JP4739870B2 - Sunglasses detection device and face center position detection device - Google Patents

Description

  The present invention relates to a sunglasses detection apparatus that detects an image of sunglasses from a face image, and a face center position detection apparatus that can accurately detect the center position of a face even when sunglasses are detected.

  In order to identify a person or read a facial expression based on a face image, it is important to detect the face position, the face center position, the orientation, and the like.

  For example, Patent Document 1 discloses a technique for processing an image obtained by photographing a human face and detecting the position of the face with high accuracy without being affected by the movement or background of the person.

Further, Patent Document 2 captures an image obtained by photographing an area where a driver can exist, detects a lateral edge image from the captured image, and detects a candidate line for a face center line in the detected lateral edge image. A plurality of face center candidate lines, and for each of the plurality of face center candidate lines, based on a value indicating the probability that the face center candidate line is the face center line, the presence / absence determination accuracy of the face Techniques for improving the above are disclosed.
JP 2004-310396 A JP 2005-011097 A

  Drivers often wear sunglasses. In the techniques disclosed in Patent Documents 1 and 2, since the size of the lens portion of the sunglasses is larger than that of a normal eye, the position of the detected face and the position of the center line of the face deviate from the accurate positions. .

The present invention has been made in view of the above problems, and an object of the present invention is to make it possible to determine the presence or absence of sunglasses from a person image.
Another object of the present invention is to make it possible to discriminate the position and center position of a face from a person image wearing sunglasses.

In order to achieve the above object, the sunglasses detection apparatus of the present invention is
Face image storage means for storing a face image obtained by photographing a face;
Edge detection means for performing black and white white edge detection processing in which a predetermined number of black pixels are matched to the size of the lens portion of sunglasses for the face image stored in the face image storage means;
Sunglasses detecting means for determining the presence or absence of an image of sunglasses of a face image based on the edge detected by the edge detecting means;
It is characterized by comprising.

  The edge detection unit performs, for example, a monochrome white edge detection process in accordance with the size of the lens portion of the sunglasses on the image data of the eye detection planned position of the face image stored in the face image storage unit.

  The edge detection means includes, for example, a first image processing means for executing a first image processing for detecting a black and white white horizontal edge or a black and white white vertical edge that matches a size of a lens portion of the sunglasses from the face image; Second image processing for executing second image processing for detecting black and white white vertical edges or black and white white horizontal edges in accordance with the size of the lens portion of the sunglasses from the face image processed by the first image processing means Means.

  For the face image stored in the face image storage means, an eye detection means for detecting eye image data is further arranged, and when the edge detection means cannot detect the eye by the eye detection means, You may comprise so that the black-and-white white edge detection process matched with the size of the lens part of sunglasses may be performed.

  For example, the face image storage means sequentially stores images obtained by photographing faces in time series, and the edge detection means performs edge detection processing on a plurality of images stored in the face image storage means. The sunglasses detecting means detects the presence / absence of sunglasses based on the result of edge detection processing for a plurality of images.

  An infrared camera for photographing a face image and supplying it to the face image storage means may be arranged.

  According to a second aspect of the present invention, there is provided a face center discriminating apparatus, and the center of the face based on the assumed position and size of the eye when sunglasses are detected by the sunglasses detecting apparatus and the sunglasses detecting apparatus. And means for discriminating.

  For example, the face center determination unit estimates the position of the eye based on the position of the sunglasses detected by the sunglasses detection device, and determines the center of the face based on the estimated position of the eye.

The face center determining means, for example, synthesizes an eye image not including sunglasses with a face image including sunglasses based on the position of the sunglasses detected by the sunglasses detection device, and determines the center of the face based on the combined image. To do.
For example, the face center determination unit obtains the center of gravity of each part of the face included in the composite image.

Further, a computer program according to a third aspect of the present invention relates to a face image storage means for storing a face image obtained by photographing a face of a computer, and the face image stored in the face image storage means. Edge detection means for performing black and white white edge detection processing in which a predetermined number of black pixels according to the size of the lens portion of the sunglasses are continuous, and based on the edge detected by the edge detection means, the presence or absence of the sunglasses image of the face image It is made to function as a sunglasses detection means to discriminate | determine.

  According to this invention, sunglasses can be detected from a face image. According to the present invention, the center of the face can be obtained based on the detection of sunglasses.

  Hereinafter, a face center position detection device having a sunglasses detection function according to an embodiment of the present invention will be described.

  As shown in FIG. 1, the face center position apparatus of the present embodiment captures a driver's face and generates a face image, a camera 10, an illumination light source 12 that illuminates the driver's face, and the driver's face center position. The computer 14 to detect and the display apparatus 16 connected to the computer 14 are provided.

  The camera 10 is composed of a CCD camera, for example, and acquires a gradation image of the driver's face. The face image generated by the camera 10 includes not only the driver's face but also its background.

  The display device 16 includes an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube), or the like, and displays a binarized image extracted from a face image taken by the camera 10.

  The computer 14 is a device that processes the face image acquired by the camera 10 and detects the center position of the face. As shown in FIG. 2, the A / D converter 21, the image memory 22, the ROM 23, A CPU 24, a RAM 25, a display control device 27, and a light source control device 28 are provided.

An A / D (analog / digital) converter 21 converts an analog image signal photographed by the camera 10 into a digital signal.
The image memory 22 stores image data generated by the camera 10 and digitized by the A / D converter 21.
The ROM 23 stores programs and fixed data for controlling the operation of the CPU.
The CPU 24 executes a program stored in the ROM 23 to perform image processing and detect the center position of the face.
The RAM 25 functions as a work rear for the CPU 24.
The display control device 27 controls the display device 16 under the control of the CPU 24.
The light source control device 28 controls turning on / off of the illumination light source 12.

FIG. 3 shows an example of fixed data stored in the ROM 23.
As shown in FIG. 3A, the ROM 23 stores coordinate data y1 and y2 that define areas (eye areas) where the eyes are expected to be located in advance in the face image captured by the camera 10.

  Further, as shown in FIG. 3B, the ROM 23 stores the expected vertical and horizontal widths (number of pixels) of the eye image included in the face image captured by the camera 10 and the image of the lens part of the sunglasses. The expected vertical width and horizontal width (number of pixels) are stored. These values are the sizes when the size of the face image is about 60 pixels in both vertical and horizontal directions (yx), and are appropriately changed according to the size of the image.

  Further, the ROM 23 stores a standard image of the eyes of the driver as shown in FIG. Note that not only eyes but also images such as eyebrows and sets of eyes and eyebrows may be prepared.

Next, the operation of the face center position detection apparatus having the above configuration will be described.
FIG. 4 is a flowchart showing a processing routine of the face center position detection program stored in the ROM 23. The CPU 24 performs processing by repeating steps according to this processing routine.

  For example, every time the internal timer passes a certain time, the CPU 24 generates an internal interrupt every 100 ms, for example, and starts the processing routine shown in FIG. The data is taken in via the D converter 21 and stored in the image memory 22 (step S11). Thereby, for example, a driver's face image as shown in FIG. 9A or FIG. 10A is obtained. Note that this face image includes images other than the face such as the background.

  Next, the CPU 24 performs preprocessing such as noise removal processing on the captured face image (step S12). For example, the CPU 24 removes the image of the background portion (stationary object) by taking the difference between the image captured immediately before and the image captured this time. As a result, for example, the background stationary object image is removed from the driver's face image shown in FIGS. 9A and 10A as shown in FIGS. 9B and 10B. . In this pre-processing, other noise removal processing and gradation adjustment are performed as necessary.

  Subsequently, the CPU 24 binarizes the gradation of each pixel constituting the face image with an arbitrary threshold as a reference (step S13). As the threshold value, for example, an average value of all the pixels forming the face image can be used. Thereby, for example, a binary image as shown in FIG. 9C or FIG. 10C is generated from the gradation face image of the driver shown in FIG. 9B or FIG. 10B.

  Subsequently, the CPU 24 reads out the area designation data (y1, y2 shown in FIG. 3A) that defines the area where the eye is expected to be present from the binary image, and reads the eyes of the binarized image. The region is cut out as shown in FIGS. 9D and 10D (step S14).

Subsequently, a black-and-white white horizontal edge is detected from the cut binary image (step S15).
Here, the black-and-white white edge is not an edge that merely becomes a boundary between white and black, but an edge that changes from white to black and further to white and has a predetermined width. Among these, the black and white white horizontal edge means a boundary extending in the horizontal direction, and the black and white white vertical edge means a boundary extending in the vertical direction.

  In the monochrome white / horizontal edge detection process, the CPU 24 detects a black pixel row having a width of 2 to 4 pixels in the vertical direction in the eye area image according to the setting shown in FIG.

  The technique itself for detecting the black-and-white white horizontal edge is arbitrary. For example, as shown in the flowchart of FIG. 5, while sequentially updating the coordinate value y, the luminance of the pixel (x, y) is determined, and when the luminance changes from white to black, the number of consecutive blacks is counted, When black changes to white, it is determined whether or not the number of continuous blacks is 2 to 4. If the continuous number is 2 to 4, the pixel is maintained; Can be used.

  Specifically, the CPU 24 first sets the coordinates (x, y) of the pixel to be processed as the initial value (1, 1) (step S101). Note that the coordinates of the origin of the eye area image are (0, 0).

  Subsequently, the CPU 24 determines the color of the pixel to be processed (the pixel at coordinates (x, y)) and the color of the pixel at the previous coordinate (x, y−1). (Step S102).

  If it is determined in step S102 that the color of the pixel at coordinates (x, y-1) is white and the color of the pixel at coordinates (x, y) is black, it is a boundary from white to black, and step S103. Then, the coordinates (x, y) of the pixel to be processed are stored, and the black continuous number counter C is set to 1.

  If it is determined in step S102 that the color of the pixel at the coordinate (x, y-1) is black and the color of the pixel at the coordinate (x, y) is black, the black is continuous, and the process proceeds to step S104. The coordinate value (x, y) of the pixel is stored, and the black continuous number counter C is incremented by one.

  If it is determined in step S103 that the color of the pixel at coordinates (x, y-1) is black and the color of the pixel at coordinates (x, y) is white, this is a boundary from black to white, and step S105. Then, it is determined whether the value of the black continuous number counter C is 2 to 4.

  If it is determined in step S105 that the value of the black continuous number counter C is not any of 2 to 4, that is, the number of continuous black pixels is 1 or 5 (step S105; No), step S103 and / or The color of the pixel at the coordinate position stored in step S104 is changed from “black” to “white” (step S106). Thereafter, the stored coordinate values are reset, and further, the black continuous number counter C is reset to 0 (step S107).

  If it is determined in step S105 that the value of the black continuous number counter C is 2 to 4, that is, the number of continuous black pixels is 2 to 4 (step S105; Yes), step S103 and / or step The black pixel at the coordinate position stored in S104 is maintained as it is, the stored coordinate value is reset, and the black continuous number counter C is reset to 0 (step S107).

  If it is determined in step S102 that the color of the pixel at coordinates (x, y-1) is white and the color of the pixel at coordinates (x, y) is white, white is continuous, and steps S103 to S107 are performed. Jump processing.

  Following the above processing, the CPU 24 determines whether or not the coordinate value y has reached the final value (step S108). If the coordinate value y has not reached (step S108; No), the next in the vertical direction (y direction) is determined. In order to process the pixel, y is incremented by 1 (step S109), and the process returns to step S102.

  On the other hand, if the coordinate value y has reached the final value in step S108 (step S108; Yes), it is determined whether or not the x-coordinate value is the final value (step S110). If the x-coordinate is not the final value (step S110; No), x = x + 1 and y = 1 are set (step S111) in order to process the pixels in the next column, and the process returns to step S102. If the x coordinate is the final value (step S110; Yes), the process is terminated.

  With such a configuration, black pixels having a continuous number of 1 or less and 5 or more in the y direction are converted into white pixels. For this reason, when the image of the cut-out eye region is an image as illustrated in FIGS. 11A and 12A (each cell represents one pixel), FIG. 11B and FIG. It is converted into an image as shown in FIG. As a result, black pixel rows that do not constitute eyes or eyebrows and that are long in the vertical direction, or pixels that have only one pixel in the vertical direction are removed.

  After the black and white white horizontal edge detection process shown in FIG. 5 is completed, the process proceeds to step S16 in FIG.

  The black-and-white white vertical edge detection process is a process of detecting a black pixel row having a width of 6 to 12 pixels in the horizontal direction from the image of the eye region processed in step S15 in accordance with the setting shown in FIG.

  Although the technique itself for detecting the black and white white vertical edge is arbitrary, for example, the technique shown in FIG. 6 can be used. The process shown in FIG. 6 is basically the same as the process shown in FIG. 5 except that the process direction is different. In step S201, the process start point pixel is set to (1, 1). The pixel coordinates (x, y) and the color of the pixel at the previous (left lateral) coordinate (x-1, y) are discriminated (step S202).

If the pixel color is changed from white to black in step S202, the coordinates (x, y) of the pixel to be processed are stored, and the black continuous number counter C is set to 1 (step S203).
If it is determined in step S202 that black pixels are continuous, the coordinate value (x, y) of the pixel is stored, and the black continuous number counter C is incremented by 1 (step S204).

If it is determined in step S202 that the color of the pixel has changed from black to white, it is determined which of the black continuous number counter C is 6 to 12 (step S205), and the continuous number of black pixels. Is determined to be 1 to 5 or 13 or more (step S205; No), the color of the pixel at the stored coordinate position is changed from "black" to "white" (step S206), and the stored coordinates The value and the black continuous number counter C are reset (step S207),
If it is determined in step S202 that the continuous number of black pixels is 6 to 12 (step S205; Yes), these black pixels are maintained as they are (step S207).

  Then, the process is sequentially advanced while updating x. When x reaches the final value, the same process is repeated for the next pixel line, and when the process reaches the final pixel line, the process ends (steps S208 to S211).

  By adopting such a configuration, for example, when the image of the eye region after the detection of the horizontal edge is the image illustrated in FIGS. 11B and 12B, the continuous number is 1 to 5 in the x direction. And 13 or more black pixels are converted into white pixels. As a result, black pixel rows that are short in the horizontal direction and do not constitute eyes or eyebrows or black pixel rows that are too long are removed.

  Next, the CPU 24 determines whether or not an eye image has been detected based on the image of the eye area after such processing (step S17 in FIG. 4).

  The CPU 24 determines that an eye image has been detected when a black region of 2 to 4 pixels in the vertical direction and 6 to 12 pixels in the horizontal direction is obtained in the processed image of the eye region ( Step S17; Yes). In this case, the CPU 24 resets the number of consecutive sunglasses detections to 0 (step S18).

  Further, the CPU 24 overwrites the original binary image with the image of the eye area after such processing, and generates a face image for measuring the center of gravity (step S19). For example, the processed eye region image shown in FIG. 9D is overwritten with the original binary image shown in FIG. 9C, and the face image for center position measurement shown in FIG. Generate. Note that processing similar to that for the eye region may be performed on the mouth region, the nose region, and the like to remove noise.

Subsequently, the center position (face position, center of gravity position) of the face is obtained using the center position measurement image obtained in step S19 (step S20). Although the center measurement method itself is arbitrary, for example, the coordinates of the center (center of gravity) position of the face can be obtained from the following equation.
X-coordinate of the center of the face = Σxi / n xi: x-coordinate value of the i-th black pixel y-coordinate of the center of the face = Σyi / n yi: y-coordinate value of the i-th black pixel i: 1 to n n Is the total number of black pixels

  When the driver wears sunglasses, the lens part of the sunglasses (the part that covers the eyes) is much larger than the human eye. For this reason, as shown in FIG. 12, the pixels of the image of the sunglasses portion are converted into white pixels by the monochrome white edge detection process executed in steps S15 and S16. For this reason, almost no black pixels remain in the image of the eye area. For this reason, in step S17, it is determined that the eye image could not be acquired (step S16; No).

For this reason, a process progresses to the sunglasses detection process of step S21.
In the sunglasses detection process of step S21, as shown in FIG. 7, the image of sunglasses is detected in the same manner as the processes of steps S14 to S16 for detecting the eye region. That is, in accordance with the setting shown in FIG. 3B stored in the ROM 23, a portion where a predetermined number of black pixels continue in the vertical and horizontal directions is detected based on the expected size of sunglasses.

  First, as shown in FIG. 7, an image of the eye area is cut out from the binarized face image (step S41).

  Subsequently, a black-and-white white lateral edge detection process for detecting sunglasses is performed on the cut-out eye region image (step S42). This process is basically the same as the eye extraction process shown in FIG. However, in step S105, the range of C is set to 6 to 12 in accordance with the size of the sunglasses in the vertical direction in accordance with the setting of FIG.

  Subsequently, a black-and-white white vertical edge detection process for detecting sunglasses is executed on the image of the eye area where black-and-white white and horizontal edges have been detected (step S43). This process is basically the same as the vertical edge detection process of the eye shown in FIG. However, in step S205, the range of C is set to 10 to 18 according to the size of the sunglasses in the horizontal direction in accordance with the setting of FIG. 3B stored in the ROM 23.

  As a result of such processing, for example, the image of the lens portion of the sunglasses as shown in FIG. 13A is processed as shown in FIG. 13B by the black and white white lateral edge detection processing (step S42), Further, processing is performed as shown in FIG. 13C by black and white white vertical edge detection processing (step S43), and the lens is detected.

Based on the image of the eye area after performing such processing, the CPU 24 determines whether or not sunglasses are detected (FIG. 4; step S22).
As shown in FIG. 13C, the CPU 24 determines that an eye image has been detected when a black region of 6 to 12 pixels in the vertical direction and 10 to 18 pixels in the horizontal direction is obtained. .

  If sunglasses are not detected (step S22; No), the flow proceeds to step S18 described above, the number of consecutive sunglasses detections is set to 0, normal image processing is performed (step S19), and the face center position is detected. (Step S20).

  On the other hand, if sunglasses are detected in step S22 (step S22; Yes), the number of consecutive sunglasses detections is incremented by 1 (step S23), and it is further determined whether the number of consecutive sunglasses detections is greater than or equal to the reference value m. (Step S24). The number of consecutive sunglasses detections is a value indicating how many times sunglasses have been detected continuously for the face images sequentially acquired at a constant cycle in step S11. If the interrupt cycle is 100 ms as described above, but sunglasses are continuously detected for m · 100 ms, it is determined Yes in step S24.

  If it is determined as Yes in step S24, sunglasses corresponding image processing is executed (step S25). In this sunglasses-compatible image processing, the positions of the left and right lenses of the sunglasses are specified, the positions of the left and right eyes are specified (estimated) from the specified lens positions, and the figure stored in the ROM 23 with this position as a reference. The eye image shown in 3 (c) is synthesized.

  Specifically, as shown in FIG. 8, the CPU 24 first determines the coordinates of the center positions of the left and right lenses of the sunglasses (step S51). The method for obtaining the center position of each lens is arbitrary. For example, the same process as the process for obtaining the center of gravity shown in step S20 may be performed for each of the black images of the lenses located on the left and right sides of the eye area.

  Next, the eye position is specified (estimated) based on the specified lens position (step S52). The eye position may be the center position of the lens, or may be obtained, for example, by adding a statistically calculated offset value to the lens center position.

  Next, the eye image (FIG. 3C) stored in the ROM 23 is combined with the eye position obtained in step S52 of the same size image as the eye region image (step S53).

  Next, the image of the portion excluding the eye area in the binary image and the eye area image generated in step S53 are synthesized, and the face image including the eyes, nose, mouth, and the like is obtained as the face image for measuring the center of gravity. Are synthesized (step S54).

  To explain based on a specific example, for example, an image of the eye region shown in FIG. 14B is cut out from the face image of the driver wearing sunglasses shown in FIG. Step S41). Subsequently, as shown in FIG. 14C, an image of the lens of the sunglasses is extracted by the black and white white horizontal edge detection process for detecting sunglasses (step S42) and the black and white white vertical edge detection process for detecting sunglasses (step S43). Is done.

  Subsequently, if sunglasses are detected in step S22 and the number of consecutive sunglasses detections is greater than or equal to m (step S24; Yes), the position of each lens of the sunglasses is specified by the sunglasses-compatible image processing in step S25 (step S51). ) Based on the lens position, the eye position is specified as shown in FIG. 14D (step S52). Subsequently, as shown in FIG. 14E, an eye image is synthesized (step S53).

  Subsequently, the binary image from which the eye area shown in FIG. 14F is removed and the eye area image shown in FIG. 14E are combined to generate a face image for measuring the center of gravity (step S54).

  Thereafter, the process proceeds to step S20 in FIG. 4, and the center position of the face is determined by the above-described process.

  As described above, according to the face center position detection device, the presence / absence of sunglasses is detected from the face image, and the center position of the face is appropriately determined based on the detection result regardless of whether or not sunglasses are worn. It becomes possible to do. By determining the center position of the face, it is possible to determine the presence / absence of the face, the position of the face, the orientation of the face, and the like.

In addition, this invention is not limited to the said embodiment, A various deformation | transformation and application are possible.
For example, in the above embodiment, the eye area is fixed at a specific position in the image, but the position of the eye area may be appropriately set according to the size and position of the face image. In this case, for example, the vertical edge sobel filter and the horizontal edge sobel filter are applied to the image acquired in step S11, the face edge is detected, the top edge, the bottom edge, the left edge, and the right edge of the face are determined. The area between the left end and the right end in the area of 3/10 to 6/10 between the lower end and the lower end may be the eye area. In addition, the method for specifying the eye region is arbitrary.

  In the above embodiment, the center (center of gravity) position of the face is obtained from the images of eyes, eyebrows (or eyes), nose, and mouth, but the face part (parts) used to determine the center position is arbitrary. It is. For example, the center may be obtained by adding ears, cheeks, hair, and the like.

  Moreover, in the said embodiment, when sunglasses were detected, the image of eyes was combined and the center position of the face was calculated | required based on the synthesized image, but the method of calculating | requiring a gravity center is arbitrary. For example, when sunglasses are detected, a center is obtained by combining the images of “eyes and eyebrows”, or a weight of about 0.3 to 0.5 times is applied to each pixel constituting the lens image of sunglasses. In addition, the center of gravity of each face (lens, eyes, nose, mouth,...) May be obtained to obtain the center of gravity of the face image. Further, the center of gravity is obtained by adding an offset value in consideration of the influence of the eye image in advance to the center of gravity obtained from the face image excluding the eye region as shown in FIG. Also good.

  In the above embodiment, the present invention has shown the embodiment in which the presence / absence of sunglasses is detected in order to detect the center position of the face. However, after detecting the sunglasses, what kind of processing is used for the result? Is optional. For example, it is possible to detect sunglasses, thereby changing the intensity of illumination and adjusting the brightness of the backlight of the screen.

  The system configuration described with reference to FIGS. 1 and 2 is also an example, and can be arbitrarily changed. For example, it is not always necessary to provide the illumination light source 12. For example, if the camera 10 is an infrared camera that captures images with far infrared rays, it is possible to acquire images of each part of the face relatively accurately without being affected by race, skin, or hair color. Become.

The above-described flowchart can be arbitrarily changed as long as the same function can be realized.
For example, in the above embodiment, when the image of sunglasses is obtained continuously m times or more, the sunglasses-compatible image processing (step 25) is executed. Image processing may be executed.

  In the above embodiment, the black-and-white white horizontal edge detection process and the black-and-white white vertical edge detection process are executed after the binary image is formed in step S13, but the gradation value of the face image is not performed without performing the binarization process. Alternatively, the black and white white edge may be directly detected.

  Further, in the present invention, the black and white white edge is for expressing a difference in gradation, and is not limited to white or black as a color, and any hue may be used. Regarding color images, the presence or absence of sunglasses may be determined in consideration of the hue of each pixel.

  In each of the above embodiments, the present invention is applied to the case where sunglasses are detected by photographing a driver. However, the present invention is not limited to this, and humans, animals, dolls, robots, etc. wear sunglasses in any scene. The present invention can be widely applied to processing for determining whether or not it is applied.

  The present invention is not limited to processing while acquiring an image with a camera. For example, for each of one or a plurality of face images photographed elsewhere, the presence or absence of sunglasses, the position of the center of the face, the orientation of the face, etc. It can be used to determine.

  Further, a computer program for causing a computer to execute the above-described processing may be stored in the ROM via an arbitrary recording medium or a network.

It is a block diagram of the face center discriminating device concerning the embodiment of the present invention. It is a block diagram which shows the structure of the computer shown in FIG. It is a figure for demonstrating the various data stored in ROM. 3 is a flowchart for explaining the operation of the face center discrimination device shown in FIG. 1. It is a flowchart for demonstrating the specific example of the black-and-white white horizontal edge detection process in the flowchart of FIG. It is a flowchart for demonstrating the specific example of the black-and-white white vertical edge detection process in the flowchart of FIG. It is a flowchart for demonstrating the specific example of the sunglasses detection process in the flowchart of FIG. It is a flowchart for demonstrating the specific example of the sunglasses corresponding | compatible image process in the flowchart of FIG. It is a figure for demonstrating the content of the face image of a driver who is not wearing sunglasses, and image processing. It is a figure for demonstrating the face image of the driver wearing sunglasses, and the content of image processing. It is a figure for demonstrating a black-and-white white edge and its detection process. It is a figure for demonstrating a black-and-white white edge and its detection process. It is a figure for demonstrating the image of the lens of sunglasses, and its detection process. It is a figure for demonstrating the process which calculates | requires the center position of a face from the face image containing the image of sunglasses.

Explanation of symbols

10 Camera 22 Image memory (Face image storage means)
23 ROM (Sunglasses detection means, center position determination means)
24 CPU (Sunglasses detection means, center position determination means)

Claims (8)

Face image storage means for storing a face image obtained by photographing a face;
Edge detection means for performing black and white white edge detection processing in which a predetermined number of black pixels are matched to the size of the lens portion of sunglasses for the face image stored in the face image storage means;
Sunglasses detection means for determining the presence or absence of sunglasses images of the face image based on the black and white white edge detected by the edge detection means;
A sunglasses detection device comprising:   The edge detection means performs black-and-white white edge detection processing in accordance with the size of the lens portion of the sunglasses for the image data of the eye detection scheduled position of the face image stored in the face image storage means. The sunglasses detection device according to claim 1. The edge detection means includes
A first image processing means for performing a first image processing for detecting, from the face image, a black and white white horizontal edge or a black and white white vertical edge that matches the size of the lens portion of the sunglasses;
Second image processing for executing second image processing for detecting black and white white vertical edges or black and white white horizontal edges in accordance with the size of the lens portion of the sunglasses from the face image processed by the first image processing means Means,
The sunglasses detecting device according to claim 1, comprising: Eye detection means for detecting eye image data for the face image stored in the face image storage means;
The edge detection means performs black and white white edge detection processing according to the size of the lens portion of the sunglasses when the eyes cannot be detected by the eye detection means.
The sunglasses detection device according to claim 1, wherein: The face image storage means sequentially stores images obtained by photographing faces in time series,
The edge detection means performs edge detection processing for a plurality of images stored in the face image storage means,
The sunglasses detection means detects the presence or absence of sunglasses based on the result of edge detection processing for a plurality of images.
The sunglasses detection device according to claim 1, wherein:   The sunglasses detecting apparatus according to claim 1, further comprising an infrared camera for photographing a face image and supplying the photographed image to the face image storage means. A sunglasses detection device according to claim 1;
Center position determination means for determining the center of the face based on the assumed eye position and size when sunglasses are detected by the sunglasses detection device;
A face center discrimination device characterized by comprising: Computer
Face image storage means for storing a face image obtained by photographing a face;
Edge detection means for performing black and white white edge detection processing in which a predetermined number of black pixels are matched to the size of the lens portion of sunglasses for the face image stored in the face image storage means;
Sunglasses detection means for determining the presence or absence of sunglasses image of the face image based on the edge detected by the edge detection means,
A computer program that functions as a computer program.

Images