1. A face detection method is characterized in that the realization method comprises the following steps:

(1) aiming at the facial features of the human face, the human face is divided into a plurality of key points for recognition, and the key points of the human face comprise: tribe, face, chin, cheekbones, apple muscle, temple, eye, nose, human middle, lip, and mouth.

(2) The method comprises the steps of distinguishing more detailed types of key points, including atrium length, face contour shape, chin contour shape, cheek bone height, apple muscle existence, temple type, eye type, nasal wing width, human middle length, lip shape and mouth angle curvature, collecting face data through a camera, measuring type information of the key points through an AI (artificial intelligence) technology, comparing the type information with data of a face key point library, calculating a difference value between the face key points and a standard structure to serve as basic face measurement data, comparing the basic face measurement data with a fixed feature library of five sense organs and face parts, and marking the feature of the five sense organs with small difference as the feature of the five sense organs of a user.

(3) Dividing the facial feature style into a juvenile state, a mature state, a sweet warm state and a high cold state, and then giving a score value corresponding to each facial feature style according to different types of information of each key point.

(4) The age feeling A is divided into three levels of younger A1, moderate A2 and more mature A3, and the calculation mode of the age feeling A is (juvenile-young) + (age-20).

(5) The cooling and heating degree B is divided into three levels of warmer B1, always B2 and colder B3, and the calculation mode of the cooling and heating degree B is high cooling-sweet heating.

(6) And combining the age feeling A and the temperature B, and judging the corresponding style of the human face according to different combinations and sexes.

2. The face detection method according to claim 1, characterized in that: when the calculated value of the age feeling A is less than or equal to 2, the people are classified as being younger A1, when the calculated value of the age feeling A is 2-11, the people are classified as being moderate A2, and when the calculated value of the age feeling A is more than or equal to 11, the people are classified as being more mature A3.

3. The face detection method according to claim 1, characterized in that: when the calculated value of the cooling and heating degree B is less than or equal to-15, the temperature is divided into a warmer B1, when the calculated value of the cooling and heating degree B is-15-9, the temperature is divided into a moderate B2, and when the calculated value of the cooling and heating degree B is greater than or equal to-9, the temperature is divided into a cooler B3.

4. The face detection method according to claim 1, characterized in that: the eye type comprises the angular degrees of the outer canthus in the left eye and the right eye, the size of the eye, the shape of the eyebrow, the distance between the two eyebrows, the existence of silkworm in bed and the existence of eye bags.

5. The face detection method according to claim 1, characterized in that: the lip shape includes a lip thickness and a lip peak type.

Background

With the development of the AI technology, the medical field is also continuously trying to combine with the artificial intelligence technology, so as to improve the diagnosis accuracy and improve the treatment result. The goal of AI is to accomplish work through learning that would otherwise only be done by humans. The medical information judgment which can be digitalized, structured and standardized can be completed by artificial intelligence. So the current applications are better in imaging, pathology and dermatology. In the medical beauty field, beauty and ugly people who judge the appearance of a person by feeling have long been used, and people who have existed in real life are often judged although the standards of beauty are acknowledged. There is no uniform standard. However, with the rapid development of imaging hardware and the high popularity of smart phones today, AI technology has made possible accurate quantitative measurements of the five sense organs and causal evidence to aesthetics.

Disclosure of Invention

The technical problem to be solved by the invention is to overcome the defects and provide a face detection method capable of judging and identifying the face style according to the face characteristic points.

In order to solve the technical problems, the technical scheme provided by the invention is as follows: a face detection method comprises the following steps:

(1) aiming at the facial features of the human face, the human face is divided into a plurality of key points for recognition, and the key points of the human face comprise: tribe, face, chin, cheekbones, apple muscle, temple, eye, nose, human middle, lip, and mouth.

(2) The method comprises the steps of distinguishing more detailed types of key points, including atrium length, face contour shape, chin contour shape, cheek bone height, apple muscle existence, temple type, eye type, nasal wing width, human middle length, lip shape and mouth angle curvature, collecting face data through a camera, measuring type information of the key points through an AI (artificial intelligence) technology, comparing the type information with data of a face key point library, calculating a difference value between the face key points and a standard structure to serve as basic face measurement data, comparing the basic face measurement data with a fixed feature library of five sense organs and face parts, and marking the feature of the five sense organs with small difference as the feature of the five sense organs of a user.

(3) Dividing the facial feature style into a juvenile state, a mature state, a sweet warm state and a high cold state, and then giving a score value corresponding to each facial feature style according to different types of information of each key point.

(4) The age feeling A is divided into three levels of younger A1, moderate A2 and more mature A3, and the calculation mode of the age feeling A is (juvenile-young) + (age-20).

(5) The cooling and heating degree B is divided into three levels of warmer B1, always B2 and colder B3, and the calculation mode of the cooling and heating degree B is high cooling-sweet heating.

(6) And combining the age feeling A and the temperature B, and judging the corresponding style of the human face according to different combinations and sexes.

Compared with the prior art, the invention has the advantages that: compared with the prior art, the invention integrates a plurality of AI technologies such as face detection, gender detection, facial shape, facial distance and position detection, facial feature points and the like, the front-mounted or rear-mounted camera of the smart phone is used for shooting and identifying, then the score is determined according to the type feature of each key point, the facial style of the user is judged by calculating the age feeling and the temperature and combining, the traditional judgment standard for comparing the facial style with the basic face is abandoned, and the judgment of the facial style is more accurate.

In an improvement, when the calculated value of the age feeling A is less than or equal to 2, the people are classified as younger A1, when the calculated value of the age feeling A is 2-11, the people are classified as moderate A2, and when the calculated value of the age feeling A is more than or equal to 11, the people are classified as more mature A3.

As an improvement, when the calculated value of the cooling and heating degree B is less than or equal to-15, the calculation value is divided into warmer B1, when the calculated value of the cooling and heating degree B is-15-9, the calculation value is divided into moderate B2, and when the calculated value of the cooling and heating degree B is greater than or equal to-9, the calculation value is divided into colder B3.

As an improvement, the eye type comprises the angular degrees of the outer canthus in the left eye and the right eye, the size of the eyes, the shape of the eyebrows, the distance between the two eyebrows, the existence of silkworm lying down and the existence of eye bags.

As a refinement, the lip shape includes a lip thickness and a lip peak type.

Drawings

FIG. 1 is a graph of a key point type-corresponding style score according to a face detection method of the present invention.

FIG. 2 is a combined style determination diagram of a face detection method according to the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings.

With reference to fig. 1, a face detection method is implemented by the following steps:

(1) aiming at the facial features of the human face, the human face is divided into a plurality of key points for recognition, and the key points of the human face comprise: tribe, face, chin, cheekbones, apple muscle, temple, eye, nose, human middle, lip, and mouth.

(2) The method comprises the steps of distinguishing more detailed types of key points, including atrium length, face contour shape, chin contour shape, cheek bone height, apple muscle existence, temple type, eye type, nasal wing width, human middle length, lip shape and mouth angle curvature, collecting face data through a camera, measuring type information of the key points through an AI (artificial intelligence) technology, comparing the type information with data of a face key point library, calculating a difference value between the face key points and a standard structure to serve as basic face measurement data, comparing the basic face measurement data with a fixed feature library of five sense organs and face parts, and marking the feature of the five sense organs with small difference as the feature of the five sense organs of a user.

(3) Dividing the facial feature style into a juvenile state, a mature state, a sweet warm state and a high cold state, and then giving a score value corresponding to each facial feature style according to different types of information of each key point.

(4) The age feeling A is divided into three levels of younger A1, moderate A2 and more mature A3, and the calculation mode of the age feeling A is (juvenile-young) + (age-20).

(5) The cooling and heating degree B is divided into three levels of warmer B1, always B2 and colder B3, and the calculation mode of the cooling and heating degree B is high cooling-sweet heating.

(6) And combining the age feeling A and the temperature B, and judging the corresponding style of the human face according to different combinations and sexes.

When the calculated value of the age feeling A is less than or equal to 2, the people are classified as being younger A1, when the calculated value of the age feeling A is 2-11, the people are classified as being moderate A2, and when the calculated value of the age feeling A is more than or equal to 11, the people are classified as being more mature A3.

When the calculated value of the cooling and heating degree B is less than or equal to-15, the temperature is divided into a warmer B1, when the calculated value of the cooling and heating degree B is-15-9, the temperature is divided into a moderate B2, and when the calculated value of the cooling and heating degree B is greater than or equal to-9, the temperature is divided into a cooler B3.

The eye type comprises the angular degrees of the outer canthus in the left eye and the right eye, the size of the eye, the shape of the eyebrow, the distance between the two eyebrows, the existence of silkworm in bed and the existence of eye bags.

The lip shape includes a lip thickness and a lip peak type.

The working principle of the invention is as follows: the method comprises the steps of collecting face data through a camera, scanning each key point of a face through an AI technology, comparing the data content of the key points with the data content of a fixed feature library, judging each feature according to comparison, enabling the feature to enter into an atrium, enabling the feature to correspond to four styles of juvenile, mature, sweet and warm, and high-cold, giving judgment scores for each style, giving a score for the mature score when the atrium is shorter than the atrium, and giving a score for the other three styles of 0, finally calculating the corresponding style scores of all the features of a user, adding the scores of each style into a whole, calculating age feeling A through (juvenile-young) + (age-20), calculating degree B through high-cold-sweet and warm, combining the calculated age feeling A with the cold and warm degrees B, judging the gender of the user according to the AI technology, and then judging the age feeling A, the cool and the warm degrees according to the age feeling A, The cold and warm degree B and the gender judge the style of the face of the user through the combined style judgment map, for example, the style of the face of a female with a younger A1 warmer B1 is a young style.

The present invention and its embodiments have been described above, and the description is not intended to be limiting, and the drawings are only one embodiment of the present invention, and the actual structure is not limited thereto. In summary, those skilled in the art should appreciate that they can readily use the disclosed conception and specific embodiments as a basis for designing or modifying other structures for carrying out the same purposes of the present invention without departing from the spirit and scope of the invention as defined by the appended claims.