JP3409538B2 - Human body detection device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a human body detection device for detecting the presence of a human body by infrared rays emitted from the human body, and more particularly to a technique for improving the detection accuracy of the human body.

[0002]

2. Description of the Related Art In a conventional human body detecting apparatus, as shown in FIG. 13, a two-dimensional infrared sensor 131 for detecting infrared rays and converting the intensity of the infrared rays into an electric signal is used to detect infrared rays obtained from a measuring direction. This infrared sensor to detect the intensity
The output signal of 131 is sequentially read by the scanning circuit 132 to generate two-dimensional infrared intensity distribution data in the measurement direction, and the processing unit 133 determines whether or not the infrared intensity distribution data has a region belonging to a predetermined intensity. A method is generally known in which a notification device 134 generates a notification when a region having a predetermined intensity exists.

[0003]

However, in such a conventional human body detecting apparatus, since the object within the predetermined infrared intensity range is detected by extracting from the output of the infrared sensor, the human body detecting apparatus can be used as a human body. It was difficult to accurately detect an object that emitted a very small amount of infrared rays. That is, even if an object satisfying a predetermined condition is detected, the output of the infrared sensor converted into an electric signal is very weak and the SN ratio is low, so that the human body and the object other than the human body cannot be distinguished. Cases often occurred.

The present invention has been made in view of such conventional problems.
An object of the present invention is to provide a human body detection device that improves the detection accuracy of a human body by performing a process that substantially improves the sensitivity of an infrared sensor.

[0005]

In order to achieve the above object, the invention as set forth in claim 1 is a human body detecting apparatus provided with a two-dimensional infrared sensor for measuring the intensity distribution of infrared rays, and the measurement is performed. A first region extracting means for extracting a region belonging to a first intensity range with respect to the infrared intensity distribution; and detecting the intensity of the infrared ray with respect to a region belonging to the first intensity range a predetermined number of times, An integrated intensity distribution measuring unit that obtains an integrated intensity distribution by integrating the detected infrared intensities, and an area that belongs to the second intensity range with respect to the integrated intensity distribution obtained by the integrated intensity distribution measuring unit is extracted. The second area extracting means and the human body determining means for determining the presence or absence of a human body based on the areas belonging to the second intensity range are included.

As described above, the SN ratio of the signal from the infrared sensor is improved by subjecting the region where the human body is considered to exist to the integration process, and the detection sensitivity of the human body can be improved, while it is extracted. Since the integration process is performed on a part of the region, the processing speed can be improved and the human body can be detected more quickly. In the invention according to claim 2, the second strength range is a strength range narrower than the first strength range.

By setting the intensity range closer to the intensity of infrared rays emitted by the human body in this way, the region where the human body is considered to exist can be specified more accurately than the region belonging to the first intensity range. The human body detection accuracy can be improved. The invention according to claim 3 includes means for measuring a distance to an infrared ray generating object, and the human body determining means is based on a relationship between a distance to the object and a size of a region belonging to the second intensity range. The presence of the human body is determined.

By thus determining whether or not the human body is present not only from the area belonging to the second intensity range but also from the distance corresponding to the area, the human body detection accuracy can be further improved. In the invention according to claim 4, the determination signal of the human body determining means is output to the notifying means for notifying when the human body is determined.

With this, when the human body is detected, it is possible to generate a notification notifying the presence of the human body.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Specific embodiments of the present invention will be described below with reference to FIGS. First, the configuration of the first embodiment is shown in the block diagram of FIG. 1 (not included in the dotted line).
Will be explained. The human body detection apparatus according to the present embodiment includes a two-dimensional infrared sensor 101 that detects the intensity of infrared rays emitted from an object and the individual detectors that make up the two-dimensional infrared sensor 101 to scan the detector. The scanning circuit A102 and the scanning circuit B103 which collect the output signal from the sensor and generate the temperature distribution image, the image memory 104 which stores the obtained temperature distribution image, and, for example, 2
An image processing device 105 that performs image processing such as binarization, area extraction, and barycentric position calculation, and the like, and the microcomputer 108 integrally controls these processes.

The image processing apparatus 10 according to the present embodiment
5 may be replaced by the microcomputer 108. Next, a method of detecting a human body by the human body detection device of the present embodiment will be described with reference to the flowchart shown in FIG. First, the infrared sensor 101 is arranged in the direction in which the human body is to be detected. Then, the scanning circuit A102 sequentially scans the individual detectors of the infrared sensor to extract the output signals from the respective detectors (201). The temperature distribution in the detection area can be obtained from the output signal of each detector of the infrared sensor. An example of the scanning method at this time is raster scanning starting from the upper right corner of the screen shown in FIG.

The temperature distribution thus obtained, that is, the intensity distribution of the output signal, is stored in the frame 1 of the image memory 104 (202), and the image processing device 105 causes the temperature range corresponding to the temperature of the human body (body temperature) (for example, A binarization process is performed in which (27-37 ° C.) is set to 1 and other temperature ranges are set to 0 (203). FIG. 4 shows an example of the image subjected to the binarization processing. By performing this binarization processing, the human body candidate regions A41a to 41d, which are temperature ranges corresponding to the temperature of the human body, and other regions
42 can be roughly separated.

Then, for the binarized image shown in FIG.
For example, by performing labeling processing, the human body candidate area A
Individually separate and extract and recognize (204). Further, each of the extracted human body candidate areas A is scanned by the scanning circuit B103 in the order shown in FIG. 5, so that the output of the infrared sensor is stored in the frame 2 of the image memory 104. That is, the data is stored in the corresponding position of the frame 2 with respect to the area where the value of the binarized image of the frame 1 is 1.

On the other hand, similarly, the detection of the output signal from each detector of the infrared sensor is performed by the scanning circuit B103 to detect the human body candidate area A, and the detected output signal is detected in the human body of the frame 2 of the image memory 104. The data is stored in the position corresponding to the candidate area A. As a result, it is possible to omit an area that does not require processing for detecting a human body, so that the calculation load can be reduced and the processing speed can be improved.

The output signal from the infrared sensor for detecting the human body candidate area A is integrated a predetermined number of times by fetching (205). That is, the output signal from the infrared sensor is captured a predetermined number of times, and the sum of the captured signals is added to obtain the sum. By performing this integration processing N times, the SN ratio of the detection value from the infrared sensor for the human body candidate area A can be improved in proportion to N ⁻¹ . Note that this integration processing may be performed using hardware such as an integration circuit.

Next, the integrated data of the frame 2 whose SN ratio has been improved by the integration processing is processed by the image processing device 105 into a temperature range narrower than the threshold value of the binarization processing (for example, 32 ±).
The binarization process is performed again at 1 ° C (206). As a result, it is possible to extract the human body candidate regions B61a, 61b in which the human body shown in FIG. 6 is considered to exist with higher accuracy than the human body candidate region A in FIG. 4 (207).

With respect to each human body candidate region B obtained by the above processing, the length Dx _i in the x direction and the length D in the Y direction are obtained.
y _i and the barycentric positions Cx _i and Cy _i are obtained, respectively.
(208). Obtained Dx _i , Dy _i , and barycentric position C
As shown in FIG. 7, x _i and Cy _i are stored in the frame 3 of the image memory 104 by the number of existing human body candidate regions B. Here, i is an index indicating the order of the human body candidate regions B, and the lengths Dx _i and Dy _i may be, for example, the maximum length for each human body candidate region B.

Next, the microcomputer 108 reads Dx _i , Dy _i and the barycentric positions Cx _i , Cy _i for each human body candidate region B from the frame 3 of the image memory 104. Then, the human body is detected by determining (209) whether or not the human body is present in the human body candidate region B based on these values or the feature amount obtained from these values.

After this determination processing, the infrared sensor 101 is again used.
Return to the measurement step (201) by and repeat the above processing. As described above, according to the present embodiment, the SN of the output signal of the infrared sensor 101 is obtained by performing the integration process on the human body candidate region A in which the human body is considered to exist.
It is possible to improve the ratio and increase the detection accuracy of the human body. In addition, since the integration process is performed on the extracted human body candidate region A, the processing speed can be improved, and thus the human body can be detected more quickly.

Next, in the configuration of the first embodiment, a second embodiment will be described in which it is determined whether or not the human body is also based on the distance to the detected object. The configuration of this embodiment will be described with reference to the block diagram of FIG. 1 (inclusive of the dotted line). The human body detection apparatus according to the present embodiment has the same structure as that of the first embodiment, except that it emits a radar beam such as a near infrared ray and measures the distance from the time until the reflected beam reaches the object to the object. The device 106, a distance data memory 107 that stores distance measurement data for an object in the detection range of the infrared sensor, and a notification device 109 that generates a notification when a human body is detected are added.

The distance measuring device 106 and the distance data memory 1
The detection process by the human body detection device added with 07 is shown in FIG.
It will be described according to the flowchart shown in FIG. Steps 301 to 308 in this flowchart are steps 201 to 208 in FIG. 2 in the first embodiment.
Since it is similar to the above, the description is omitted here. In the present embodiment, the infrared sensor according to the first embodiment
In parallel with the measurement of the temperature distribution by 101, the infrared sensor 10
The distance is measured by the distance measuring device 106 with respect to the measurement range of 1 (309). For example, it is possible to irradiate the measurement range with a radar beam by raster scanning shown in FIG. 3 and measure the distance from the arrival time of the reflected beam at each measurement position. This distance measurement result is stored in the distance data memory 10
Store in 7 according to the measurement position.

Here, the distance data memory 107 is
M × n of the frame 1 of the image memory 104 shown in FIG. 10, which is composed of m × n elements arranged two-dimensionally as shown in FIG.
There is a 1: 1 correspondence with each pixel. That is, the measured value D (x, y) of the distance to the arbitrary measurement position x, y is the intensity T of the infrared rays stored in the frame 1 of the image memory 104.
It corresponds to (x, y).

Next, an image is displayed by the microcomputer 108.
From frame 3 of the image memory 104 to the human body candidate region B
Dx_i, Dy_i, And the center of gravity position Cx_i, Cy_iRead
Including the measured value D (Cx_i, Cy
_i) Is read from the distance data memory 107. Where the object
If you know the distance to the size of the human body on the image to some extent
Since it can be predicted,
Measured distance D (Cx _i, Cy_i) Of the human body corresponding to
Hx on the image in the y and y directions_i, Hy_iSeeking
Meru (310).

These Hx _i and Hy _i are, for example, the difference between Hx _i and Dx _i , and Hy _i and D, with respect to the sizes Dx _i and Dy _{i of the} human body candidate regions B detected by the infrared sensor.
The differences of y _i are calculated and compared (311), and when the respective differences are equal to or less than a predetermined value, that is, the human body candidate region B is approximately the same size as the human body, the human body candidate region B is If it is judged that it exists, the notification device 109 causes the notification to be generated.
(312). After this notification, the process returns to the measurement step (301) by the infrared sensor 101 again. If the size is not the same as the size of the human body, the process returns to the measurement step (301) by the infrared sensor 101 again without generating the notification.

As described above, according to the present embodiment, since the distance to the human body can be measured, it is possible to generate a notification according to the detected distance to the human body.
For example, it is possible to generate a weak notification when the detected human body is at a distant position and give a strong notification as the human body approaches. Further, in addition to the size in the x direction and the y direction of the human body candidate region B extracted by improving the SN ratio of the output signal from the infrared sensor, the determination according to the distance of the human body candidate region B is added. The accuracy can be further increased.

Next, a third embodiment for detecting a human body using two infrared sensors will be described. This third
FIG. 11 is a block diagram showing the configuration of the human body detecting device according to the embodiment. In the present embodiment, an infrared sensor 1 is used instead of the distance measuring device 106 having the configuration of the second embodiment.
It is configured with 01a and 101b. A method of detecting a human body by the human body detecting device using the two infrared sensors 101a and 101b will be described with reference to FIG. First, infrared sensor 10
1a, 101b to the step of FIG. 2 in the first embodiment
The processes from 201 to step 208 are performed to detect the human body candidate regions B123a and 123b of the binarized images 122a and 122b, respectively. The obtained images 122a and 122b are superposed by aligning the positions of the images to obtain the candidate regions B123a and 123b.
The shift amount W of b is determined, and the distance L from the shift amount W to the human body 121 is determined by a map or calculation formula prepared in advance.

From the obtained distance L, step 310 in FIG.
According ～312, the body size of the image Hx _i,
Hy _i is calculated, and these are calculated as the size D of the human body candidate region B.
Compare with x _i , Dy _i . If the size of the human body candidate region B and the size of the human body are substantially equal, the notification is generated, and if they are not equal, the notification is not generated and the measurement by the infrared sensor is repeated again.

As described above, according to the present embodiment, the distance to the human body candidate region B can be obtained with a simple configuration by measuring the distance from the two infrared sensors provided at different positions to the object. The human body can be detected accurately.

[0029]

According to the first aspect of the present invention, the SN ratio of the signal from the infrared sensor is improved by performing the integration process on the region where the human body is considered to exist, and the detection accuracy of the human body is improved. On the other hand, since the integration processing is performed on the extracted partial area, the processing speed can be improved and the human body can be detected more quickly.

According to the second aspect of the present invention, the second intensity range is set to a narrower intensity range than the first intensity range, whereby the infrared intensity range closer to the intensity of infrared rays emitted by the human body is set. The region belonging to the second intensity range can be specified more accurately than the region belonging to the first intensity range, and the region in which the human body is considered to exist can be specified more accurately, thereby further improving the detection accuracy of the human body. You can

According to the third aspect of the present invention, the human body detection accuracy is further improved by determining whether or not the human body is present not only from the area belonging to the second intensity range but also from the distance corresponding to the area. Can be made. In the invention according to claim 4, the determination signal of the human body determining means is output to the notifying means for notifying when the human body is determined.

Thus, when the human body is detected, it is possible to generate a notification for notifying the presence of the human body.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration / function in the first and second embodiments.

FIG. 2 is a flowchart of human body detection processing according to the first embodiment.

FIG. 3 is a diagram for explaining a scanning method when detecting an output signal of an infrared sensor.

FIG. 4 is a diagram in which an intensity distribution of a signal from an infrared sensor is binarized.

FIG. 5 is a diagram illustrating a method of scanning a human body candidate region A after binarization processing.

FIG. 6 is a diagram showing a result of extraction of a human body candidate region B.

FIG. 7 is a diagram showing a configuration of a frame 3 of the image memory.

FIG. 8 is a flowchart of human body detection processing according to the second embodiment.

FIG. 9 is a diagram showing a configuration of a distance measurement memory.

FIG. 10 is a diagram showing a configuration of a frame 1 of an image memory.

FIG. 11 is a block diagram showing the configuration and functions of the third embodiment.

FIG. 12 is a diagram illustrating a distance detecting method according to a third embodiment.

FIG. 13 is a block diagram showing the configuration and functions of a conventional human body detection device.

[Explanation of symbols]

41a, 41b, 41c, 41d Human body candidate area A 61a, 62b Human body candidate area B 101,101a, 101b Infrared sensor 131 human body

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G01V 8/20 G01J 1/02 G01K 13/04 G06T 1/00 G08B 13/18

Claims (4)

(57) [Claims] 1. A human body detection device equipped with a two-dimensional infrared sensor for measuring an infrared intensity distribution, wherein a first region for extracting a region belonging to a first intensity range from the measured infrared intensity distribution is detected. Area extraction means, integrated intensity distribution measuring means for detecting the intensity of the infrared rays for a predetermined number of times in the area belonging to the first intensity range, and integrating the detected infrared rays to obtain an integrated intensity distribution. A second area extracting means for extracting an area belonging to a second intensity range from the integrated intensity distribution obtained by the integrated intensity distribution measuring means; and a human body based on the area belonging to the second intensity range. A human body detecting device comprising: a human body determining means for determining the presence or absence of the human body. 2. The human body detection device according to claim 1, wherein the second intensity range is an intensity range narrower than the first intensity range. 3. The human body determining means includes means for measuring a distance to an infrared ray generating object, and the human body judging means detects the presence of the human body based on the relationship between the distance to the object and the size of a region belonging to the second intensity range. The human body detection device according to claim 1, wherein the human body detection device according to claim 1 or 2. 4. The human body detection device according to claim 1, wherein the determination signal of the human body determination means is output to an informing means for informing at the time of human body determination.