CN214896642U

CN214896642U - Face detection system

Info

Publication number: CN214896642U
Application number: CN202121193501.8U
Authority: CN
Inventors: 王正平; 林国勇; 王杉; 董善志; 边振兴
Original assignee: Shanghai Haiwen Information Technology Co ltd
Current assignee: Shanghai Haiwen Information Technology Co ltd
Priority date: 2021-05-31
Filing date: 2021-05-31
Publication date: 2021-11-26
Anticipated expiration: 2031-05-31

Abstract

The utility model relates to a face identification technical field, concretely relates to face detection system, face detection system is including detecting the body, the lag, the protection piece, ring gear and step motor, detect body and lag fixed connection, it is located the inside of lag to detect the body, the lag has the opening, protection piece and lag sliding connection, the protection piece is located the opening part, step motor fixed mounting is in the inside of lag, ring gear and step motor's output fixed connection, one side of protection piece is provided with the tooth form groove, the ring gear agrees with the tooth form groove mutually, setting through the above-mentioned structure, be stained with on the detection lens that detects the body to impurity such as preventing dust, make the detection camera lens that detects the body keep comparatively clean and tidy state, thereby improve the recognition effect of this face detection system.

Description

Face detection system

Technical Field

The utility model relates to a face identification technical field especially relates to a face detection system.

Background

An important application of face recognition is identity authentication, and in many important occasions, the traditional identity authentication method is slow and inefficient, so people try to find more efficient and convenient identity authentication schemes, face recognition is one of the considerations, and compared with other types of biological recognition, face recognition has the advantages of non-mandatory, non-contact, hidden and concurrency, and the facial expression of a person can reflect the psychological state and emotion of the person, so that necessary auxiliary information can be provided for people.

Most of the existing face recognition systems are fixedly exposed in the air, and impurities such as dust and the like are possibly attached to the lens due to long-time standing, so that the recognition effect is influenced.

SUMMERY OF THE UTILITY MODEL

An object of the utility model is to provide a face detection system to protect the lens, thereby improve the recognition effect.

For realizing the above-mentioned purpose, the utility model provides a face detection system, face detection system is including detecting body, lag, protection piece, ring gear and step motor, detect the body with lag fixed connection, it is located to detect the body the inside of lag, the lag has the opening, the protection piece with lag sliding connection, the protection piece is located the opening part, step motor fixed mounting is in the inside of lag, the ring gear with step motor's output fixed connection, one side of protection piece is provided with the tooth form groove, the ring gear with the tooth form groove agrees with mutually, the lag is right the detection body protects, prevents that impurity such as dust from falling on detecting the lens to improve recognition effect.

The face detection system further comprises a fixing frame, a supporting frame is arranged at the bottom of the protecting sleeve, the fixing frame is provided with a containing groove, the supporting frame is connected with the fixing frame in a sliding mode, and the bottom of the supporting frame is located inside the containing groove, so that the height of the detection body is changed.

The face detection system further comprises a fixing structure, the fixing structure comprises a movable column and a compression spring, the support frame is provided with an installation cavity, the fixing frame is provided with a plurality of clamping grooves at intervals along the length direction of the fixing frame, the two ends of the compression spring are respectively fixedly connected with the support frame and the movable column, the compression spring is located in the installation cavity, the movable column is connected with the support frame in a sliding mode, the movable column is close to one end of the compression spring is located in the installation cavity, the movable column is far away from one end of the compression spring is located in one of the clamping grooves, and therefore the support frame is fixed.

The fixing frame is provided with a plurality of anti-skid lugs, and the anti-skid lugs are distributed at the bottom of the fixing frame at intervals so as to increase the friction force between the fixing frame and the ground.

The outer side surface of the support frame is provided with a positioning block, the positioning block and the support frame are integrally formed, a positioning opening is formed in the inner side surface of the installation cavity in an extending mode along the length direction of the fixing frame, the positioning block is located in the positioning opening, and the support frame is prevented from rotating relative to the fixing frame.

The face detection system further comprises two U-shaped rings, the two U-shaped rings are fixedly connected with the fixing frame respectively, and the two U-shaped rings are located on two opposite side faces of the fixing frame respectively, so that the face detection system can be moved conveniently through the U-shaped rings.

The face detection system further comprises a mounting plate, the mounting plate is fixedly connected with the protecting sleeve, the mounting plate is located at the top of the protecting sleeve, and the mounting plate can be detached so as to be convenient for maintenance of the detection body.

The face detection system further comprises a cooling fan, the cooling fan is fixedly mounted on the mounting plate, the cooling fan is located inside the protection sleeve and can cool the detection body, and impurities such as dust can be prevented from entering the protection sleeve.

The utility model discloses a face detection system, when the inductor does not sense the human body, the protection piece is located open-ended inside is in order to right the opening shelters from, makes impurity such as dust can't get into the inside of lag is stained with and attaches on the detection lens of detection body, when the inductor senses the human body, step motor operation makes the protection piece is no longer right the opening shelters from, the detection lens of detection body discerns the people's face, and discernment is accomplished the back, step motor gyration makes the protection piece resets, the protection piece is right the opening is sheltered from to prevent impurity such as dust from being stained with and attaching on the detection lens of detection body, make the detection lens of detection body keeps comparatively clean and tidy state, thereby improves the recognition effect of this face detection system.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a cross-sectional view of a face detection system provided by the present invention.

Fig. 2 is a cross-sectional view taken along line a-a of fig. 1 according to the present invention.

Fig. 3 is a cross-sectional view taken at B-B of fig. 1 according to the present invention.

Fig. 4 is a partial enlarged view of the point C in fig. 1 according to the present invention.

Fig. 5 is a cross-sectional view taken at D-D of fig. 2 according to the present invention.

Fig. 6 is a partial enlarged view of the point E in fig. 2 according to the present invention.

1-detection body, 2-protective sleeve, 21-opening, 22-support frame, 221-installation cavity, 222-positioning block, 3-protective sheet, 31-tooth groove, 4-gear ring, 5-stepping motor, 6-fixing frame, 61-accommodating groove, 62-clamping groove, 63-anti-skid bump, 64-positioning opening, 7-fixing structure, 71-movable column, 72-compression spring, 8-U-shaped ring, 9-installation plate and 10-cooling fan.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present invention, and should not be construed as limiting the present invention.

In the description of the present invention, it is to be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are merely for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. In addition, in the description of the present invention, "a plurality" means two or more unless specifically limited otherwise.

Please refer to fig. 1 to 6, the utility model provides a face detection system, face detection system is including detecting body 1, lag 2, protection piece 3, ring gear 4 and step motor 5, detect body 1 with lag 2 fixed connection, it is located to detect body 1 the inside of lag 2, lag 2 has opening 21, protection piece 3 with lag 2 sliding connection, protection piece 3 is located opening 21 department, step motor 5 fixed mounting is in the inside of lag 2, ring gear 4 with step motor 5's output fixed connection, one side of protection piece 3 is provided with tooth shape groove 31, ring gear 4 with tooth shape groove 31 agrees with mutually.

In this embodiment, the detection part of the detection main body 1 uses a customized face detection library, and after a dll library is configured, a corresponding header file is imported into a project file, so that an interface of the library can be used, the library provides four different interfaces for different application occasions, and the detection main body 1 uses an interface with the best detection effect but the most time-consuming one: int & facedetect & multiview & reinformance (parameter list …), where the function form list includes a pointer pointing to the area storing the detection result, a pointer pointing to the area storing the data of the input picture, the length and width of the input picture and some other parameters, the function returns a pointer pointing to the area storing the detection result, and the key is to extract the data pointed by the pointer, assuming that the returned pointer is a pResult, the first integer data pointed by the pResult indicates the number of faces detected in the input picture, the next consecutive memory area stores face information of each person, each person occupies 142 integer data, the first four numbers indicate the left side of the top left corner and the bottom right corner of the face box, the fifth data indicates information named neighbor, the sixth data indicates the corner of the face, the range is 0-60 degrees, and then every two data form a coordinate point, marking 68 characteristic points of the human face, wherein the characteristic points mark the position and the contour of the contour, eyes, eyebrows, nose and mouth of the human face and occupy 136 data, and the detection based on monitoring only needs to take out pictures from a camera at certain time intervals, send the pictures into a detection function, draw a detection structure on the pictures and output the detection structure to a user interface; the method is characterized in that a FaceNet neural network model based on Tensorflow is adopted in the characteristic extracting and identifying part of the detection body 1, a Python API is adopted, functions which directly use Tensorflow are compiled by Python, the functions defined by Python are converted into C-form definition functions through Cython, Cython can automatically generate header files and source files corresponding to the functions, and the functions defined by the C-form can be used by importing the generated files into a project. All usage related to Tensflow is encapsulated in a class called TensflowInference, which encapsulates the operations including acquisition and maintenance of session objects, initialization of Python environment, loading of network model, computation mapping features and all C + + and Python data interworking, and completely hides all Python related contents from users outside the class. Two of these needs to be noted: (1) session object: it is worth noting that the computation in the Tensorflow is completed through a session object, the session object stores the structural information and weight information of the dataflow graph, the network model used here is very huge (141MB), if the model is called from a disk into a memory again in each computation, the time consumption of the program is seriously increased (about two minutes are consumed for loading the model each time), therefore, the reasonable design is that the model is loaded into the memory when the whole system is initialized, the session object is obtained to reside in the memory, and the member method of the session object is called to complete one computation (about 0.3-0.4 second is consumed for each computation). In Python scripts, the acquisition of a session object is acquired through an interface tf.session () provided by the tensflow, however, before the session object is acquired, the structure information of a data flow graph needs to be loaded into a default graph, and then the weight in the graph is restored to the session object, so that the subsequent session object can correctly execute the forward calculation of FaceNet. The obtained session object will appear in C + + in the type PyObject; (2) data interaction of C + + and Python: the data interaction mode is carried out through the form factor and the return value of the function, and mainly comprises two directions, namely flowing the data from the C + + part into Python, and returning the result of the calculation executed by the Python part to C + +. The detection body 1 is fixedly arranged inside the protection sleeve 2, a detection lens of the detection body 1 is aligned to the opening 21, the stepping motor 5 is fixedly arranged inside the protection sleeve 2, the gear ring 4 is fixedly connected with an output end of the stepping motor 5, the tooth-shaped groove 31 of the protection sheet 3 is matched with the gear ring 4, the protection sheet 3 is arranged at the opening 21, the gear ring 4 is driven to rotate by the stepping motor 5 when the stepping motor 5 operates, the protection sheet 3 is driven to move by the gear ring 4 so that the protection sheet 3 can shield the opening 21, an inductor is arranged inside the protection sleeve 2, when the inductor does not sense a human body, the protection sheet 3 is positioned inside the opening 21 to shield the opening 21, so that impurities such as dust and the like cannot enter the protection sleeve 2 and adhere to the detection lens of the detection body 1, when the sensor senses a human body, the stepping motor 5 operates to enable the protective sheet 3 to not shield the opening 21, the detection lens of the detection body 1 identifies the human face, after identification is completed, the stepping motor 5 rotates to enable the protective sheet 3 to reset, the protective sheet 3 shields the opening 21 to prevent impurities such as dust from being attached to the detection lens of the detection body 1, so that the detection lens of the detection body 1 keeps a tidy state, and the identification effect of the human face detection system is improved.

Further, the human face detection system further comprises a fixing frame 6, a supporting frame 22 is arranged at the bottom of the protecting sleeve 2, the fixing frame 6 is provided with an accommodating groove 61, the supporting frame 22 is connected with the fixing frame 6 in a sliding manner, and the bottom of the supporting frame 22 is located inside the accommodating groove 61;

the face detection system further comprises a fixing structure 7, the fixing structure 7 comprises a movable column 71 and a compression spring 72, the support frame 22 is provided with an installation cavity 221, the fixing frame 6 is provided with a plurality of clamping grooves 62 at intervals along the length direction of the fixing frame, two ends of the compression spring 72 are respectively fixedly connected with the support frame 22 and the movable column 71, the compression spring 72 is located inside the installation cavity 221, the movable column 71 is slidably connected with the support frame 22, one end of the movable column 71, which is close to the compression spring 72, is located inside the installation cavity 221, and one end of the movable column 71, which is far away from the compression spring 72, is located inside one of the clamping grooves 62;

the fixed frame 6 is provided with a plurality of anti-skid lugs 63, and the anti-skid lugs 63 are distributed at the bottom of the fixed frame 6 at intervals;

a positioning block 222 is arranged on the outer side surface of the support frame 22, the positioning block 222 and the support frame 22 are integrally formed, a positioning opening 64 is formed in the inner side surface of the mounting cavity 221 in an extending manner along the length direction of the fixing frame 6, and the positioning block 222 is located inside the positioning opening 64;

the face detection system further comprises two U-shaped rings 8, the two U-shaped rings 8 are fixedly connected with the fixing frame 6 respectively, and the two U-shaped rings 8 are located on two opposite side faces of the fixing frame 6 respectively.

In the present embodiment, the supporting frame 22 can slide up and down inside the receiving groove 61, so that the position of the detecting body 1 can be adjusted up and down, thereby increasing the practicability of the face detecting system, the supporting frame 22 and the fixing frame 6 are fixed by the fixing structure 7, in the case of no external force, one end of the movable post 71 away from the compression spring 72 is pushed out of the mounting cavity 221 by the compression spring 72 and is located inside one of the engaging grooves 62, thereby fixing the supporting frame 22, when the position of the detecting body 1 needs to be adjusted, the movable post 71 is pressed into the mounting cavity 221, the movable post 71 completely enters the mounting cavity 221, so that the supporting frame 22 can slide inside the receiving groove 61, and after the position of the supporting frame 22 is moved, the movable column 71, under the support of the compression spring 72, comes out from the inside of the mounting cavity 221 and enters the corresponding engaging groove 62, so as to fix the supporting frame 22, so as to fix the position of the detection body 1, and the bottom of the fixing frame 6 is provided with a plurality of the protection protrusions arranged at intervals, so as to increase the friction force between the fixing frame 6 and the ground, so as to improve the anti-skid performance of the face detection system, the outer side surface of the supporting frame 22 is provided with the positioning block 222, the positioning block 222 can slide inside the positioning opening 64, the positioning block 222 is engaged inside the positioning opening 64 so as to limit the supporting frame 22 from rotating relative to the fixing frame 6, the U-shaped rings 8 are arranged on the two opposite outer side surfaces of the fixing frame 6, so that a worker can move the fixing frame 6 through the U-shaped rings 8, so that the worker can change the position of the face detection system.

Further, the face detection system further comprises an installation plate 9, the installation plate 9 is fixedly connected with the protection sleeve 2, and the installation plate 9 is located at the top of the protection sleeve 2;

the face detection system further comprises a heat radiation fan 10, the heat radiation fan 10 is fixedly installed on the installation plate 9, and the heat radiation fan 10 is located inside the protection sleeve 2.

In this embodiment, the mounting plate 9 is fixedly connected to the protecting cover 2 through a plurality of screws, so that the mounting plate 9 can be disassembled to facilitate the maintenance or replacement of the detecting body 1 installed inside the protecting cover 2, the mounting plate 9 is fixedly provided with the heat dissipating fan 10, the heat dissipating fan 10 can dissipate heat of the detecting body 1, so that the detecting body 1 can better operate, and the wind direction of the heat dissipating fan 10 faces the opening 21, so that impurities such as dust are more difficult to enter the protecting cover 2 to adhere to the detecting lens of the detecting body 1, thereby improving the recognition effect of the face detecting system.

While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims

1. A face detection system, characterized in that,

face detection system is including detecting body, lag, protection piece, ring gear and step motor, detect the body with lag fixed connection, it is located to detect the body the inside of lag, the lag has the opening, the protection piece with lag sliding connection, the protection piece is located the opening part, step motor fixed mounting be in the inside of lag, the ring gear with step motor's output fixed connection, one side of protection piece is provided with the tooth form groove, the ring gear with the tooth form groove agrees with mutually.

2. The face detection system of claim 1,

the face detection system further comprises a fixing frame, a supporting frame is arranged at the bottom of the protecting sleeve, the fixing frame is provided with a containing groove, the supporting frame is in sliding connection with the fixing frame, and the bottom of the supporting frame is located inside the containing groove.

3. The face detection system of claim 2,

face detection system still includes fixed knot structure, fixed knot constructs including activity post and compression spring, the support frame has the installation cavity, the mount is provided with a plurality of block grooves along self length direction interval, compression spring's both ends respectively with the support frame with activity post fixed connection, compression spring is located the inside of installation cavity, the activity post with support frame sliding connection, the activity post is close to compression spring's one end is located the inside of installation cavity, the activity post is kept away from compression spring's one end is located one of them the inside in block groove.

4. The face detection system of claim 3,

the fixing frame is provided with a plurality of anti-skid lugs, and the anti-skid lugs are distributed at the bottom of the fixing frame at intervals.

5. The face detection system of claim 4,

the lateral surface of support frame is provided with the locating piece, the locating piece with support frame integrated into one piece, edge on the medial surface of installation cavity the length direction extension of mount is provided with the location mouth, the locating piece is located the inside of location mouth.

6. The face detection system of claim 5,

the face detection system further comprises two U-shaped rings, the two U-shaped rings are fixedly connected with the fixing frame respectively, and the two U-shaped rings are located on two opposite side faces of the fixing frame respectively.

7. The face detection system of claim 6,

the face detection system further comprises an installation plate, the installation plate is fixedly connected with the protection sleeve, and the installation plate is located at the top of the protection sleeve.

8. The face detection system of claim 7,

the face detection system further comprises a cooling fan, the cooling fan is fixedly installed on the installation plate, and the cooling fan is located inside the protection sleeve.