Method and system for carrying out same identification on personal identity on basis of phrenological geometrical characteristics

A technology of geometric features and personnel, applied in the field of technical investigation, can solve the problems of high requirements for shooting environment and attitude, and large impact of shooting time interval, and achieves the effect of good robustness and high reference value.

Inactive Publication Date: 2016-03-16
CHONGQING ZHONGKE YUNCONG TECH CO LTD
3 Cites 3 Cited by

AI-Extracted Technical Summary

Problems solved by technology

However, due to technical limitations, this type of method currently has high requirements for the shooting environment and posture, and is greatly affected by factors such as shoot...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Abstract

The present invention discloses a method and a system for carrying out same identification on a personal identity on the basis of phrenological geometrical characteristics. The method comprises: identifying a facial image of a to-be-identified person to determine a plurality of markers used for marking the phrenological geometrical characteristics of the to-be-identified person; generating a plurality of marking lines according to the plurality of markers; and calculating the phrenological geometrical characteristics according to the plurality of markers and the plurality of marking lines, and calculating a first characteristic vector according to the phrenological geometrical characteristics; according to the first characteristic vector and a pre-calculated second characteristic vector, calculating a similarity between the facial image of the to-be-identified person and a docket facial image, which is relevant to a suspected identity of the to-be-identified person, in a public security system, wherein the second characteristic vector is a characteristic vector calculated according to the phrenological geometrical characteristics of the docket facial image; when the similarity is greater than a predetermined threshold, determining that the to-be-identified person is coincided with the suspected identity.

Application Domain

Technology Topic

Feature vectorPublic security

Image

  • Method and system for carrying out same identification on personal identity on basis of phrenological geometrical characteristics
  • Method and system for carrying out same identification on personal identity on basis of phrenological geometrical characteristics
  • Method and system for carrying out same identification on personal identity on basis of phrenological geometrical characteristics

Examples

  • Experimental program(2)

Example

[0017] First embodiment
[0018] figure 1 It shows a flow chart of the method for performing identity authentication of a person based on cranial geometric features provided by the first embodiment of the present invention. See figure 1 The first embodiment of the present invention provides a method for the same authentication of a person's identity based on cranial geometric features, which may include:
[0019] Step S11: Recognizing the facial image of the person to be identified to determine a plurality of marking points for marking the cranial features of the person to be identified.
[0020] The facial image is preferably a frontal facial image of a person to be identified, and the person to be identified usually refers to a suspect who is being searched and pursued by a public security agency.
[0021] In a specific embodiment, the left eye, right eye, nose, mouth, cheek, and chin in the facial image of the person to be identified can be automatically identified using, for example, the cascade classifier, and then it can be determined that it can better reflect the person to be identified. The location of 13 marking points of cranial features. The marking points may include left eyebrow center point, right eyebrow center point, between eyebrow points, left outer eye point, right outer eye point, left inner eye point, right inner eye point, lower nose point, left mouth corner point, right mouth corner point, The submandibular point, the left wide end point and the right wide end point, where the left eyebrow center point is at the center of the left eyebrow, the right eyebrow center point is at the center of the right eyebrow, the eyebrow center point is the center point between the two eyebrows, and the left outer eye point is at the left The junction of the upper and lower eyelids on the outer side of the eye cleft, the right outer eye point is located at the junction of the upper and lower eyelids on the outer side of the right cleft eye, the left inner eye point is located at the upper and lower eyelid junction on the inner side of the left cleft eye, and the right inner eye point is located on the upper and lower sides of the inner right cleft The eyelid junction, the subnasal point is the deepest point of the angle formed by the lower edge of the nasal septum and the upper lip skin transition. The left corner point is at the junction of the upper and lower lips on the left side of the mouth, and the right corner point is at the junction of the upper and lower lips on the right side of the mouth, and the submandibular point Located at the center of the bottom of the chin, the left wide end is located at the most prominent position on the left side of the facial contour, and the right wide end is located at the most prominent position on the right side of the facial contour in the horizontal direction.
[0022] The recognition of the facial image of the person to be identified to determine multiple marking points for marking the cranial features of the person to be identified may include: left eyebrow center point, right eyebrow center point, and eyebrow on the face image of the person to be identified Recognize between the middle point, left outer eye point, right outer eye point, left inner eye point, right inner eye point, undernose point, left mouth corner point, right mouth corner point, submandibular point, left face wide end point and right face wide end point for recognition ; Determine and store the recognized left eyebrow center point, right eyebrow center point, inter-brow point, left outer eye point, right outer eye point, left inner eye point, right inner eye point, lower nose point, left mouth corner point, and right mouth corner point , The position of the submandibular point, the left face width end point, and the right face width end point in the facial image.
[0023] Step S12, generating multiple marking lines according to the multiple marking points.
[0024] In a specific embodiment, the generating multiple marking lines based on the multiple marking points may include: connecting the left outer eye point and the right outer eye point to generate an outer eye point line; The point between the eyebrows is perpendicular to the midline of the line connecting the outer eye points, the upper end of the midline reaches the center of the forehead and the lower end of the midline reaches the bottom of the mandibular chin; the generation passes through the subnasal point and is perpendicular to the front The undernose line of the midline, the length of the undernose line is close to the distance between the left side wide end point and the right side wide end point; connect the left mouth corner point and the right mouth corner point to generate a mouth corner line; generate passing through the submandibular point And the submandibular line perpendicular to the midline. When the mouth of the person to be identified is in the closed position in the face image of the person to be identified, the mouth corner line usually passes through the intersection of the upper and lower lips. That is, in step S12, 5 marking lines are generated using the 13 marking points determined in step S11.
[0025] Step S13: Calculate cranial geometric features based on the multiple marking points and the multiple marking lines, and calculate a first feature vector based on the cranial geometric features.
[0026] In a specific embodiment, the calculating cranial geometric features based on the multiple marking points and the multiple marking lines may include: calculating the left eyebrow center point or the right eyebrow center point to the lower nose The first distance of the line (L 0 ); Calculate the second distance from the left outer eye point or the right outer eye point to the lower nose line (L 1 ); Calculate the third distance from the left inner eye point or the right inner eye point to the lower nose line (L 2 ); Calculate the fourth distance from the corner of the mouth to the lower line of the nose (L 3 ); Calculate the fifth distance from the submandibular line to the subnasal line (L 4 ); Calculate the sixth distance from the left outer eye point to the median line (L 5 ); Calculate the seventh distance from the right outer eye point to the median line (L 6 ); Calculate the eighth distance from the left inner eye point to the median line (L 7 ); Calculate the ninth distance from the right inner eye point to the median line (L 8 ); Calculate the tenth distance from the left side wide end point to the median line (L 9 ); Calculate the eleventh distance from the wide end of the right face to the median line (L 10 ).
[0027] Correspondingly, the calculating the first feature vector according to the cranial geometric features may include: constructing a scale factor vector including 10 components according to the first distance to the eleventh distance, wherein the scale factor vector The first distance is used as the denominator for each component, and the second distance to the eleventh distance is used as the numerator for each component; the scale factor vector is unitized to obtain the first feature vector.
[0028] Specifically, the constructed scale factor vector can be Where σ m1 =L 1 /L 0 , Σ m2 =L 2 /L 0 , Σ m3 =L 3 /L 0 , Σ m4 =L 4 /L 0 , Σ m5 =L 5 /L 0 , Σ m6 =L 6 /L 0 , Σ m7 =L 7 /L 0 , Σ m8 =L 8 /L 0 , Σ m9 =L 9 /L 0 , Σ m10 =L 10 /L 0 , The first feature vector obtained is: among them
[0029] Step S14, according to the first feature vector And the pre-computed second eigenvector Calculate the similarity between the facial image of the to-be-identified person and the recorded facial image related to the suspected identity of the to-be-identified person in the public security system, wherein the second feature vector is based on the cranial phase of the recorded facial image Feature vector for geometric feature calculation.
[0030] The suspected identity of the person to be identified may refer to the suspected identity of the person to be identified by the public security agency. For example, a criminal detained in a prison escaped and became a fugitive in search and pursuit by the public security agency. When a public security officer finds something similar to the fugitive in a certain place When a person is a person, the person is a person to be identified, and the suspected identity of the person to be identified is a fugitive. The recorded facial image may be a facial image filed against the fugitive in the public security system, which is preferably a frontal facial image. Second eigenvector It is a feature vector calculated using a method similar to that of calculating the first feature vector for the recorded facial image. specifically, Second eigenvector It can be pre-calculated, or it can be calculated when needed.
[0031] In a specific embodiment, the calculation of the facial image of the person to be identified is related to the suspected identity of the person to be identified in the public security system based on the first feature vector and the pre-calculated second feature vector The similarity between the recorded facial images may include: calculating the first feature vector With the pre-computed second eigenvector Euclidean distance between And the reciprocal of the Euclidean distance is As the similarity between the face image of the person to be identified and the registered facial image related to the suspected identity of the person to be identified in the public security system, where
[0032]
[0033] Step S15: When the similarity is greater than a preset threshold, it is determined that the person to be identified matches the suspected identity.
[0034] When calculated When it is greater than a preset threshold, it can be determined that the person to be identified matches the suspected identity. For example, in the above example, it can be determined that the person to be identified is a fugitive. The preset threshold may be set by the user, and it is not limited to a specific value. It can be seen that the Euclidean distance The smaller the value, the higher the similarity between the face image of the person to be identified and the registered face image related to the suspected identity of the person to be identified in the public security system; on the contrary, the Euclidean distance The larger the value of, the lower the similarity between the facial image of the to-be-identified person and the registered facial image related to the suspected identity of the to-be-identified person in the public security system.
[0035] When there are multiple filing facial images related to the suspected identity of the person to be identified in the public security system, for example, when there are first-generation ID card images and two second-generation ID card images related to the suspected identity of the person to be identified in the public security system If there is a large visual difference between the first-generation ID image and the second-generation ID image, the first feature vector and the pre-calculated sum can be calculated for each of the registered facial images. The Euclidean distance between the second feature vectors corresponding to the registered facial image, and the reciprocal of the Euclidean distance is used as the similarity between the facial image of the person to be identified and the registered facial image. Correspondingly, when each of the calculated similarities is greater than a preset threshold, it is determined that the person to be identified matches the suspected identity.
[0036] According to the method for identifying the same identity of a person based on the cranial geometric features provided by the embodiment of the present invention, a plurality of marking points for marking the cranial features of the person to be identified are determined by recognizing the facial image of the person to be identified, and according to the determined The marking points generate multiple marking lines, the first characteristic vector of the cranial geometric features of the person to be identified is calculated according to the marking points and the marking lines, and the calculation is obtained based on the first characteristic vector and the pre-calculated second characteristic vector. When the similarity between the facial image of the identified person and the registered facial image related to the suspected identity of the person to be identified in the public security system is greater than a preset threshold, it is determined that the person to be identified matches the suspected identity. The method provided by the embodiment of the present invention uses the morphometric information of the cranial marking points and the marking line to perform the same authentication of the identity of the person, which is suitable for the search and pursuit of the suspect involved, and has good conditions for the age, posture change, and shooting conditions of the person to be identified. Robustness can provide high reference value for on-site decision-making of case handlers.

Example

[0037] Second embodiment
[0038] figure 2 It shows a schematic block diagram of a system for performing identity authentication of a person based on cranial geometric features according to the second embodiment of the present invention. Refer to figure 2 In the second embodiment of the present invention, the system 200 for performing identity authentication of persons based on cranial geometric features may include a determining unit 210, a generating unit 220, a first calculating unit 230, a second calculating unit 240, and a determining unit 250.
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

no PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Similar technology patents

Improved MSER image matching algorithm

InactiveCN106529591AReduce running timeRobustCharacter and pattern recognitionScale invarianceImage structure
Owner:HUNAN VISION SPLEND PHOTOELECTRIC TECH

Position sensor-free vector control device for built-in permanent magnetic synchronous motor

ActiveCN102361430AAchieving accurate field orientation controlRobustElectronic commutation motor controlVector control systemsShaft voltageThree-phase
Owner:哈尔滨工大正元信息技术有限公司

Classification and recommendation of technical efficacy words

  • Robust
  • High reference value
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products