JPH11328364A - Area estimating device and display processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the display processor which can effectively be used to support the visual angle of a driver by displaying an area including the whole body of an object person of detection like a pedestrian discriminatingly from other areas. SOLUTION: The area estimating device decides the head position of the object person of detection on an infrared-ray image photographed by an infrared- ray camera 1 by a head position deciding means 31 and according to information on the head position in the infrared-ray image, an area determining means 32 determines the area equivalent to the body of the object person of detection on the infrared-ray image. An area emphasizing means 33 emphasizes the area equivalent to the body of the object person of detection discriminatingly from other areas and displays this infrared-ray image on a display monitor 4. Consequently, the whole body of the object person of detection like a pedestrian can be emphasized and displayed discriminatingly on the infrared-ray image to be displayed on the display monitor 4.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an area estimating apparatus for determining a body area of a detection target from an infrared image captured by infrared imaging means such as an infrared camera, and a detection target of the detection target determined by the area estimating apparatus. The present invention relates to a display processing device that displays a body region separately from other regions.

[0002]

2. Description of the Related Art As an effective means for detecting the presence of a person by distinguishing it from other objects, there is an infrared sensor including an infrared camera. Since the infrared sensor performs the determination based on the temperature of the target, a simple configuration is possible as compared with a means for examining the shape and movement of the target.

A conventional example using such an infrared detection technique is known from, for example, Japanese Patent Application Laid-Open No. 2-143047. In this conventional example, a room is divided into a plurality of regions, an infrared sensor corresponding to each region is provided, and if a person is present in each region, each person is detected and reflected in the wind direction control of the air conditioner. is there.

Japanese Patent Application Laid-Open No. Hei 6-117,836 discloses that
As another conventional example, there is disclosed a technique of applying a neural network or a pattern recognition mechanism to an infrared image obtained by an infrared sensor to perform detailed human individual identification.

Japanese Patent Laid-Open Publication No. Hei 5-137030 describes another conventional example. This conventional example is a technique for assisting a driver's field of view by detecting infrared rays emitted from the body of a detection target by an infrared camera in a low field of view such as at night and displaying the infrared ray on a display device.

[0006]

However, such a conventional technique has not solved the following points.
That is, since the body of the detection target is usually covered with clothes, infrared rays are not uniformly emitted from the entire body. For this reason, a deviation may occur in the specification of the region corresponding to the entire body. Become controlled, and it may not be possible to adjust the wind direction to warm the entire body.

[0007] Further, when applying to the driver's view assist, it is necessary for the driver to visually recognize a pedestrian-like detection candidate and judge whether or not it is a target to be noted from its shape and movement. It is desirable that the position of the displayed detection candidate is displayed in such a manner that it can be more emphasized in the image so that it can be visually recognized at a glance. desirable. However, since the divergence of infrared light in the clothing part is kept low, in the infrared image taken by the conventional infrared camera, especially in the image of distant pedestrians, only the head with exposed skin is emphasized. In some cases, it may be difficult to determine a streetlight or the like.

The present invention has been made to solve such a conventional technical problem, and an area estimating apparatus capable of detecting an area including the whole body of a detection target using an infrared image. The purpose is to provide.

The present invention also provides a display processing device which can clearly display an area including the entire body of a detection target such as a pedestrian and the like, separately from other areas, and can effectively use it for assisting the driver's visual angle. The purpose is to provide.

[0010]

According to a first aspect of the present invention, there is provided an area estimating apparatus for determining a head position of a person to be detected from a photographing means for photographing an infrared image and the infrared image photographed by the photographing means. Area determining means for determining a region corresponding to the body of the detection target person in the infrared image based on information on a head position in the infrared image determined by the head position determining means. Means.

According to a second aspect of the present invention, in the area estimating apparatus of the first aspect, the area determining means includes a body of the detection target person as a region corresponding to the body of the detection target person. The area is determined.

According to a third aspect of the present invention, in the area estimating apparatus according to the first aspect, the area determining means determines whether or not the area corresponding to the body of the detection target is present on the infrared image. The position, the installation height of the photographing means, the depression angle of the photographing direction, and the height data of the detection target person given in advance are determined.

In the region estimating apparatus according to the first to third aspects of the present invention,
The head position of the person to be detected is determined by the head position determining unit from the infrared image captured by the capturing unit, and the area determining unit determines the detection target in the infrared image based on the information on the head position in the infrared image. The area corresponding to the person's body is determined.

Thus, the position of the whole body of the detection target in the infrared image can be specified, and this information is reflected on, for example, the wind direction control of the air conditioner, thereby not only the head of the person but also the whole body. It becomes possible to control the wind direction toward. Further, if this information is used for assisting the driver's visual angle, it can be applied to classify and highlight the entire body of the detection target person such as a pedestrian from the infrared image displayed on the display means. It is possible to more effectively assist the driver's visual angle.

A display processing device according to a fourth aspect of the present invention includes the region estimating device according to any one of the first to third aspects, and further corresponds to the body of the detection target person determined by the region determining means in the infrared image. The image processing apparatus includes image processing means for dividing an area from other areas, and display means for displaying an output image of the image processing means.

According to a fifth aspect of the present invention, in the display processing apparatus of the fourth aspect, the image processing means (1) specifies a boundary line in a corresponding area in the infrared image, (2) Determining the area in the infrared image by one or a combination of (1) making the brightness or color of the area different from the other areas, and (3) painting an area other than the area in the infrared image with a predetermined color. The region corresponding to the body of the detection target person determined by the means is divided from other regions.

In the display processing device according to the fourth and fifth aspects of the present invention, a region including the entire body of the detection target person in the infrared image is determined by the region estimating device according to the first to third aspects.
The corresponding area is displayed separately from other areas. And
For this display, (1) a boundary line is clearly specified in the corresponding area in the infrared image, (2) the brightness or color of the corresponding area in the infrared image is different from other areas, or (3) One or a combination of a plurality of methods of filling an area other than the corresponding area in the infrared image with a predetermined color is adopted.

Thus, if the present apparatus is used for assisting the driver's visual angle, the whole body of the detection target person such as a pedestrian can be highlighted in the infrared image displayed on the display means. It is possible to effectively assist the driver's viewing angle.

[0019]

As described above, according to the region estimating apparatus according to the first to third aspects of the present invention, the position of the whole body of the detection target in the infrared image can be specified, and this information is used for, for example, an air conditioner. Thus, the wind direction can be controlled not only for the head of the person but also for the whole body. Also, if this information is used to assist the driver's viewing angle,
It can be applied to classify and highlight the entire body of the detection target person such as a pedestrian from the infrared image displayed on the display means, and it is possible to more effectively assist the driver's viewing angle. .

According to the display processing device of the present invention, when the present device is used for assisting the driver's visual angle, a detection target such as a pedestrian is detected from the infrared image displayed on the display means. Of the entire body can be highlighted,
An effective driver's viewing angle assistance is possible.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 shows a hardware configuration of a display processing device according to a first embodiment of the present invention, and FIG. 2 shows a functional configuration thereof. The display processing device according to the first embodiment is mounted on a vehicle, and is an infrared camera 1 that captures a road scene ahead of the vehicle with a predetermined field of view in terms of hardware. The infrared image captured by the infrared camera 1 is digitized. Infrared image frame memory 2 to be stored, an area that includes the whole body of the detection target person such as a pedestrian from an infrared image as an area estimation device, and is further divided into other areas as a display processing device. CPU3 to process to
A display monitor 4 displays an infrared image processed by the CPU 3.

The functions processed by the CPU 3 are shown for each function. As shown in FIG. 2, the infrared image data stored in the infrared image frame memory 2 is fetched.
A head detection processing unit 31 that scans for each pixel, searches for an area where bright parts are concentrated, and determines the area of the head of the detection target person;
For the head region determined by the head detection processing unit 31, the body region determination processing unit 32 determines a region including the body of the detection target person by a method described later, and the detection determined by the body region determination processing unit 32 An area emphasis processing unit 33 is provided which converts the entire body area of the subject into a display form that is divided from other areas by a method described later, and outputs the entire infrared image data.

Although this embodiment relates to a display processing device, the infrared camera 1, the infrared image frame memory 2, and the head detection processing unit 31 and the body region determination processing unit 32 in the CPU 3 perform the area estimation of the present invention. An apparatus is configured, and the present embodiment configures a display processing apparatus using the area estimation apparatus.

Next, the operation of the above-configured region determining device and the display processing device using the same will be described. Infrared camera 1
Captures a road scene ahead of the vehicle by infrared rays with the set field of view and angle of view, and stores them in the infrared image frame memory 2 frame by frame. The CPU 3 fetches infrared image data from the infrared image frame memory 2, determines a region including the body of the pedestrian who is the detection target by each processing shown in the flowchart of FIG. 3, and further displays the region to be separated from other regions. The processing of converting the image data into a form and displaying the image data on the display monitor 4 is repeatedly executed.

For this purpose, first, the head detection processing section 31 fetches one frame of infrared image data from the infrared image frame memory 2, scans each pixel, and detects an area corresponding to the pedestrian's head (step). S1, S2). For this purpose, for example, a technique described in Japanese Patent Application Laid-Open No. 9-128548 is used. Is determined as a region corresponding to the pedestrian's head.

Subsequently, the body region determination processing unit 32
An area corresponding to the pedestrian's body is determined (step S3). The determination of the body equivalent region is performed as follows. Referring to FIG. 4, the actual distance fe from the center of the horizontal angle of view of the image to the pedestrian's head and the height eb of the pedestrian on the screen are:
The following equations 1 to 3 exist between the camera installation height Hcam, the depression angle θ, the height Hped of the pedestrian, and the horizontal distance L from the camera.

[0027]

(Equation 1)

(Equation 2)

(Equation 3) Here, when the average height of an adult is applied as a representative value as the height Hped of the pedestrian, the above-mentioned relational expression between fe and eb can be obtained. On the other hand, referring to FIG. 5, a value corresponding to fe on the screen can be obtained based on the detected position of the pedestrian's head detected by head detection processing unit 31. Thus, the value of eb according to L can be obtained.

Based on the position (= e) corresponding to the pedestrian's head thus obtained and the size of the area in the height direction (= eb), a figure containing the pedestrian is applied on the screen. The area corresponds to the pedestrian's body.

In this embodiment, the following long side e
b, a rectangular area with a short side eb / 2 is used.

(A) The position of the upper side is defined as vertical coordinate e,
(B) The horizontal center line is the horizontal coordinate, (c) the vertical size is eb, and (d) the horizontal size is eb / 2.
And

Instead of using such a rectangular area as an area corresponding to the body of a pedestrian as a detection target, for example, an ellipse having a long axis eb and a short axis eb / 2 and a head position as a vertex. It is also possible to use regions of other shapes, such as regions. When determining a region corresponding to a pedestrian's body on an image, when a pedestrian located at a relatively short distance is set as a detection target, the size of the head region is checked, and the head and body are determined. Assuming that the ratio to the whole, that is, the value of α: β in FIG. 5 is a fixed value, the size eb in the height direction is simply eb =
α + β can also be used. These can be similarly applied to the following second and third embodiments.

When the area corresponding to the head of the pedestrian and the body including the pedestrian is determined in this way, the area emphasis processing section 33 performs emphasis processing so that the area can be clearly distinguished from other areas (step S4). . Here, the emphasis processing is realized by bordering the outline of a rectangular area corresponding to the body with a conspicuous color.

When the emphasizing process is completed, the CPU 3 outputs the emphasized infrared image data to the display monitor 4 and displays it (Step S5).

The image processing by the CPU 3 will be described with reference to an actual screen. FIG. 6A shows a relationship between a head position of a pedestrian who is a detection target, a detection position on a screen, and a region (vertical region) corresponding to a body on an infrared image OG by the infrared camera 1. I have. On the other hand, FIG. 3B shows an example A of an image actually displayed on the display monitor 4. The rectangular area 11 corresponding to the body described above is highlighted in a conspicuous color to display the rectangular area 11 on the screen. The driver can recognize at a glance that the detection target person exists at the position of the area.

As described above, in the display processing device of the first embodiment, the position of the detection candidate including the clothing part other than the head of the pedestrian is determined as the display processing device adapted to assist the driver's view. Highlighting can be performed in the image, which makes it easier for the driver to visually recognize the pedestrian-like detection candidate and determine whether or not it is a target to be noted from its shape or movement.

Next, a display processing apparatus according to a second embodiment of the present invention will be described with reference to FIGS. 1, 2, and 7 to 9. The configuration of the display processing device according to the second embodiment is the same as that of the first embodiment shown in FIGS. However, the display processing executed by the CPU 3 is different. In other words, the region emphasis processing unit 33 of the CPU 3 in the second embodiment performs black-and-white inversion processing only on the region 11 for emphasizing the region 11 corresponding to the body of the detection target person to separate the region from other regions. There is a feature in that. Others are the same as in the first embodiment.

The image processing display by the display processing device according to the second embodiment will be described with reference to the flowchart of FIG. As in the first embodiment, the head detection processing unit 31 detects the pedestrian's head region from the infrared image fetched from the infrared image frame memory 2, and the body region determination processing unit 32 executes the first embodiment. The area 11 corresponding to the pedestrian's body is determined by the same processing as in the embodiment (steps S1 to S3).

Subsequently, the area emphasis processing section 33 performs a black-and-white inversion process on the area 11 determined by the body area determination processing section 32 (step S40), and then displays the entire infrared image on the display monitor 4 (step S40). Step S5).

Here, the black-and-white inversion processing will be described. The infrared image OG is scanned for each pixel, and it is checked whether or not the coordinates of the pixel are included in the area 11 determined in step S3. If it is included, black-and-white inversion processing is performed on the corresponding pixel of the infrared image according to the brightness inversion scale as shown in FIG. Then, the pixel data after the inversion processing is replaced with the data of the corresponding pixel of the original infrared image. This is repeated for all the pixels of the infrared image OG, and when the processing is completed, the entire image is displayed on the display monitor 4. The display image B thus obtained is
The result is as shown in FIG.

Thus, according to the display processing device of the second embodiment, the same effect as that of the first embodiment is obtained, and further, the infrared image B in which only the area corresponding to the body of the detection target is black and white inverted. Can be displayed, so that a pedestrian who does not emit light by itself can be displayed in a form in which it can be displayed in an infrared image brighter than the surrounding area, and can be clearly distinguished from other areas.

In the second embodiment, the area 11 corresponding to the body of the person to be detected is displayed in a black-and-white inversion. However, the present invention is not limited to this. For example, a color that can be distinguished from other areas is used. Alternatively, the brightness may be changed.

Next, a display processing apparatus according to a third embodiment of the present invention will be described with reference to FIGS. 1, 2, 10 and 11. The configuration of the display processing apparatus according to the third embodiment is the same as that of the first embodiment shown in FIGS. However, the display processing executed by the CPU 3 is different, and the region emphasis processing section 33 of the CPU 3 in the third embodiment is different.
Is an area 1 corresponding to the body of the detection target person as described later.
In order to emphasize 1, the corresponding area 11 in the infrared image OG
It is characterized in that a portion other than the region is painted with a predetermined dark color to be separated from other regions. Others
This is the same as in the first embodiment.

An image processing display by the display processing apparatus according to the third embodiment will be described with reference to a flowchart of FIG. As in the first embodiment, the head detection processing unit 31 detects the pedestrian's head region from the infrared image fetched from the infrared image frame memory 2, and the body region determination processing unit 32 executes the first embodiment. The area 11 corresponding to the pedestrian's body is determined by the same processing as in the embodiment (steps S1 to S3).

Subsequently, the region emphasis processing section 33 performs a process of filling a portion other than the area 11 in the infrared image OG with a predetermined dark color in order to emphasize the area 11 determined by the body region determination processing section 32. (Step S41),
Thereafter, the entire infrared image is displayed on the display monitor 4 (step S5). The display image C thus obtained is as shown in FIG.

As described above, according to the display processing device of the third embodiment, the same effects as those of the first embodiment can be obtained, and further, the infrared image in which the area other than the area corresponding to the body of the detection target is darkly filled. Since C can be displayed, an area that does not need to be emphasized becomes a display reminiscent of darkness without any objects outside the vehicle, especially when used during driving at night.
It is possible to make the driver clearly recognize the presence of the pedestrian.

Although the first to third embodiments show the display processing devices of the present invention, the infrared camera 1, the infrared image frame memory 2, and the head of the CPU 3 incorporated in those display processing devices. An area estimation device of the present invention can be configured from the section detection processing section 31 and the body area determination processing section 32, and only this section is used to detect the position of the detection target person for controlling the wind direction of the air conditioner. Therefore, it is possible to control the wind direction not only for the human head but also for the entire body.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a hardware configuration according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating a functional configuration of a CPU in the above embodiment.

FIG. 3 is a flowchart of an image display process of a CPU according to the embodiment.

FIG. 4 is an explanatory diagram of image processing according to the embodiment.

FIG. 5 is an explanatory diagram of image processing according to the embodiment.

FIG. 6 is an explanatory diagram showing an example of a display image according to the embodiment.

FIG. 7 is a flowchart of an image display process of a CPU according to the second embodiment of the present invention.

FIG. 8 is an explanatory diagram of a black-and-white inversion process according to the embodiment.

FIG. 9 is an explanatory diagram showing an example of a display image according to the embodiment.

FIG. 10 is a flowchart of an image display process of a CPU according to the third embodiment of the present invention.

FIG. 11 is an explanatory diagram showing an example of a display image according to the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Infrared camera 2 Infrared image frame memory 3 CPU 4 Display monitor 31 Head detection processing part 32 Body area determination processing part 33 Area enhancement processing part

Claims (5)

[Claims] A photographing means for photographing an infrared image; a head position discriminating means for discriminating a head position of a detection target person from the infrared image photographed by the photographing means; An area estimating apparatus comprising: area determining means for determining an area corresponding to the body of the detection target person in the infrared image based on the information on the determined head position in the infrared image. 2. The apparatus according to claim 1, wherein the area determining unit determines an area equivalent to the body of the detection target person as a region corresponding to the body of the detection target person. Area estimation device. 3. The area determination means determines an area corresponding to the body of the detection target person, the position of the detection target person on the image of the infrared image, the installation height of the imaging means, and the depression angle of the imaging direction. The region estimation apparatus according to claim 1, wherein the determination is performed using height data of the detection target person given in advance. 4. An image processing means for dividing an area corresponding to the body of the detection target person determined by the area determination means in the infrared image from other areas, and an output image of the image processing means is displayed. A display processing device using the area estimation device according to claim 1, further comprising a display unit. 5. The image processing means separates an area corresponding to the body of the detection target person determined by the area determination means from the other areas in the infrared image by one or more of the following combinations: The display processing device according to claim 4, wherein: (1) A boundary line is clearly specified in the corresponding area in the infrared image. (2) The brightness or color of the corresponding area in the infrared image is different from other areas. (3) Other than the corresponding area in the infrared image. Area is painted in a predetermined color.