KR100983346B1 - System and method for recognition faces using a infra red light - Google Patents

Abstract

The present invention relates to a face recognition system and method using IR illumination, and more particularly to a face recognition system and method that is robust to the external environment by applying the IR illumination.

In the face recognition system using IR illumination according to an embodiment of the present invention, the IR illumination unit for irradiating light in the infrared region, the image acquisition to obtain an input image including a predetermined face by the irradiated light using an IR filter A feature information extracting unit extracting feature information of a predetermined face included in the input image, and a face recognizing a face of the input image by comparing the extracted feature information with feature information of a person image stored in a database. It includes a recognition unit.

Description

System and method for recognition faces using a infra red light}

The present invention relates to a face recognition system and method using IR illumination, and more particularly to a face recognition system and method that is robust to the external environment by applying the IR illumination.

As information society develops, identification technology for identifying people is becoming important, and biometric technology using human features for personal information protection and identification using computers is being researched. Among the biometric technologies, the face recognition technology may be convenient to identify the user in a non-contact manner, unlike a recognition technology requiring a user's special operation or action such as fingerprint recognition and iris recognition.

Face recognition technology is one of the core technologies for multimedia database search, and can be used for video summary, face search, security, and surveillance system using face information.

However, in the general face recognition technique, the recognition result is sensitively changed depending on images captured at various angles, various expressions, lighting, and image processing for the same person, thereby degrading performance.

Accordingly, there is a need for a system and method for performing face recognition reliably by performing face recognition robustly in a surrounding situation in face recognition.

An object of the present invention is to provide a face recognition system and method that is robust to external illumination by illumination of irradiating light in the infrared region or image acquisition filtering light in the infrared region in image acquisition.

Problems to be solved of the present invention are not limited to the above-mentioned problems, and other tasks not mentioned will be clearly understood by those skilled in the art from the following description.

One aspect of the face recognition system using the IR illumination of the present invention to achieve the problem to be solved is an IR illumination unit for irradiating light in the infrared region; An image acquisition unit for acquiring an input image including a predetermined face by the irradiated light using an IR filter; A feature information extracting unit which extracts feature information about a predetermined face included in the input image; And a face recognition unit configured to recognize the face of the input image by comparing the extracted feature information with feature information of the person image stored in the database.

Another aspect of the face recognition system using the IR illumination of the present invention to achieve the problem to be solved is an illumination intensity sensing unit for sensing the external illumination; An image acquisition unit for obtaining an input image by passing light of an area having a different wavelength according to the sensed illuminance; A feature information extracting unit which extracts feature information about a predetermined face included in the input image; And a face recognition unit configured to recognize the face of the input image by comparing the extracted feature information with feature information of the person image stored in the database.

One aspect of the face recognition method using the IR illumination of the present invention to achieve the problem to be solved is the step of irradiating light in the infrared region; Acquiring an input image including a predetermined face by the irradiated light using an IR filter; Extracting feature information about a predetermined face included in the input image; And comparing the extracted feature information with feature information of a person image stored in a database to recognize a face of the input image.

Another aspect of the face recognition method using the IR illumination of the present invention to achieve the problem to be solved is the step of sensing the external illumination; Acquiring an input image by passing light of an area having a different wavelength according to the sensed illuminance; Extracting feature information about a predetermined face included in the input image; And comparing the extracted feature information with feature information of a person image stored in a database to recognize a face of the input image.

Advantages and features of the present invention and methods for achieving them will be apparent with reference to the embodiments described below in detail with the accompanying drawings. However, the present invention is not limited to the embodiments set forth below, but may be implemented in various forms, and only the present embodiments make the disclosure of the present invention complete, and those of ordinary skill in the art to which the present invention belongs. It is provided to fully inform the person having the scope of the invention, the invention is defined only by the scope of the claims.

Hereinafter, the present invention will be described with reference to the drawings for a block diagram or a processing flowchart for explaining a face recognition system and method using IR illumination according to embodiments of the present invention. At this point, it will be understood that each block of the flowchart illustrations and combinations of flowchart illustrations may be performed by computer program instructions. Since these computer program instructions may be mounted on a processor of a general purpose computer, special purpose computer, or other programmable data processing equipment, those instructions executed through the processor of the computer or other programmable data processing equipment may be described in flow chart block (s). It creates a means to perform the functions. These computer program instructions may be stored in a computer usable or computer readable memory that can be directed to a computer or other programmable data processing equipment to implement functionality in a particular manner, and thus the computer usable or computer readable memory. It is also possible for the instructions stored in to produce an article of manufacture containing instruction means for performing the functions described in the flowchart block (s). Computer program instructions may also be mounted on a computer or other programmable data processing equipment, such that a series of operating steps may be performed on the computer or other programmable data processing equipment to create a computer-implemented process to create a computer or other programmable data. Instructions for performing the processing equipment may also provide steps for performing the functions described in the flowchart block (s).

In addition, each block may represent a portion of a module, segment, or code that includes one or more executable instructions for executing a specified logical function (s). It should also be noted that in some alternative implementations, the functions noted in the blocks may occur out of order. For example, the two blocks shown in succession may in fact be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending on the corresponding function.

The term '~' or 'module' used in this embodiment refers to software or a hardware component such as an FPGA or an ASIC, and the '~' or 'module' plays certain roles. However, '~' or 'module' is not meant to be limited to software or hardware. The 'unit' or 'module' may be configured to be in an addressable storage medium or may be configured to play one or more processors. Thus, as an example, a 'unit' or 'module' may include components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and the like. Procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. Functions provided within components and 'parts' or 'modules' may be combined into a smaller number of components and '~ parts' or 'modules' or additional components and '~ parts' or 'modules' Can be further separated into '.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a face recognition system using IR illumination according to an embodiment of the present invention. Referring to FIG. 1, the face change detection system 100 according to an exemplary embodiment of the present invention may include an IR illuminator 110, an image acquirer 120, a face detector 130, a feature information extractor 160, and a face. The recognition unit 170 and the database 300 may be included.

The IR illuminator 110 serves to irradiate light in the infrared region. According to an embodiment of the present invention, an input image including a predetermined face is obtained by irradiating light in an infrared region. For example, in order to minimize the influence of the visible light band in obtaining an image in the IR band, infrared light of a specific wavelength band (850 nm to 950 nm) may be irradiated artificially. In this case, the face area is the exposure center of the illumination, and the intensity of the illumination is adjusted so that saturation of data does not occur by the illumination.

As described above, by using the light in the infrared region, it is possible to implement a stable face recognition system having less influence on artificial lighting changes such as sunlight or fluorescent light. In addition, the face recognition system may be operated regardless of the external environment by acquiring an image by the light in the infrared region with respect to the external environment that varies depending on the time zone at the entrance or the entrance.

The image acquisition unit 120 acquires an input image from the outside. The image acquirer 120 may acquire an image captured by infrared illumination. Alternatively, the image acquisition unit 120 may further include an IR (Infra Red) filter (not shown) that filters the light in the infrared region while acquiring a general image.

For example, when an input image is acquired by an image input sensor, an input image may be obtained by converting an image signal of a subject incident through a predetermined lens into an electrical signal. Here, the image input sensor may include a charge coupled device (CCD), a CMOS, and other image acquisition means known in the art. In addition, an analog / digital converter for converting the electrical signal obtained by the image input sensor into a digital signal and a digital signal processor (DSP) for receiving a digital signal converted by the analog / digital converter and processing the image signal It is possible to obtain a predetermined input image by the.

On the other hand, the IR filter (not shown) serves to pass only the light of a predetermined infrared region of the light obtained by the image input sensor. For example, the IR filter may serve to pass light in the infrared spectrum in the wavelength range of 700 nm or more. The IR filter may block light in the visible light region irradiated by the external environment and pass the light irradiated by the IR illuminating unit 110.

The face detector 130 extracts a face region from the input image as a face image. The face detector 130 may extract an eye, a nose, a mouth, etc. which are specific components in the face after detecting the approximate face from the input image, and extract the face region based on the extracted face. For example, if the position of both eyes is detected, the distance of both eyes can be obtained. The face detector 130 may extract a face region from the input image as a face image based on the distance between the two eyes, thereby reducing the influence of the background of the input image or the change of the human hair style. In addition, the face detector 130 may normalize the size of the face region using the extracted face region information. By normalizing the size of the face region, unique features such as distance between two eyes and distance between eyes and nose in the face region can be calculated at the same scale level.

The feature information extractor 160 may extract feature information about the face image from the face image. Here, the feature information refers to information representing unique characteristics of a predetermined face included in the face image. For example, the feature information may be selected by using Principal Component Analysis (PCA), Linear Discriminate Analysis (LDA), and the like.

As another example, 2-Dimensional Principal Component Analysis (2D-PCA) may be used to extract feature information of a face image based on a two-dimensional image matrix. Since the size of the projection matrix is smaller than that of the general PCA method, the 2D-PCA technique can be easily applied to portable devices, low-end embedded systems, etc., because the amount of computation and required memory can be relatively low. Therefore, face recognition can be performed within a short time, so that face recognition can be performed in real time.

The face recognition unit 170 compares the extracted feature information with feature information of the person image stored in the database and recognizes the face of the input image. The face recognition unit 170 may recognize the face of the person image and the face in the input image as the face of the same person when the distance difference between the extracted feature information and the feature information of the person image stored in the database is within a threshold.

The database 300 stores characteristic information of the plurality of person images and / or person images. Here, the person image refers to a face image matching a predetermined person. Alternatively, the database may receive a plurality of person images from outside, extract feature information about each person image, and store the feature information. The feature information of each person image may be calculated in advance or may be calculated at the time of input according to an embodiment of the present invention and stored in a database.

The database 300 may provide feature information of the plurality of person images and / or the plurality of person images to the face recognition unit. Meanwhile, the database 300 may further store personal information such as a name, age, address, and the like of a registered person, a security level, and the like.

For example, in the environment where night and day are continuously changed or when there is a great influence on external lighting such as indoors, the image acquisition unit 120 acquires an image with uneven illuminance according to the lighting environment to detect the face. It may not be right. However, according to an embodiment of the present invention, the IR illumination unit 110 irradiating the light in the infrared region may acquire the image by the light in the infrared region irrespective of the brightness of the outside, thereby stably face recognition.

2A schematically illustrates a face recognition system using IR illumination according to an embodiment of the present invention, and FIG. 2B shows a circuit diagram of the IR illumination unit of FIG. 2A.

Referring to FIG. 2A, a face recognition system using IR illumination according to an embodiment of the present invention includes an IR illumination unit 110, an image acquisition unit 120, and a face recognition processing unit 200 embedded in the system. ) May be included. In this case, the face recognition processor 200 includes the face detector 130, the feature information extractor 160, and the face recognizer 170 as described above with reference to FIG. 1. An image may be processed to recognize a predetermined face.

The IR illumination unit 110 may be configured of a plurality of IR LEDs 115 formed in a circular shape surrounding the image acquisition unit 120. The IR LED 115 is a light emitter made using a light emitting diode (LED), and serves to irradiate light in the infrared region. The IR LED 115 has a low power consumption characteristic of the light emitting diode, excellent durability, and can be mounted and operated in a compact space such as a face recognition system. On the other hand, the image acquisition unit 120 will be described in more detail later.

Referring to FIG. 2B, a predetermined source voltage Vcc is applied at one end of the IR illuminator 110. When a predetermined gate voltage is applied from the embedded face recognition processor 200, the source current is grounded. While passing, the IR LED 115 can irradiate infrared light having a predetermined wavelength band or a specific wavelength.

FIG. 3A schematically illustrates an image input unit in a face recognition system using IR illumination according to an embodiment of the present invention, and FIG. 3B schematically illustrates a bandwidth of light input to the image input unit of FIG. 3A.

Referring to FIG. 3A, an image acquirer 120 according to an embodiment of the present invention may include a lens 122, an image input sensor 125, and an IR filter 127. The lens 122 receives light from the outside, reduces the external image, and transmits the image to the image input sensor 125. The image input sensor 125 converts the light reflected from the subject into an electrical image signal to generate an input image. The IR filter 127 passes light in the infrared region and blocks light in the visible and ultraviolet regions. The IR filter 127 is a bandpass filter and transmits light belonging to a wavelength of a specific region to the image input sensor 125 by passing only light belonging to a predetermined wavelength band.

Referring to FIG. 3B, a wavelength region of light transmitted through the IR filter 127 is illustrated. Referring to FIG. 3B, the light passing through the IR filter 127 has zero or extremely low transmittance for visible and ultraviolet regions. However, in the infrared region, the transmittance of light rapidly decreases for a certain wavelength band and then decreases. For example, it may serve as a band pass filter allowing transmission of infrared light in a specific wavelength band of 850 nm to 950 nm.

As described above, in an embodiment of the present invention, the IR LED 115 and the IR filter 127 are introduced to acquire and process a predetermined input image including a face, thereby affecting sensitive external lighting in a general visible light region. It is possible to increase the reliability of the face recognition system by acquiring a face image by acquiring an input image having a predetermined illuminance consistently without being received.

4 is a block diagram of a face recognition system using IR illumination according to another embodiment of the present invention. 4, the facial recognition system 400 using IR illumination according to another embodiment of the present invention includes an IR illumination unit 110, an illumination intensity detector 410, a first image acquirer 420, and a second image. It may include an acquirer 430, a face detector 130, a feature information extractor 160, a face recognizer 170, and a database 300. Here, the IR illuminator 110, the face detector 130, the feature information extractor 160, the face recognition unit 170, and the database 300 have been described in detail with reference to FIG. 1, and thus will be omitted.

The illuminance sensor 410 detects illuminance in the surrounding environment of the face recognition system 400 when an image is acquired. The illuminance sensor 410 may detect illuminance of the surrounding environment by a predetermined illuminance sensor (not shown). The illuminance detector 410 transmits an image acquisition command to the first image acquirer 420 when the detected illuminance is within a predetermined range, and the second image when the detected illuminance is outside the predetermined range. The image acquisition command is transmitted to the acquirer 430. Here, the predetermined range refers to a range of illuminance in which a general image acquisition sensor acquires a high quality input image by an average ambient brightness. For example, it refers to a predetermined illuminance range in which features such as an eye line or a mouse line are extracted relatively sharply from an input image, or a 50 to 150 Lux range that operates properly even in a general face recognition system.

For example, when the detected illuminance is obtained during general indoor or daytime, the input image may be acquired by the first image acquisition unit 420 having a general image acquisition sensor without using a specific light. have. Here, the general image acquisition sensor receives the visible light region to receive the input image.

On the other hand, when the detected illuminance is darker than a normal room or becomes brighter than a predetermined threshold, the IR illuminator 110 irradiates the infrared light, and the second image acquirer 430 having the IR filter is predetermined. An input image can be obtained.

The face included in the predetermined input image may be recognized by processing the obtained input image through the face detector 130, the feature information extractor 160, and the face recognizer 170.

As described above, by detecting the illuminance around the face recognition system 400 by the illuminance sensor, the input image is acquired according to the sensed illuminance and used for face recognition, thereby making it possible to reliably face recognition. Therefore, it is possible to implement a face recognition system that is relatively robust to the surrounding environment.

In another exemplary embodiment, face recognition may be performed by simultaneously operating or sequentially operating the first image acquisition unit 420 and the second image acquisition unit 430 to acquire input images, respectively. Therefore, when face recognition is performed, a relatively familiar interface may be provided by outputting and displaying the input image acquired by the first image acquisition unit 420 to the user.

As another embodiment, face recognition may not be performed properly when face recognition is performed by an input image acquired by the first image acquisition unit 420. In this case, when it is determined that the face recognition error or the face recognition failure is performed, the face recognition unit 170 causes the IR illumination unit 110 and the second image acquisition unit 430 to acquire the input image. By performing face recognition again based on the input image acquired by the light in the infrared region, even if the face recognition is not properly performed by the external environment, the face recognition can be stably performed.

5 is a block diagram of a real recognition unit for real recognition in the face recognition system using IR illumination according to an embodiment of the present invention. The face recognition system using the IR illumination according to an embodiment of the present invention may recognize the real object by detecting eye blink or mouth closing with respect to the face image extracted from the face detector 130. This is to secure the real-time recognition and security by operating in conjunction with the real recognition in face recognition.

Referring to FIG. 5, the real object recognizer 500 extracts an eye region from a plurality of input images or a face image extracted from the face detector 130, and the size of the pupil in the extracted eye region. It may include a flicker detection unit 520 for detecting the eye blink using a change or a change in the eye-line (eye-line).

The eye region extractor 510 extracts an eye region from the face image or the converted image. Here, the eye region includes an eye line and may include an eye outline formed by the eye line.

As an example of a method of extracting an eye region, the eye region may be extracted using an eye-specific outline or an eye-specific brightness. For example, eye-specific contours can be extracted using a pattern of common eye-lines. When a similar line is extracted from the face image or the converted image using a plurality of general eye-line samples, this may be used as an eye region. As such, since the shape of the eye has a geometrically constant shape, the eye region may be extracted by the variable template technique.

As another example, the brightness of the eye region may be darker than other parts of the face image due to the pupil. Based on the property of a relatively dark area, the eye area may be extracted by calculating an average brightness for any number of pixels (for example, 5 × 5 pixels). Therefore, in the face image using the contrast information, two eye regions appear in isolation, and the eye regions can be easily extracted.

Meanwhile, the eye region extraction technique is not limited to the above technique, and various known techniques may be applied to extract the eye region.

The flicker detection unit 520 detects eye flicker by using a change in eye size or a change in eye-line in the extracted eye region. In the state where the eye region is extracted with respect to the plurality of face images or the converted images, the blink detector 520 detects the shape of the pupil and the eye-line of the extracted eye region, so that the pupil is not detected and the upper line of the eye-line is not detected. Judging when the interval between the and the lower line is narrow, eyes are judged, and when the pupil is detected and the interval between the upper line and the lower line of the eye line is maintained by the experimental value Can be. Alternatively, when the change in the area of the eye contour formed by the eye-line is equal to or greater than a certain amount or the area change ratio is equal to or greater than the threshold, it may be determined that the eye blinks.

As described above, the person included in the input image may be determined to be a real person by detecting eye blinks. Alternatively, when the number of blinks is greater than or equal to the threshold, it may be determined as the real thing.

Referring to FIG. 5 again, the real object recognizer 500 may include a mouth region extractor 530 which extracts a mouth region from a face image or a converted image, and upper and lower portions of a mouth-line in the extracted mouth image. It may include an opening and closing detection unit 540 for detecting the mouth opening and closing using the position.

The mouth region extractor 530 extracts a mouth region from a face image or a converted image, and as an example of a method of extracting a mouth region, a technique similar to a technique of extracting an eye region may be applied. Here, the mouth region includes a mouth-line formed by the lips and may include a mouth contour formed by the mouth-line.

The opening detection unit 540 determines whether the mouth is open or closed using the upper and lower positions of the mouse line in the mouth region. Accordingly, when the mouth line is closed and the open state is detected by comparing the mouse lines in the mouth regions respectively extracted from the plurality of face images or the converted images, it may be determined that the mouth is closed. Here, the mouse-line refers to the line which abuts each other when the mouth is closed by the upper and lower lips.

As described above, the person included in the input image may be determined to be a real person by detecting mouth opening and closing. Alternatively, it may be determined as a real object when the number of opening and closing is greater than or equal to a threshold.

According to an embodiment of the present invention, the face recognition robust to the surrounding environment may be processed by the IR illumination unit 110 and the face recognition robust to the illumination of the surrounding environment may be processed by software according to the input image processing.

6A and 6B are exemplary views to which the LBP technique is applied in the transform image acquisition unit in the face recognition system according to an embodiment of the present invention.

6A and 6B, a process of obtaining an LBP pattern based on pixel values of adjacent pixels 142 around a predetermined center pixel 141 in a face image is shown.

The LBP technique is as follows.

Here, LBP (x _c , y _c ) represents the LBP transformed value at the (x _c , y _c ) center pixel 330, and g _c is the face at the (x _c , y _c ) center pixel 141. It is a pixel value of an image, and g _p represents the pixel value in the peripheral pixel 142 of (x _c , y _c ). Herein, the pixel value may represent an intensity value when the gray value is converted into a gray scale image or a single channel image. At the same time, p increments by one, turning clockwise or counterclockwise with respect to a point in the (x _c , y _c ) pixel.

For example, if the LBP technique is applied to a 3 × 3 pixel area, a binary pattern 143 having a binary value of 1 or 0 is compared by comparing eight peripheral pixel values based on corresponding pixel values of a face image. You can get it. Therefore, when the LBP technique is applied to a 3 × 3 pixel region, a binary pattern 143 consisting of eight binary numbers can be obtained, and the binary pattern 143 becomes a pattern value 144 of the corresponding pixel. Can be. Therefore, the binary image 143 may be obtained for each pixel of the face image, and the converted image may be calculated by substituting the pattern value 144 for each pixel.

As shown in FIG. 6B, the pixel value of the center pixel 141 is 28 in the 3 × 3 pixel area, and has a value of 1 when the surrounding pixels 142 are greater than 28, and a value of 0 when it is less than or equal to 28. . Therefore, the pixel value is rotated clockwise around the first peripheral pixel 145 having the pixel value of 250, and the pixel value is compared to the eighth peripheral pixel through the 8-bit binary pattern (1, 1, 0, 1, 0). , 0, 0, 1). If the binary pattern (1, 1, 0, 1, 0, 0, 0, 1) is identified as a binary number and converted into a decimal number, the pattern value 144 of 209 can be obtained with respect to the center pixel 141. .

As described above, by applying the LBP technique to each pixel of the face image, a predetermined pattern value is derived for each pixel, thereby obtaining a converted image composed of the pattern values. Since the converted image is determined by the relationship between the pixel value or the brightness between the pixel and the surrounding pixels, the influence of the overall lighting or interference can be reduced, so that the face recognition can be stably performed even for the bright image or the dark image as a whole. Can be done.

7 is a diagram schematically illustrating a process of calculating a projection matrix by a feature information extracting unit in a face recognition system according to an embodiment of the present invention.

As shown in FIG. 7, according to the 2D-PCA technique, an image matrix B _j may be added to be averaged to obtain a covariance matrix G _t , and a projection matrix X including eigenvectors of the obtained covariance matrix may be obtained. .

The image matrix B _j may be obtained by the following equation.

는 M개의 샘플영상의 평균영상이다. 각 샘플영상에 대하여 평균영상을 빼 준 행렬에 연산을 수행함으로써 각 샘플영상에 대한 영상행렬(B_j)을 구할 수 있다. 한편, 여기서 각 샘플영상은 임의의 얼굴영상으로서, 임의의 얼굴영상으로부터 LBP 기법에 의해 변환된 변환영상이 사용될 수도 있다.Here, B _j is the image matrix for the j- th sample image, A _j is the j- th sample image in the M sample images,

Is an average image of M sample images. An image matrix B _j for each sample image may be obtained by performing an operation on a matrix obtained by subtracting an average image from each sample image. Meanwhile, each sample image may be an arbitrary face image, and a converted image converted by the LBP technique from an arbitrary face image may be used.

Based on the obtained image matrix B _j , a covariance matrix G _t may be obtained by the following equation.

Here, G _t is a covariance matrix, M is the number of sample images, and B _j is a j-th image matrix in M sample images. Therefore, after generating M respective image matrices for M sample images, the covariance matrix can be obtained by averaging them.

Therefore, eigenvectors of the covariance matrix maximizing the total scatter from which the sample images are projected can be obtained by the following equation.

X ₁ , ..., X _d are eigenvectors of the covariance matrix, and d is an index of the maximum eigenvector. Accordingly, the projection matrix X may be generated by using each eigenvector as a column vector.

Therefore, by multiplying the given transformation image by the projection matrix X, the feature information Y consisting of d projection feature vectors Y _k can be obtained.

Here, Y is feature information of a face image or a converted image, and may be referred to as a feature matrix for a given image. The feature information Y may be represented by Y ₁ , ..., Y _d consisting of column vectors. In addition, F is a face image matrix representing a face image with respect to a given input image, and when a predetermined face image is given, feature information about the face image may be obtained.

As described above, calculating the feature information by applying the 2D-PCA technique may reduce the computational and memory requirements by obtaining a projection matrix based on a small covariance matrix G _t . In addition, in general PCA, a covariance matrix G is based on a two-dimensional image matrix B _j in one embodiment of the present invention, compared to aligning a one-dimensional vector from an input image and extracting feature points based on the vector. _t ) and the projection matrix can be significantly reduced in the size of the matrix, thereby reducing the time required for computation, which can be applied to face recognition and low-end systems in real time.

Meanwhile, the face recognizing unit 500 may recognize the face by comparing the feature information obtained by Equation 5 with the feature information about the registered person registered in the database.

The comparison of the feature information may be based on the following equation.

)및 데이터베이스에 등록된 등록인물에 대한 특징정보(

)와의 거리를 비교하여 유사도를 측정하고, 상기 유사도가 임계치 범위 내인 경우 동일인물로 인식 할 수 있다.Therefore, as shown in Equation 6, the characteristic information of the converted image obtained from the input image (

And feature information about registered persons registered in the database (

The similarity is measured by comparing the distance with), and if the similarity is within the threshold range, it can be recognized as the same person.

As described above, according to the exemplary embodiment of the present invention, face recognition may be robust against the influence of lighting and the like, and it may be easily applied to a low specification system or an embedded system by reducing the computation time and the size of the memory.

FIG. 8 is a view comparing a face rejection rate (FRR) in general face recognition and face recognition according to an embodiment of the present invention. Here, the FRR (False Reject Rate) indicates a recognition rate that the face of the input image is misjudged or not determined by others.

Referring to FIG. 8, in the general face recognition system, the change of the face recognition rejection rate is relatively large according to the change of illuminance. For example, the face recognition rejection rate was significantly higher than 50% in the low illumination section (0 ~ 10 Lux, 300 ~ 5000 Lux), and the face recognition rejection rate was 100% especially in the high illumination section of 500 Lux or higher. Face recognition is not possible at all.

In an embodiment of the present invention, a bandpass filter for transmitting light having a wavelength of 850 nm to 950 nm is used as an IR filter. It can be seen that the face recognition rejection rate is 0% for almost all illuminances when applying an IR filter that passes light having a wavelength of 850 nm to 950 nm. It can be seen that the effect of external illumination is significantly reduced by applying an IR filter that passes light of 850 nm to 950 nm. More specifically, even if the external illuminance increases from 0, the facial recognition rejection rate up to 1000 Lux is almost 0%. In addition, even if it corresponds to 3000 Lux, 5000 Lux it can be seen that the face recognition rejection rate is less than 10%. It can be seen that the face recognition rejection rate is significantly lower than the relatively common face recognition system.

In addition, although the face recognition rejection rate of the general face recognition system is considerably high in a low illuminance region and a high illuminance region, the face recognition system according to an embodiment of the present invention acquires an input image by using IR illumination and an IR filter, and thus faces By performing the recognition, the face recognition rejection rate is significantly lowered regardless of the external lighting.

9 is a flowchart illustrating a face recognition method using IR illumination according to an embodiment of the present invention. Referring to FIG. 9, first, the IR illumination unit 110 irradiates light having a wavelength in the infrared region (S910). The image acquisition unit 120 acquires an input image including a predetermined face by using an IR filter (S920). The image acquisition unit 120 may acquire a plurality of input images in order to determine whether a real object is present.

Meanwhile, the external illuminance may be sensed and the input image may be acquired differently according to the first image acquirer 420 or the second image acquirer 430. In this case, the input image may be acquired by receiving light from a visible light region when the light intensity is within a predetermined illuminance range according to the detected illuminance range, and the IR image may be obtained by applying an IR filter when it is out of the predetermined range. .

When the input image is obtained, the face region is extracted from the input image (S930). A predetermined pixel area including a face may be extracted from the input image as a predetermined face image. Extracting such a face region may be performed by a technique of extracting a general face region.

Face recognition is performed on the faces included in the input image. Among the plurality of face images, a binary pattern is obtained by comparing pixel values of neighboring pixels with respect to a predetermined face image, and a converted image is calculated based on the binary pattern (S940). In order to increase the person recognition rate, a face recognition may be performed by selecting an image having a relatively clear eye line from a plurality of face images. However, the process of calculating the converted image may be omitted for the input image obtained under an environment in which noise of an external environment such as lighting is not significant.

Feature information is extracted from the extracted face region or the converted image (S960). PCA or LDA techniques may be applied to extract feature information. In addition, 2D-PCA can be applied to feature information extraction.

For example, according to the 2D-PCA method, each image matrix of each sample image is calculated based on a plurality of sample images, and a projection matrix including eigenvectors of the covariance matrix obtained by averaging each image matrix is obtained. Can be. The calculated projection matrix and the face image may be calculated to extract feature information about the face image or the converted image.

The face of the face image is recognized by comparing the feature information of the extracted face image and the registered person information (S970). When the distance between the feature information of the face image and each feature information included in the registered person information is calculated and positioned within a distance within a threshold value, the face in the input image may be determined to be the face of the same person as the registered person.

As described above, according to an embodiment of the present invention by irradiating the light of the wavelength of the infrared region by the IR illumination unit 110, when obtaining the image by applying only the IR wavelength band of light to obtain the image Stable face recognition can be achieved despite external lighting changes.

In the above examples, the term 'embedded system' or 'embedded system' is used for some explanations, but the example of the present invention is not limited to the embedded system, but includes a PC including a mobile / embedded system. It should be understood that it can be applied to various computer computing devices such as server computers.

Although embodiments of the present invention have been described above with reference to the accompanying drawings, those skilled in the art to which the present invention pertains may implement the present invention in other specific forms without changing the technical spirit or essential features thereof. You will understand that. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive.

1 is a block diagram of a face recognition system using IR illumination according to an embodiment of the present invention.

2A is a view schematically showing a face recognition system using IR illumination according to an embodiment of the present invention.

FIG. 2B is a diagram illustrating a circuit diagram of the IR illumination unit of FIG. 2A.

3A is a view schematically showing an image input unit in a face recognition system using IR illumination according to an embodiment of the present invention.

FIG. 3B is a diagram schematically illustrating a bandwidth of light input to the image input unit of FIG. 3A.

4 is a block diagram of a face recognition system using IR illumination according to another embodiment of the present invention.

5 is a block diagram of a real object recognition unit for real object recognition in a face recognition system using IR illumination according to an embodiment of the present invention.

6A and 6B are exemplary views to which the LBP technique is applied in the transform image acquisition unit in the face recognition system according to an embodiment of the present invention.

7 is a diagram schematically illustrating a process of calculating a projection matrix by a feature information extracting unit in a face recognition system according to an exemplary embodiment of the present invention.

FIG. 8 is a view comparing a face rejection rate (FRR) in general face recognition and face recognition according to an embodiment of the present invention.

9 is a flowchart of a face recognition method using IR illumination according to an embodiment of the present invention.

<Explanation of symbols on main parts of the drawings>

110: IR illuminator 115: IR LED

       120: image acquisition unit 127: IR filter

       130: face detection unit 160: feature information extraction unit

       170: face recognition unit 200: face recognition processing unit

       300: database 410: illuminance detector

       420: First image acquisition unit 430: Second image acquisition unit

 500: real recognition unit

Claims (11)

delete delete An illuminance detecting unit detecting an external illuminance; An image obtaining unit which obtains an input image by passing light of an area having a different wavelength according to the detected illuminance; A feature information extracting unit which extracts feature information about a predetermined face included in the input image; And And a face recognition unit configured to recognize the face of the input image by comparing the extracted feature information with feature information of a person image stored in a database. The apparatus of claim 3, wherein the image acquisition unit A first image acquisition unit that receives the visible light region and receives the input image when the illuminance is within a predetermined range; And If the illuminance is out of a predetermined range includes a second image acquisition unit for applying the IR filter to obtain the input image, And the predetermined range is in the range of 50 to 150 Lux. The method of claim 3, wherein When the illuminance is out of a predetermined range further comprises an IR illumination unit for irradiating light in the infrared region, And the predetermined range is in the range of 50 to 150 Lux. The method of claim 3, wherein If the face recognition unit does not recognize the face of the input image, the IR illumination unit irradiates light in the infrared region, The image acquisition unit obtains the input image by applying an IR filter, facial recognition system using IR illumination. The method of claim 4, wherein the IR filter Face recognition system using IR illumination, which is a bandpass filter that passes light of wavelength 850 nm to 950 nm. delete Sensing external illuminance; Acquiring an input image by passing light of an area having a different wavelength according to the sensed illuminance; Extracting feature information about a predetermined face included in the input image; And And comparing the extracted feature information with feature information of a person image stored in a database to recognize a face of the input image. The method of claim 9, wherein the obtaining of the input image comprises: When the illuminance is within a predetermined range, receiving the visible light region to obtain the input image; And Obtaining an input image by applying an IR filter when the illuminance is out of a predetermined range; The predetermined range is 50 to 150 Lux range, facial recognition method using IR illumination. The method of claim 9, If the illuminance is out of a predetermined range further comprises the step of irradiating light in the infrared region, The predetermined range is 50 to 150 Lux range, facial recognition method using IR illumination.

Images