JPS63311106A - Apparatus for recognition crew's position of vehicle - Google Patents

Abstract

PURPOSE:To enhance a recognition rate by detecting the movement of crew within a narrow range rapidly and accurately, by providing a pattern detection means for detecting the pattern of the state in a compartment, a detection means for detecting an abnormal region and a means for setting a predetermined region at the position corresponding to the abnormal region detection result. CONSTITUTION:The image of the crew M1 of a vehicle is caught by a light emitting part and an image detection part to obtain image data M2. Further, the predetermined region of the driver M1 of the vehicle, pref. the nose thereof present in a predetermined region M3 is detected from the data of the region M3 set in the image data M2. The position of the region M3 in the image data M2 is set by estimating the position of the predetermined region of the crew M1 of the vehicle in accordance with the detection result of an abnormal region detection means M6, that is, the region of abnormality such as the positional movement of the head from the predetermined pattern, pref., the boundary line of the ceiling and door part of the vehicle detected by a pattern detection means M5. By this method, the part where it is estimated that there is the head of the driver 1 is set, and the detection of the apex nasi and the confirmation of the position of an eye are performed from the predetermined region in the vicinity of said set part and, therefore, accurate detection becomes possible without lowering detection efficiency and a detection region can be made narrow.

Description

DETAILED DESCRIPTION OF THE INVENTION Object of the Invention [Industrial Field of Application] The present invention relates to a device for recognizing the riding position of a vehicle occupant for detecting the position of a driver or the like inside the vehicle and using the detected position for various controls. .

[Conventional technology] Conventionally, the position of the driver of a car, etc., especially the position of the eyes, is recognized and the position and angle of rearview mirrors, headrests, air-conditioned air outlets, steering wheels, etc. are automatically adjusted, or when the driver falls asleep. A device has been proposed that detects the
No. 4, etc.).

As one means of accurate position recognition using these devices, there is a device that attempts to detect the position of the nose, etc. by using a portion of a relatively wide image captured by an imaging device. The reason why a part of the image is used is because the imaging means is fixed and cannot be taken at an arbitrary angle, so by capturing a wide range of images and capturing a part of the image within an arbitrary range, the imaging angle of the imaging means can be adjusted. This is to achieve the same effect as changing the area, and furthermore, to improve detection efficiency only in the vicinity of a specific part and prevent misidentification. Once the position of this specific part, such as the eyes or nose, is determined,
Since it is empirically known that the detection area is within a certain narrow range, the detection area can be set in advance.

[Problems to be Solved by the Invention] However, if there are individual differences or if the seat where the driver is present moves, the detection area may deviate from the preset specific area. There is a high possibility that a specific part cannot be detected or that it is erroneously detected, and the range that should be detected becomes wider, making it impossible to detect it quickly.

Structure of the Invention Therefore, the present invention employs the following structure for the purpose of solving the above problems.

[Means for Solving the Problems] That is, the gist of the present invention is, as illustrated in FIG. In the vehicle occupant riding position recognition device that sets M3 and detects a predetermined part of the vehicle occupant M1 from the predetermined area M3, the device further includes: a pattern that detects the state of the predetermined pattern M4 existing in the vehicle interior on the image data; a detection means M5, an abnormal region detection means M6 on the image data of the predetermined pattern M4, and a position corresponding to the detection result of the abnormal region detection means M6;
A device for recognizing the riding position of a vehicle occupant, comprising: a predetermined region setting means M7 for setting the predetermined region M3.

[Operation] The vehicle occupant riding position recognition device captures an image of the vehicle occupant M1 and obtains image data M2. Furthermore, this recognition device uses the data of a predetermined area M3 set within the image data amount 2,
A predetermined part of the vehicle occupant M1 existing in the overcrowded predetermined region M3 is detected. The position of this predetermined region M3 within the image data amount 2 is determined based on the detection result of the abnormal region detection means M6, that is, the abnormal region, of the predetermined pattern detected by the pattern detection means M5, and the predetermined region of the vehicle occupant M1. is set by estimating the position of.

[Example code] Next, one embodiment of the present invention will be described with reference to the drawings.

The present invention is not limited to these, but includes various embodiments without departing from the gist thereof.

FIG. 2 is a perspective view showing the instrument panel part and its surroundings of a vehicle in which the passenger position recognition device of this embodiment is installed. It represents driving a car face to face. A light emitting unit 5 that illuminates the upper body of the driver 1 from one direction to the left is located diagonally forward to the left of the driver's seat 2 of the instrument panel 3, that is, in front of the passenger seat. An image detection unit 6 that captures a dimensional image is provided.

Here, the light emitting unit 5 uses an infrared strobe that emits infrared light so as not to cause the driver 1 to feel dazzled when irradiating the light, and as shown in FIG. A lens 5b for broadly irradiating the driver with the infrared light emitted by the light emitter 5a, an infrared filter 5c that transmits infrared light but not visible light, these infrared light emitters 5a, and the lens. 5b and an infrared filter 5c, and a case 5d for mounting them at a predetermined position on the instrument panel 3.
It is composed of. Further, as shown in FIG. 4, the image detection unit 6 includes an infrared filter 6a that transmits infrared light, and a focal length for forming an image of the body of the driver 1 on the imaging surface of a solid-state image sensor 6e, which will be described later. For example, 12.5mm lens 6
b, a liquid crystal aperture element 6C for adjusting the amount of light, a phototransistor rod d used to automatically adjust the aperture of the liquid crystal aperture element 6c and detecting the amount of light transmitted through the liquid crystal aperture element 6C, and the above-mentioned infrared diaphragm element 6C. A MOS type solid-state image sensor 6e consisting of a photosensor and a switching circuit, which extracts an image formed on an imaging surface through a filter 6a, a lens 6b, and a liquid crystal aperture element 6c as an electrical signal by switching scanning; A solid-state camera is used, which includes a case 6f for storing each part and attaching to the instrument panel 3.

Next, the overall configuration of the riding position recognition device of this embodiment will be explained based on the block diagram shown in FIG. 5.

As shown in the figure, in addition to the light emitting unit 5 and the image detecting unit 6, this recognition device processes the image of the driver 1 captured by the image detecting unit 6, and processes the riding position of the driver 1 (this embodiment The camera is equipped with an image processing unit 8 that ultimately recognizes the position of the eyes. This image processing unit 8 is configured to include the image detection unit 6
an image human power section 10 that can control the image signal obtained by the image detection section 6, convert the image signal obtained by the image detection section 6 into a digital signal, and temporarily store the converted digital signal as image data; An irradiation signal output unit 12 outputs an irradiation signal for causing the driver 1 to emit light. A central processing unit (C
PU) 14, a read-only memory (ROM) 16 in which control programs and data for executing eye position recognition processing by the CPU 14 are stored in advance, and data used for executing arithmetic processing are temporarily stored. A random access memory (RAM) 1B that is read and written, a path line 20 that connects each of the above parts and serves as a path for image signals and control signals,
It is composed of a power supply circuit 22 that supplies power to each of the above-mentioned parts.

The image input section 10 also includes a synchronization signal generation circuit 10a that generates a vertical synchronization signal and a horizontal synchronization signal to the image detection section 6.
For example, the image signal obtained by the image detection unit 6 is
An A/D conversion circuit 10b that converts image data into digital signals representing the brightness of 384 pixels horizontally and 240 pixels vertically per frame, and an image data storage circuit 10c that can store the converted image data. It is composed of.

Next, the operation of the eye position recognition process executed by the image processing section 8 will be explained along the flowcharts shown in FIGS. 6(a) and 6(b).

This process is executed when a request for driver 10's position recognition is made manually by a signal of a certain period or a signal from a switch, and as described above, in this embodiment, driver 1's position recognition is The position of the eyes will be recognized as the riding position.

When the process starts, step 110 is first executed,
In the image detection section 6, an irradiation image 71 as shown in FIG. 7 obtained when the light emitting section 5 is made to emit light is sent to the image input section
The irradiation image data G is stored as irradiation image data G consisting of digital signals.

As described above, the image processing unit 10 includes a synchronization signal generation circuit 10a, an A/D conversion circuit 10b, and an image data storage circuit 10c, and includes a vertical synchronization signal and a horizontal synchronization signal output from the synchronization signal generation circuit 10a. The image detection section 6 is controlled according to the signal, the image signals output from the image detection section 6 are sequentially captured, and the image signals converted into digital signals via the A/D conversion circuit 10b are stored as image data G in the storage circuit 10c. Therefore, the main field image data can be easily read by simply operating the image processing unit 10 and causing the light emitting unit 5 to emit light in synchronization with the vertical synchronization signal output from the synchronization signal generation circuit 10a. It can be executed. In other words, as shown in FIG. 8, first, after a predetermined time ΔT has elapsed from the rising time to of the vertical synchronization signal, the irradiation signal S is output to the light emitting unit 5, so that the vertical retrace time Δ
The light emitting unit 5 is caused to emit light in coincidence with t1, and an irradiation image as shown in FIG. 7 obtained by the light emission is read as image data G.

The image data G that is A/D converted and stored in the image processing unit 10 represents the brightness of pixels that divide the image obtained by the image detection unit 6 into 384 horizontal by 240 vertical pixels, as described above. As shown in Figure 9, the brightness G of each pixel is
(x, y) is stored.

In this way, the light emitting section 5, the image detection section 6, and the image human power section 10 are controlled in step 110, and when the image data G is read in the image human power section 10, the subsequent step 120 is executed. The process of setting a predetermined area in the image data G for detecting upright positioning is performed as shown in the flowchart shown in FIG. 6(b).

First, in step 121, as shown in FIG. 10, a process is performed to detect a boundary region whose brightness exceeds a predetermined value Lt in the horizontal direction. That is, the value of each brightness G (x, y) on the screen is
Search horizontally from (0, O) to the right to G (383, 0), determine whether the brightness is equal to or greater than a predetermined 1LfiL7,
The first X coordinate xO that exceeds a predetermined value fiLt is stored. Thereafter, the brightness in the horizontal direction is determined every few lines. For example, next, G(0,5) to G (383,5) are determined, and then G(0,10) to G (383,1
0) is 11 and the judgment is made. In this way, XO to xn are obtained. These XO to xn are arranged as shown in FIG. 11 in screen coordinates. That is, this embodiment corresponds to the pattern of the boundary 73 between the ceiling and the door of a single room.

Therefore, the difference dxO between adjacent values of XO~xn is usually
~dxn-1 has a certain relationship.

Here, dxi =xi+1-xi. For example, if the boundary between the ceiling of the passenger compartment and the door is a straight line, dxO~dx
n-1 is constant. Therefore, if dxi is abnormally small or large and irregular compared to other values,
The occupant 1 should exist between the image detection unit 6 and the boundary 73 inside the vehicle interior, as shown in FIG.

Therefore, in step 123, dxO to dxn-1 are determined and irregular portions thereof are detected. If it is known in advance that the boundary is a predetermined straight line, processing is performed in order from dxO to determine that there is no abnormality if dxi is within a predetermined value range, for example, 11≦dxi≦17. After making the determination in this way, as shown in Figure 12-11'1-, the dxi that was first determined to be abnormal
The coordinate G (tx, ty) on the image data of one value of x1→1 is detected as an abnormal site.

Next, in step 127, assuming that the xi+1 portion is the driver's head, the nose existing region 75 is determined in correspondence with the above coordinates G Ctx, ty). For example, G (tx.

ty), -25 in the X-axis direction, +6 in the Y-axis direction, and 4 with the distant position as the upper left end! A rectangular area with ia 108 and width 80 is determined as the nose existing area 75.

In this way, the process reaches step 130 in FIG. 6(a),
Bright spot search processing for the nose existing region 75 is performed. That is,
Search for a pixel on the leftmost side of this area that has a brightness above a certain level, and find its position G (x, y)
A bright spot search process is performed in which the bright spot is stored. This process is a process for limiting the range to search for the nose when detecting the position of the driver's 1 nose in the next step 140,
The position of the determined pixel is used as the starting point for searching for the nose.

In addition, as for the zero bright point search process, starting from the upper left pixel in the nose existing region lj- area 75 in FIG. This can be done by memorizing the location of

When the bright spot search is executed in this way and the position of the pixel that exists on the leftmost side of the nose existing region 75 and has a brightness above a certain level is determined, in the subsequent step 140 as described above, this position is used as the starting point. The process of searching for the nose and detecting the position of the nose is performed, and this process uses a preset standard pattern of the nose as shown in Figure 13 in the vicinity of the pixel found by the bright spot search. The position of the nose is detected by examining the degree of correlation between the image data and the image data, and determining the position of the image where the correlation is maximum as the position where the nose exists.

That is, as shown in FIG. 14, centering on the pixel position G (x, y) determined in step 130 above, 5 pixels each on the upper, lower, left and right sides, for a total of 121 pixels, are
The center position m (0
, 0), the correlation value between the image data and the standard pattern is calculated, and the pixel position where the value is the largest is detected as the position G (+1X, ny) of rhinitis (tip of the nose). Here, the center position m(0,0) in the standard pattern of the nose in Fig. 13 is preset as the part corresponding to rhinitis, and the correlation value can be obtained from the following formula % formula % () . As for the values of l and J above, the standard pattern of the nose is configured as data with a spread of 16 above, 7 below, 1 on the left, and 6 on the right with respect to the center position m (0°0). Therefore, from (-1) to (
+6), an integer value between (-16) and (+7) may be used for j.

In addition, the correlation value is the center position m (
0, 0), a total of 121 pixels, 5 each on the top, bottom, left and right, centering on the pixel G (x, y) on the image data shown in Figure 14, are applied. , as the values of X and Y in the above formula, X is an integer value between X±5 and Y is an integer value between y±5.
A maximum of 121 types of correlation values are calculated by combining these, and the pixel G (X, Y) on the image data that has the maximum value is detected as the nose position G (nx, n).

Once the nose position G (nx, ny) has been determined in this way, the process moves to the subsequent step 150, where the eye positions are determined based on this nose position G (nx, ny). To find the position of the eyes, first find the pixel position G (nx+12, n
y-13), and examine the degree of correlation between the image data and the standard eye pattern shown in FIG. 15, which has been set in advance near this position, and determine the eye position G(mx, my) . Note that this determination method is based on step 1 above.
Since it can be performed similarly to the position determination of the nose in No. 40, the explanation will be omitted.

In the subsequent step 160, the above step 15 is performed.
Eye position G (mx
. 1-1 Assuming that there is no movement of the driver 1 in the left-right (Z) direction shown in FIG. The position of the eye is recognized by determining the three-dimensional position Mt2-coordinates (X, Y, 0') of the eye (left eye), but this process requires that the image detection unit 5 is fixed. Therefore, the position G of the standard point P on the image data (IIX,
py) is stored in advance, and this reference position G (px,
This can be easily carried out by calculating the deviation of the eye position G (mx, my) with respect to p.

As explained above, the vehicle driver's boarding position recognition device of the present embodiment detects the head of the driver 1 in the image data detected by the image detection unit 6 from a predetermined pattern of detected abnormal parts in the vehicle interior. The nose tip is detected from a predetermined area near this part and the position of the eyes is recognized. Therefore, accurate detection is possible without reducing detection efficiency, and the detection area It can also be made narrower. Therefore, even if the driver moves, the position of the eyes can be detected quickly and accurately.

In the configuration of this embodiment, the processing in step 120 by the image processing section 8 corresponds to processing as the pattern detection means M5, the abnormal region detection means M6, and the predetermined area setting means M7.

In the above example, an existing pattern was used, but
Paint, tape, or the like that is highly infrared absorbing or reflective and invisible to the naked eye may be applied or pasted on the interior of the vehicle in advance. In this case, it is easy to form a pattern in a location where a driving vehicle can be easily detected, and the appearance is not impaired.

Alternatively, an existing predetermined pattern may be stored in advance, and pattern abnormalities may be detected by comparing image data only around the corresponding pattern with the stored pattern data. In this case, detection time is extremely shortened and detection accuracy is also improved.

Furthermore, although the pattern detection direction is horizontal in the above embodiment, it may be vertical or diagonal depending on the target pattern. In addition, the nose existing region 75 was set at a predetermined position and within a predetermined size range from the pattern abnormal region G (tx, t), but the abnormal region G (tx,
If the position of G (tx, ty) changes, the image of the driver in the image is enlarged or reduced.
), the abnormal region G (
tx, ty) may be changed. In this case as well, more accurate and rapid detection is possible.

Effects of the Invention Since the present invention is configured as described above, it is possible to relatively accurately set a predetermined area in which a predetermined part of a vehicle occupant is to be detected, in accordance with the body shape and moving position of the vehicle occupant. Therefore, no matter how the vehicle occupant moves, the predetermined region can be detected quickly and accurately within a narrow range, improving the recognition rate. As a result, more accurate automatic control of vehicle equipment can be realized, and the safety of vehicle operation etc. is improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram illustrating the basic configuration of the present invention, FIG. 2 is a perspective view showing the instrument panel of a vehicle in which an eye position recognition device according to an embodiment of the present invention is housed, and its surroundings; 3 is an explanatory diagram of the light emitting section, FIG. 4 is an explanatory diagram of the image detecting section, FIG. 5 is an overall configuration diagram of the eye position recognition device of this embodiment, and FIG. 6 (a) is a flowchart of the image processing routine. , Fig. 6 (b) is a flowchart of the routine for setting the nose existing area, Fig. 7 is an explanatory diagram of the irradiation image, Fig. 8 is a timing chart illustrating the image data loading process, and Fig. 9 is a flowchart of the routine for setting the nose existing area. FIG. 10 is an explanatory diagram of the irradiation image data structure stored in the section, FIG. 10 is an explanatory diagram of boundary region detection, and FIG.
Fig. 12 is an explanatory diagram of abnormal region detection, Fig. 12 is an explanatory diagram of the relationship between the abnormal region and the nose existing region, Fig. 13 is an explanatory diagram of the standard pattern of the nose, and Fig. 14 is an explanatory diagram of the nose position detection process. , 1st
Fig. 5 is an explanatory diagram of the standard eye pattern, Fig. 16 is an explanatory diagram of the three-dimensional position of the eyes, (a) is a plan view of the driver, (a) is an explanatory diagram of the three-dimensional position of the eyes;
b) is its side view. Ml...Vehicle occupant M2...Image data M3...
・Predetermined area M4...Predetermined pattern M5...Pattern detection means-1-M6...-Abnormal part detection means Ml...Predetermined area setting means 1...Driver 5...Light-emitting part 6- ...Image detection section 8...Image processing section 10...Image human power section 1
4...CPU73...Boundary between the ceiling of the vehicle compartment and the door 75-...Nose existing area

Claims (1)

[Scope of Claims] 1. A device for recognizing the riding position of a vehicle occupant, which images the riding state of the vehicle occupant, sets a predetermined area in the captured image data, and detects a predetermined part of the vehicle occupant from the predetermined area. further comprising: pattern detection means for detecting a state on the image data of a predetermined pattern existing in the vehicle interior; abnormal part detection means on the image data of the predetermined pattern; and according to the detection result of the abnormal part detection means. A device for recognizing the riding position of a vehicle occupant, comprising: predetermined area setting means for setting the predetermined area at a position where the vehicle occupant is seated. 2. The device for recognizing the riding position of a vehicle occupant according to claim 1, wherein the predetermined pattern is a boundary line between a ceiling and a door portion of the vehicle. 3. The device for recognizing the riding position of a vehicle occupant according to claim 1 or 2, wherein the predetermined part of the vehicle occupant is the position of the nose.