JP2009208716A - Operator-discriminating device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an operator-discriminating device can discriminate an operator for operating an on-vehicle device is a driver or a passenger. <P>SOLUTION: An extraction unit 32a of an image processing unit 32 sets first and second detection frames 320, 321 on first and second edges 131, 132 of an armrest 13, and extracts an edge in the longitudinal direction of the armrest 13 in the detection frames. When an arm of either a driver 2A or a passenger 2B is on the armrest 13, a discrimination unit 32b detects the end of the edge to discriminate an operator for operating a switch unit 41. Further, when both arms are on the arm rest 13, the extraction unit 32a of the image processing unit 32 sets third and fourth detection frames 322, 323 in a vicinity of the switch unit 41 to extract the edge, and the discrimination unit 32b discriminates the operator for operating the switch unit 41 based on the number of pixels constituting the edges. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an operator discriminating device that can discriminate whether an operator who operates an in-vehicle device is a driver or a passenger.

  As a conventional technique, a display operation unit that is provided in front of a driver seat and a passenger seat of a vehicle and displays operation items of the in-vehicle device, a driver's left hand extended to the display operation unit and the display operation unit, and an assistant An operator who processes the image data captured by the far-infrared camera and the far-infrared camera provided in the center of the driver's seat and the passenger seat on the vehicle center line so that the right hand of the seated occupant can both be viewed and captured. There is known an operator discriminating device including an image processing unit for discriminating (see, for example, Patent Document 1).

According to this operator discriminating device, the operator is discriminated based on whether the center of gravity of the operator's hand extracted from the image data is in the left region or the right region of the reference line set in the image. Based on the determination result and the running state of the vehicle, the permitted range of the operation items displayed on the display operation unit can be limited.
JP 2004-067031 A

  In the conventional operator discriminating device, the far-infrared camera is installed between the driver's seat and the passenger seat, and thus there is a possibility that the view from the rear seat may be blocked. In addition, the far-infrared camera used in the conventional operator discrimination device is expensive and difficult to obtain.

  Further, when the conventional operator discriminating device uses a far infrared camera, a hot object such as a container containing a hot beverage, a high-temperature electronic device, an object heated by direct sunlight, etc. is in the vicinity of the display operation unit. When present, the operator's hand discrimination system could deteriorate.

  Accordingly, an object of the present invention is to provide an operator discriminating device that discriminates an operator who operates an in-vehicle device with high accuracy.

(1) In order to achieve the above object, the present invention is provided at a position where an operator's arm for operating an operation unit of an in-vehicle device provided at a position operable by a driver and a passenger passes through the upper part, A structure having a first edge on the driver side and a second edge on the passenger side, and imaging provided at a position where imaging can be performed so as to include the first and second edges of the structure. And first and second detection frames are set to include the first and second edges based on the image data captured by the image capturing unit and the image capturing unit, and the first and second detection frames are set in the first and second detection frames, respectively. An extraction unit that extracts the first and second edges by performing a filtering process that extracts edges based on the first and second edges, and a labeling process on the extracted first and second edges The first edge by the arm of the operator When recognized as a plurality of edges, the driver determines that the driver is the operator, and when the second edge is recognized as the plurality of edges by the operator's arm, the passenger is An operator discriminating apparatus comprising: an image processing unit including a discriminating unit that discriminates that the user is an operator.

  According to the configuration described above, it is possible to accurately determine the operator who operates the in-vehicle device.

(2) In order to achieve the above object, the present invention is provided at a position where an operator's arm for operating an operation unit of an in-vehicle device provided at a position operable by a driver and a passenger passes through the upper part, A structure having a first edge on the driver side and a second edge on the passenger side, and imaging provided at a position where imaging can be performed so as to include the first and second edges of the structure. And first and second detection frames are set to include the first and second edges based on the image data captured by the image capturing unit and the image capturing unit, and the first and second detection frames are set in the first and second detection frames, respectively. An extraction unit that extracts the first and second edges by performing a filtering process that extracts edges based on the first and second edges, and a labeling process on the extracted first and second edges And the first and second by the arm of the operator When each edge is recognized as a plurality of edges, the extraction unit provides a third detection frame on the driver side and a fourth detection frame on the passenger side between the first and second detection frames. , At a position close to the operation unit, and performing the filtering process for extracting edges within the third and fourth detection frames to extract the third and fourth edges, and then extracting the third edge. When the first pixel number is equal to or larger than the second pixel number based on the first pixel number constituting the second pixel number and the second pixel number constituting the fourth edge, the driver is the operator And an image processing unit including a determination unit that determines that the fellow passenger is the operator when the first number of pixels is smaller than the second number of pixels. An operator discrimination device characterized by the above is provided.

  According to the configuration described above, the operator who operates the in-vehicle device can be determined with higher accuracy.

  According to such a configuration, it is possible to provide an operator discrimination device that accurately discriminates an operator who operates the in-vehicle device.

  Embodiments of an operator discrimination device according to the present invention will be described below in detail with reference to the drawings.

[Embodiment]
(Configuration of vehicle 1)
FIG. 1 is a schematic diagram of a vehicle equipped with an operator discrimination device according to an embodiment of the present invention. The operator discriminating device in the present embodiment is mounted on the vehicle 1 and, as an example, shall discriminate an operator who operates a switch unit 41 of a car navigation unit as an in-vehicle device to be described later. Note that although a car navigation unit will be described as an example of the present embodiment, the present invention is not limited to this and may be applied to an electronic device such as a car audio. In addition, the display unit 40 is divided into two screens on the driver 2A side and the passenger side, different operation menus are displayed on each screen, and the operator who operates the switch unit 41 is determined, so that the determined operator side Only the operation menu displayed on the screen may be operable, but the present invention is not limited to this. Here, the passenger refers to a person sitting next to the driver 2A.

  The vehicle 1 is provided between a vehicle body 10 on which an engine and various electronic devices (not shown) are mounted, a driver seat 11 on which a driver 2A is seated, a driver seat 11 and a passenger seat described later. An armrest 13 as a structure, a CCD (Charge-Coupled Device) camera 31 as an imaging unit provided on the ceiling of the vehicle main body 10 facing the armrest 13, and a car navigation described later that displays images, operation menus, and the like. The display part 40 of a part and the switch part 41 as an operation part of the car navigation part provided in front of the armrest 13 are comprised roughly. The position of the switch unit 41 is not limited to this, and the switch unit 41 may be installed in the vicinity of the display unit 40 or may be a touch panel type display unit, and is not limited thereto.

(Configuration of operator discrimination device 3)
FIG. 2 is a block diagram of the operator discrimination device according to the embodiment of the present invention. The operator discrimination device 3 includes an armrest 13, a CCD camera 31, and an image processing unit 32 that analyzes an image (image data) captured by the CCD camera 31 and discriminates an operator. . The image processing unit 32 further includes an extraction unit 32a that extracts horizontal edges based on the image, and a determination unit 32b that performs labeling processing based on the extracted edges and discriminates the operator. Connected to the unit 4. The image processing unit 32 stores therein an operator determination table 33 (to be described later) serving as a reference for determining an operator. The car navigation unit 4 is schematically configured to include a display unit 40 and a switch unit 41.

  The CCD camera 31 may be a CMOS (Complementary Metal Oxide Semiconductor) camera having a low-cost, low-resolution fixed image sensor. Assume that the CCD camera 31 images the inside of the vehicle based on a predetermined time interval.

(About armrest 13)
FIG. 3 is an upper view inside the vehicle according to the embodiment of the present invention. The vehicle 1 includes a driver seat 11 on which a driver 2A is seated and a passenger seat 12 on which a passenger 2B is seated, and an armrest 13 is provided therebetween. The longitudinal side of the armrest 13 on the driver 2 </ b> A side is defined as a first edge 131, and the longitudinal side on the passenger 2 </ b> B side is defined as a second edge 132. It should be noted that a detection frame, which will be described later, is set and a reference structure for discriminating the operator is not limited to the armrest 13 and has a feature that an edge can be easily extracted, and the edge is divided by the operator's arm. In other words, an edge recognized as one edge may be recognized as a plurality of edges.

(About the operator determination table 33)
FIG. 4 is an operator determination table according to the embodiment of the present invention. The operator determination table 33 is stored in the image processing unit 32 as an example, and the edge cut 330a indicates an edge divided by the operator's arm, and the number of pixels 330b. Is the number of pixels constituting the third and fourth edges, which will be described later, and the judgment 330c is whether the driver 2A or the passenger 2B is based on the edge break 330a and the number of pixels 330b The determination result is shown. The image processing unit 32 refers to the operator determination table 33 based on the obtained result, and determines whether the operator who operates the switch unit 41 is the driver 2A or the passenger 2B.

(Operation)
The operation of the operator discriminating apparatus according to the embodiment of the present invention will be described in detail with reference to FIGS. 1 to 4 and FIGS.

  FIG. 5A is a schematic diagram of an image of the vicinity of the imaged armrest according to the embodiment of the present invention, and FIG. 5B is a diagram within the first detection frame according to the embodiment of the present invention. FIG. 5C shows a binary image, which is a binary image in the second detection frame according to the embodiment of the present invention. First, the binary images 320a and 321a in the first and second detection frames 320 and 321 will be described.

  As illustrated in FIG. 5A, the extraction unit 32 a of the image processing unit 32 places the first and second detection frames 320 and 321 on the first and second armrests 13 on the basis of the image data output from the CCD camera 31. It is set so that the second edges 131 and 132 are included.

  When the first and second detection frames 320 and 321 are set, the extraction unit 32a calculates a luminance differential value in the short direction of the armrest 13 with respect to the image data in the first and second detection frames 320 and 321. Filter processing, and then binarization processing are performed to extract the edge in the longitudinal direction of the armrest 13, and the determination unit 32b performs labeling processing on the extracted edge.

  At this time, in the first detection frame 320, the extraction unit 32a sets the edge having the maximum number of pixels as the edge based on the first edge 131, and for example, the first edge 131a illustrated in FIG. To extract. Further, in the second detection frame 321, the extraction unit 32a sets the edge having the maximum number of pixels as an edge based on the second edge 132, for example, the second edge 132a illustrated in FIG. Extract.

  The image processing unit 32 stores the above extraction result, for example, compares the stored first and second edges 131a and 132a with the newly extracted edge, and the image captured by the CCD camera 31 (image). Data).

(When only the driver's 2A arm 2a is on the armrest 13)
FIG. 6A is a schematic diagram of an image of the vicinity of the armrest imaged according to the embodiment of the present invention, and FIG. 6B is a diagram in the first detection frame according to the embodiment of the present invention. FIG. 6C is a binary image, and is a binary image in the second detection frame according to the embodiment of the present invention. Next, detection of the arm 2a when the driver 2A operates the switch unit 41 will be described.

  When the driver 2A operates the switch unit 41, the arm 2a of the driver 2A passes over the armrest 13, so that an image captured by the CCD camera 31 is as shown in FIG. An image is captured in a state where a part of one edge 131 is hidden. At this time, it is assumed that the arm 2b of the passenger 2B is not on the armrest 13, as shown in FIG. The case where the arm 2a of the driver 2A and the arm 2b of the passenger 2B are simultaneously on the armrest 13 will be described later.

  The extraction unit 32a performs a filtering process for calculating a luminance differential value in the short direction of the armrest 13 on the image in the first detection frame 320, and then performs a binarization process. The determination unit 32b performs a labeling process on the extracted edge, and the first edge 131 is partially hidden by the arm 2a. Therefore, as shown in FIG. The edge 131a is recognized as a plurality of edges of the third and fourth edges 131b and 131c.

  The determination unit 32b divides the first edge 131a when the number of pixels constituting the third and fourth edges 131b and 131c is less than a predetermined number compared to the number of pixels constituting the first edge 131a. And the operator determination table 33 is referred to, and it is determined that the driver 2A is an operator who operates the switch unit 41. It is assumed that the predetermined number can be freely set by the designer of the operator discrimination device 3.

  In the above description, the case where the driver's 2A arm 2a is on the armrest 13 has been described, but the case where only the passenger 2B's arm 2b is on the armrest 13 is also determined by the same method.

(When the driver's 2A arm 2a and the passenger 2B's arm 2b are on the armrest 13)
FIG. 7A (a) is a schematic diagram of an image of the vicinity of the imaged armrest according to the embodiment of the present invention, and FIG. 7A (b) is within the first detection frame according to the embodiment of the present invention. 7A (c) is a binary image in the third detection frame according to the embodiment of the present invention, and FIG. 7B (d) is a second image according to the embodiment of the present invention. FIG. 7B (e) is a binary image in the fourth detection frame according to the embodiment of the present invention. As an example, the binary images 322a and 323a represent 5 × 8 pixels, but the present invention is not limited to this.

  In the above-described example, when the arm 2a of the driver 2A and the arm 2b of the passenger 2B are not on the armrest 13 at the same time, the arm of the operator who operates the switch unit 41 can be detected with high accuracy. However, when the driver's 2A arm 2a and the passenger 2B's arm 2b are on the armrest 13 at the same time, the operator can be identified only by image processing within the first and second detection frames 320 and 321. There is a fear that it cannot be done. Therefore, hereinafter, a method will be described in which the extraction unit 32a sets the third and fourth detection frames 322 and 323 in the armrest 13 on the image and discriminates the operator. Note that the sizes of the third and fourth detection frames 322 and 323 are not limited to this.

  It is assumed that the driver 2A and the passenger 2B are seated in the driver seat 11 and the passenger seat 12, respectively. After the driver 2A turns on the vehicle 1, the CCD camera 31 of the operator discrimination device 3 starts imaging.

  First, the extraction unit 32 a sets the first and second detection frames 320 and 321, and calculates the luminance differential value in the short direction of the armrest 13 with respect to the images in the first and second detection frames 320 and 321. Filter processing, and then binarization processing. The extraction unit 32a extracts the edges in the first and second detection frames 320 and 321, and the determination unit 32b detects whether or not the edges are cut (S1).

  Since the arm 2a of the driver 2A and the arm 2b of the passenger 2B are simultaneously on the armrest 13 as shown in FIG. 7A (a), the determination unit 32b is shown in FIGS. 7A (b) and 7B (d). As shown, the edge of the first edge 131 is divided into third and fourth edges 131b and 131c, and the edge of the second edge 132 is divided into sixth and seventh edges 132b and 132c. Since the operator cannot be discriminated in this state (S2; Yes), the extraction unit 32a is placed in the region near the switch unit 41 on the armrest 13 in the third and fourth detection frames shown in FIG. 7A (a). 322 and 323 are set (S3).

  The extraction unit 32a performs a filtering process for calculating a luminance differential value in the short direction of the armrest 13 on the images in the third and fourth detection frames 322 and 323, and then performs a binarization process. Since the binary images 322a and 323a shown in FIGS. 7A (c) and 7B (e) are obtained based on this binarization processing, the determination unit 32b has a pixel value “1” from the binary images 322a and 323a. The number of pixels constituting the fifth edge 131d (third edge) and the eighth edge 132d (fourth edge) is calculated (pixels painted in black in the binary images 322a and 323a).

  Since the third and fourth detection frames 322 and 323 are set in the vicinity of the switch unit 41 installed on the armrest 13, the pixel value in the third and fourth detection frames 322 and 323 is “1”. It can be considered that the more pixels that are, the edge of the operator's arm is represented. Therefore, the determination unit 32 b calculates a pixel having a pixel value “1” for each of the third and fourth detection frames 322 and 323, and determines the operator based on the number of pixels and the operator determination table 33.

  The number of pixels in which the number of pixels in the third detection frame 322 shown in FIG. 7A (c) is “1” is C (5 in the figure as an example), and the fourth detection frame shown in FIG. 7B (e). Assuming that the number of pixels in which the number of pixels of 323 is “1” is D (11 in the figure as an example), the determination unit 32b is based on the operator determination table 33 (1) When C ≧ D, The operator is determined to be the driver 2A. (2) When C <D, the operator is determined to be the passenger 2B. When C = D, the reason why the determination unit 32b determines that the operator is the driver 2A is that the driver 2A is likely to be driving and the car navigation unit 4 is operated for a long time. This is because it may lead to an accident. Therefore, since it is safer to restrict the operation of the driver 2A, when C = D, the determination unit 32b determines that the operator is the driver 2A and restricts the operation.

  For the above reason, the determination unit 32b calculates the number of pixels constituting the fifth and eighth edges 131d and 132d based on the binary images 322a and 323a shown in FIGS. 7A (c) and 7B (e). Since C <D, the operator is determined to be the passenger 2B based on the operator determination table 33 (S4).

  Here, in step 2, when the first and second edges 131a and 132a in the first and second detection frames 320 and 321 are not cut simultaneously (S2; No), the determination unit 32b proceeds to step 4. Go ahead and determine the operator. When neither the first edge 131a nor the second edge 132a is cut, the determination unit 32b does not perform determination, and the extraction unit 32a determines whether the first and second edges are based on a predetermined time interval. Extraction processing of edges within the detection frames 320 and 321 is performed.

  The image processing unit 32 determines that the operator is the passenger 2B based on the above processing, and then controls the car navigation unit 4 to display the operation menu permitted for the passenger 2B on the display unit 40.

(effect)
According to the above-described embodiment, image processing is performed only on a specific region on the image through which the operator's arm always passes. Therefore, it is determined whether the arm 2a of the driver 2A or the arm 2b of the passenger 2B. This can be determined by a simple algorithm, and a small and inexpensive computer can be used. Further, since a small and inexpensive computer can be used, space and cost are saved.

  The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from or changing the technical idea of the present invention.

It is the schematic of the vehicle carrying the operator discrimination device which concerns on embodiment of this invention. It is a block diagram of the operator discriminating device concerning an embodiment of the invention. It is an upper view inside a vehicle concerning an embodiment of the invention. It is an operator determination table which concerns on embodiment of this invention. (A) is the schematic of the image of the imaged armrest vicinity which concerns on embodiment of this invention, (b) is a binary image in the 1st detection frame which concerns on embodiment of this invention. And (c) is a binary image in the second detection frame according to the embodiment of the present invention. (A) is the schematic of the image of the imaged armrest vicinity which concerns on embodiment of this invention, (b) is a binary image in the 1st detection frame which concerns on embodiment of this invention. And (c) is a binary image in the second detection frame according to the embodiment of the present invention. (A) is the schematic of the image of the imaged armrest vicinity which concerns on embodiment of this invention, (b) is a binary image in the 1st detection frame which concerns on embodiment of this invention. And (c) is a binary image in the third detection frame according to the embodiment of the present invention. (D) is a binary image in the second detection frame according to the embodiment of the present invention, and (e) is a binary image in the fourth detection frame according to the embodiment of the present invention. is there. It is a flowchart which concerns on embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Vehicle, 2A ... Driver, 2B ... Passenger, 2a, 2b ... Arm, 3 ... Operator discrimination | determination apparatus, 4 ... Car navigation part, 10 ... Vehicle main body, 11 ... Driver seat, 12 ... Passenger seat, DESCRIPTION OF SYMBOLS 13 ... Armrest, 31 ... CCD camera, 32 ... Image processing part, 32a ... Extraction part, 32b ... Discrimination part, 33 ... Operator determination table, 40 ... Display part, 41 ... Switch part, 131 ... First edge, 132 2nd edge, 131a ... 1st edge, 131b ... 3rd edge, 131c ... 4th edge, 131d ... 5th edge, 132a ... 2nd edge, 132b ... 6th edge, 132c ... Seventh edge, 132d ... eighth edge, 320 ... first detection frame, 320a ... binary image, 321 ... second detection frame, 321a ... binary image, 322 ... third detection frame, 322a ... Binary image, 323 ... Fourth detection frame, 323a ... 2 binary image, 330a ... edge tear, 330b ... number of pixels, 330c ... determination

Claims (5)

The arm of the operator who operates the operation unit of the in-vehicle device provided at a position operable by the driver and the passenger is provided at a position passing through the upper part, the first edge on the driver side, and the passenger A structure having a second edge on the side;
An imaging unit provided at a position capable of imaging so as to include the first and second edges of the structure;
First and second detection frames are respectively set so as to include the first and second edges based on image data captured by the imaging unit, and the first and second detection frames in the first and second detection frames are set. And an extraction unit for extracting the first and second edges by performing a filtering process for extracting an edge based on the second edge, and a labeling process for the extracted first and second edges, When the first edge is recognized as a plurality of edges by the operator's arm, the driver determines that the operator is the operator, and the operator's arm has a plurality of the second edge. An image processing unit having a determination unit that determines that the fellow passenger is the operator when recognized as the edge;
An operator discrimination device comprising:   The operator discriminating apparatus according to claim 1, wherein the structure is an armrest provided between a driver seat and a passenger seat of a vehicle.   3. The operator determination device according to claim 2, wherein the imaging unit is a charge-coupled device (CCD) camera or a complementary metal oxide semiconductor (CMOS) camera provided on a ceiling of the vehicle. The extraction unit is provided with a third detection frame on the driver side between the first and second detection frames, and a fourth detection frame on the passenger side at a position close to the operation unit, Performing the filtering within the third and fourth detection frames to extract third and fourth edges;
When the first and second edges are respectively recognized as a plurality of the edges, the determination unit includes a first number of pixels constituting the third edge and a second constituting the fourth edge. When the first pixel number is equal to or greater than the second pixel number based on the number of pixels, the driver is determined to be the operator, and the first pixel number is the second pixel number. The operator discriminating apparatus according to claim 3, wherein when the number of passengers is smaller, the passenger is determined to be the operator. The arm of the operator who operates the operation unit of the in-vehicle device provided at a position operable by the driver and the passenger is provided at a position passing through the upper part, the first edge on the driver side, and the passenger A structure having a second edge on the side;
An imaging unit provided at a position capable of imaging so as to include the first and second edges of the structure;
First and second detection frames are respectively set so as to include the first and second edges based on image data captured by the imaging unit, and the first and second detection frames in the first and second detection frames are set. And an extraction unit for extracting the first and second edges by performing a filtering process for extracting an edge based on the second edge, and a labeling process for the extracted first and second edges, When the first and second edges are recognized as a plurality of edges by the operator's arm, a third part is provided on the driver side between the first and second detection frames by the extraction unit. A detection frame and a fourth detection frame on the passenger side are provided at positions close to the operation unit, and the filtering process is performed to extract an edge in the third and fourth detection frames, and the third and fourth detection frames are performed. After extracting 4 edges, When the first pixel number is equal to or greater than the second pixel number based on the first pixel number constituting the edge and the second pixel number constituting the fourth edge, the driver An image processing unit having a determination unit that determines that the passenger is the operator and determines that the fellow passenger is the operator when the first pixel number is smaller than the second pixel number;
An operator discrimination device comprising:

Images