detected;
generating an th derived face data set related to the face data to be detected according to the face data to be detected, wherein the th derived face data set comprises a plurality of different derived face data;
respectively carrying out age discrimination on the derived face data in the th derived face data set to generate an age distribution interval corresponding to the th derived face data set;
judging whether the age distribution interval is matched with an th reference age interval;
and if so, acquiring age data corresponding to the face data to be detected according to the age distribution interval.
The face recognition method as claimed in claim 1, wherein the face data to be tested generates the -th derived face data set based on a preset face data generation model, and further comprises, before the generating the face data to be tested into the -th derived face data set related to the face data to be tested:
acquiring reference face data;
generating a second derivative face data set related to the reference face data according to the reference face data, wherein the second derivative face data set comprises a plurality of different derivative face data;
respectively carrying out age discrimination on the derived face data in the second derived face data set to generate an age distribution interval corresponding to the second derived face data set;
judging whether the age distribution interval corresponding to the second derivative face data set is matched with a second reference age interval or not;
and if not, updating the model parameters corresponding to the face data generation model until the age distribution interval corresponding to the second derivative face data set is matched with the second reference age interval after the age is judged.
The face recognition method according to claim 1, wherein the generating age distribution intervals corresponding to the th derived face data set by age-discriminating the derived face data in the th derived face data set comprises:
respectively carrying out age discrimination on the plurality of derived face data through a preset age discrimination model to generate an age distribution interval corresponding to the th derived face data set;
the age distinguishing model is used for distinguishing age data corresponding to the characteristic data according to the characteristic data extracted from the derived face data.
The face recognition method of claim 3, wherein the age discrimination model includes a plurality of feature data discrimination models,
the feature data discrimination model is used to discriminate the feature data of a preset category, and obtain age data corresponding to the feature data of the preset category according to the feature data.
The face recognition method according to claim 3, wherein before the step of performing age discrimination on the derived face data in the th derived face data set through a preset age discrimination model to generate an age distribution interval corresponding to the th derived face data set, the method further comprises:
acquiring reference face data, wherein the reference face data comprises age data;
extracting feature data corresponding to the reference face data according to the reference face data;
and associating the characteristic data with age data corresponding to the characteristic data to generate the age discrimination model.
A face recognition method as claimed in claims 3 to 5, wherein the feature data is derived from the face data using deep learning and convolutional neural networks.
The face recognition method according to claims 3-5, wherein the age discrimination model is formed by associating and collecting a plurality of feature data discrimination models according to preset weight coefficients.
The face recognition method of claim 1, wherein the determining whether the age distribution interval matches an th reference age interval comprises:
judging whether the correlation degree between an age distribution interval obtained by age discrimination of the derived face data of and a reference age interval is higher than a preset threshold value or not;
if yes, acquiring age data corresponding to the face data to be detected according to the age distribution interval, wherein the age data comprises:
if the correlation degree is higher than a preset threshold value, acquiring age data corresponding to the face data to be detected according to the age distribution interval;
and if the correlation degree is lower than a preset threshold value, adjusting the interval position of the age distribution interval so as to enable the age distribution interval to be finally matched with the th reference age interval.
The face recognition method as claimed in claim 1, wherein the generating -th derived face data set related to the face data to be tested according to the face data to be tested comprises:
carrying out face aging treatment and face rejuvenation treatment on the face data to be detected according to preset different age groups to generate a plurality of derived face data which represent different ages and are related to the face data to be detected;
and continuously distributing the plurality of derived face data according to the age corresponding to the derived face data to form -th derived face data sets.
The face recognition method according to claim 9, wherein if yes, obtaining age data corresponding to the face data to be detected according to the age distribution interval includes:
acquiring age median data of the age distribution interval according to the age distribution interval;
and taking the age median data as age data corresponding to the face data to be detected.
A face recognition device, wherein the face recognition device comprises:
, an obtaining unit for obtaining the face data to be measured;
an generation unit, configured to generate, according to the face data to be tested, a th derived face data set related to the face data to be tested, where the th derived face data set includes a plurality of different derived face data;
an judging unit, configured to respectively perform age judgment on the derived face data in the th derived face data set, and generate an age distribution section corresponding to the th derived face data set;
a th judging unit for judging whether the age distribution section matches with the th reference age section, and
and the second acquisition unit is used for acquiring age data corresponding to the face data to be detected according to the age distribution interval if the face data to be detected is in the age distribution interval.
The face recognition apparatus as claimed in claim 11, wherein the face data under test generates the -th derived face data set based on a preset face data generation model, the face recognition apparatus further comprising:
a third acquisition unit configured to acquire reference face data;
a second generating unit, configured to generate a second derivative face data set related to the reference face data according to the reference face data, where the second derivative face data set includes a plurality of different derivative face data;
a second judging unit, configured to respectively perform age judgment on the derived face data in the second derived face data set, and generate an age distribution interval corresponding to the second derived face data set;
a second judging unit, configured to judge whether an age distribution section corresponding to the second derivative face data set matches a second reference age section; and
and if not, updating the model parameters corresponding to the face data generation model until the age distribution interval corresponding to the second derivative face data set is matched with the second reference age interval after the age is judged.
The face recognition apparatus as claimed in claim 11, wherein the decision unit comprises:
an judging subunit, configured to respectively judge ages of the plurality of derived face data according to a preset age judging model, and generate an age distribution interval corresponding to the derived face data set;
the age distinguishing model is used for distinguishing age data corresponding to the characteristic data according to the characteristic data extracted from the derived face data.
The face recognition method of claim 13, wherein the age discrimination model includes a plurality of feature data discrimination models,
the feature data discrimination model is used to discriminate the feature data of a preset category, and obtain age data corresponding to the feature data of the preset category according to the feature data.
The face recognition apparatus according to claim 13, wherein the th decision unit further comprises:
, an obtaining subunit, configured to obtain reference face data, where the reference face data includes age data;
the extracting subunit is used for extracting feature data corresponding to the reference face data according to the reference face data; and
-a generating subunit for associating the feature data with age data corresponding to the feature data to generate the age discrimination model.
The face recognition apparatus according to any of claims 13-15, wherein the feature data is obtained from face data using deep learning and convolutional neural networks.
The face recognition apparatus according to any of claims 13-15, wherein the age discrimination model is formed by associating and aggregating a plurality of feature data discrimination models according to a preset weighting factor.
The face recognition apparatus as claimed in claim 11, wherein the th judging unit comprises:
an th judging subunit, configured to judge whether a correlation between an age distribution interval obtained after age discrimination of each of the derived face data and a th reference age interval is higher than a preset threshold;
the second acquisition unit includes:
the second obtaining subunit is configured to, if the correlation is higher than a preset threshold, obtain age data corresponding to the face data to be detected according to the age distribution interval; and
and the adjusting subunit is used for adjusting the interval position of the age distribution interval if the correlation degree is lower than a preset threshold value, so that the age distribution interval is finally matched with the th reference age interval.
The face recognition apparatus as claimed in claim 11, wherein the generation unit comprises:
the second generation subunit is used for carrying out face aging treatment and face rejuvenation treatment on the face data to be detected according to preset different age groups to generate a plurality of derived face data which are related to the face data to be detected and represent different ages; and
and a second generation subunit, configured to perform continuous distribution processing on the plurality of derived face data according to the age sizes corresponding to the plurality of derived face data to form th derived face data set.
The face recognition apparatus according to claim 19, wherein the second acquisition unit comprises:
a third obtaining subunit, configured to obtain age median data corresponding to the th derived face data set after the continuous distribution processing in the age distribution section according to the result of the age determination in the age distribution section, and
and the confirming subunit is used for taking the age median data as age data corresponding to the face data to be detected.
storage medium, wherein the storage medium has stored therein a plurality of instructions adapted to be loaded by a processor to perform the method of face recognition according to claims 1-10.
electronic device, comprising a processor electrically connected to the memory, the memory storing instructions and data, the processor being configured to execute the face recognition method according to claims 1-10.
The present invention relates to the field of data processing technologies, and in particular, to a face recognition method, an apparatus, a storage medium, and an electronic device.
With the rapid development of electronic devices, , which is essential for people's daily life, especially for various applications in mobile terminals, different functions provide users with diversified user experiences, and people are increasingly unable to leave the entertainment and convenience brought by various applications.
Sometimes people may be interested in people's ages in photos when viewing the photos or other image information.
The embodiment of the invention provides face recognition methods, devices, storage media and electronic equipment, which can improve the face recognition efficiency and accuracy.
, the embodiment of the invention provides the following technical scheme that face recognition methods comprise the following steps:
acquiring data of a face to be detected;
generating an th derived face data set related to the face data to be detected according to the face data to be detected, wherein the th derived face data set comprises a plurality of different derived face data;
respectively carrying out age discrimination on the derived face data in the th derived face data set to generate an age distribution interval corresponding to the th derived face data set;
judging whether the age distribution interval is matched with an th reference age interval;
and if so, acquiring age data corresponding to the face data to be detected according to the age distribution interval.
In a second aspect, the embodiment of the present invention further provides the following technical solutions:
A face recognition device, wherein the face recognition device comprises:
, an obtaining unit for obtaining the face data to be measured;
an generation unit, configured to generate, according to the face data to be tested, a th derived face data set related to the face data to be tested, where the th derived face data set includes a plurality of different derived face data;
an judging unit, configured to respectively perform age judgment on the derived face data in the th derived face data set, and generate an age distribution section corresponding to the th derived face data set;
a th judging unit for judging whether the age distribution section matches with the th reference age section, and
and the second acquisition unit is used for acquiring age data corresponding to the face data to be detected according to the age distribution interval if the face data to be detected is in the age distribution interval.
In a third aspect, the embodiments of the present invention further provide the following technical solutions:
storage medium, wherein the storage medium has stored therein a plurality of instructions adapted to be loaded by a processor to perform the face recognition method as described above.
In a fourth aspect, the embodiments of the present invention further provide the following technical solutions:
electronic device, comprising a processor and a memory, wherein the processor is electrically connected with the memory, the memory is used for storing instructions and data, and the processor is used for executing the face recognition method.
The embodiment of the invention provides face recognition methods, devices, storage media and electronic equipment, which can improve the face recognition efficiency and accuracy.
Fig. 1 is a schematic flow chart illustrating an implementation of a face recognition method according to an embodiment of the present invention.
Fig. 2 is a flowchart of another implementation of the face recognition method according to the embodiment of the present invention.
Fig. 3 is a block diagram of implementation of age determination on derived face data according to an embodiment of the present invention.
Fig. 4 is a schematic flow chart of an implementation process of training a face data generation model according to an embodiment of the present invention.
Fig. 5 is a schematic flow chart illustrating an implementation of obtaining an age discrimination model according to an embodiment of the present invention.
Fig. 6 is a schematic structural diagram of a face recognition apparatus according to an embodiment of the present invention.
Fig. 7 is another schematic structural diagram of the face recognition apparatus according to the embodiment of the present invention.
Fig. 8 is a structural schematic diagram of another of the face recognition apparatus according to the embodiment of the present invention.
Fig. 9 is a schematic structural diagram of an th decision unit according to an embodiment of the present invention.
Fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Fig. 11 is another schematic structural diagram of the electronic device according to the embodiment of the present invention.