JP2018129090A - Detection apparatus, detection system, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a detection apparatus, a detection system, and a program for detecting presence of a person with high accuracy in accordance with a moving state of the person, while reducing the load on detection processing.SOLUTION: A detection apparatus 100 includes a detection unit 31 for detecting presence of a person in a detection area in an analysis image captured by an imaging unit 2. The detection unit 31 has moving detection mode that detects presence of a person in the detection area, and stay detection mode that detects presence of a person, on the basis of a moving amount of the person in a stay area in which the person is estimated to stay in the detection area.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a detection device, a detection system, and a program.

  Conventionally, in order to detect an intruder, the area of the image in the difference image is obtained with respect to the movement of the intruder, and if the area of the image movement due to the movement of the intruder is not less than a certain value so as to distinguish it from disturbance, A detection device that detects an intruder is known (see, for example, Patent Document 1).

  In addition, an image processing apparatus having a first contour extracting unit, a second contour extracting unit, and a third contour extracting unit has been proposed (see, for example, Patent Document 2). The first contour extraction unit detects the detection target when the movement amount and movement of the detection target are large, and the second contour extraction unit selects the detection target when the movement amount and movement of the detection target are relatively small. The third contour extracting means detects the detection target when the detection target is stationary. This image processing apparatus includes a plurality of detection modes for detecting the presence / absence of a detection target in a detection range in which the size of the detection region is different using each contour extraction unit, so that even if the movement of the detection target is fast, The detection object can be detected even when it is slow or stationary.

JP 2000-348268 A JP 2004-295416 A

  However, when the intruder moves little or stays within the detection region, the image described in Patent Document 1 is recognized as a disturbance because the area of the image movement due to the intruder movement is reduced. End up. Therefore, the thing of patent document 1 may become unable to detect an intruder when the intruder moves little or stays in the detection area.

  Moreover, the thing of patent document 2 always detects the presence or absence of a detection target by repeating the 1st outline extraction means-the 3rd outline extraction means. For example, when the detection target moves greatly, the image processing apparatus described in Patent Document 2 first determines whether or not the detection target is stationary, and then determines whether or not the detection target is moving slightly. Since it is determined whether or not a large movement is occurring, the load on the detection process is large.

  Accordingly, the present invention has been made in view of the above-described reasons, and the detection device and the detection capable of detecting the presence or absence of a person with high accuracy according to the movement state of the person with a small load on the detection process. The purpose is to provide a system.

  The detection apparatus of the present invention includes a detection unit that detects the presence or absence of a person in the detection region of the analysis image captured by the imaging unit, and the detection unit detects movement of the person in the detection region. And a stay detection mode for detecting the presence or absence of a person based on the amount of movement of the person in the stay area where the person is estimated to be stagnant in the detection area.

  The detection system of this invention is provided with a detection apparatus and the load controlled based on the detection result of a detection part.

  The program of the present invention includes a detection unit that detects the presence or absence of a person in the detection region of the analysis image, and the detection unit includes a movement detection mode that detects the presence or absence of a person in the detection region; A program for functioning as a detection device having a stay detection mode for detecting the presence or absence of a person based on the amount of movement of a person in a stay area where a person is estimated to be stagnant in the detection area .

  The detection device and the detection system of the present invention have a small load on the detection process, and can accurately detect the presence or absence of a person according to the movement situation of the person.

FIG. 1 is a block diagram of a detection device 100 and a detection system 120 according to Embodiment 1 of the present invention. 2A and 2B show an example of the identification region 22 created by the creation unit 32 of the detection apparatus 100 according to Embodiment 1 of the present invention for the analysis image, and the derivation unit 33 derives the positional change amount of the identification region 22. It is a conceptual diagram which shows an example to do. FIG. 2A is a conceptual diagram showing the detection area 21 of the analysis image in a frame before two consecutive frames of the analysis image. FIG. 2B is a conceptual diagram showing the detection region 21 of the analysis image in a frame after two consecutive frames of the analysis image. 3A and 3B are conceptual diagrams illustrating an example in which the deriving unit 33 derives the amount of movement of the person when the detection apparatus 100 according to Embodiment 1 of the present invention detects the presence or absence of a person in the stay detection mode. is there. FIG. 3A is a conceptual diagram showing a stay area 23a in a frame before two consecutive frames according to Embodiment 1 of the present invention. FIG. 3B is a conceptual diagram showing a stay area 23b in a frame after two consecutive frames according to Embodiment 1 of the present invention. FIG. 4 is a flowchart in which the detection apparatus 100 according to the first embodiment of the present invention switches the detection mode based on the position change amount of the identification region 22. FIG. 5 is a flowchart in which the detection apparatus 100 according to the first embodiment of the present invention detects the presence or absence of a person in the movement detection mode. FIG. 6 is a flowchart in which the detection apparatus 100 according to the first embodiment of the present invention detects the presence or absence of a person in the stay detection mode. FIG. 7 is a block diagram of the detection device 200 and the detection system 220 according to Embodiment 2 of the present invention. FIG. 8 is a conceptual diagram showing an example of filter image processing according to the second embodiment of the present invention. FIG. 9 is a flowchart in which the detection device 200 according to the second embodiment of the present invention detects the presence or absence of a person in the stay detection mode. FIG. 10 is a block diagram of a detection device 600 and a detection system 620 according to Embodiment 6 of the present invention. FIG. 11 is a flowchart of an operation when the detection apparatus 600 according to the sixth embodiment of the present invention switches the detection mode based on the position change amount of the identification region 22. 12A, 12B, and 12C are diagrams showing analysis images of three consecutive frames when a person 70 enters the detection area 21 from outside the detection area 21 and stops there. FIG. 13 is a conceptual diagram illustrating an example of an operation when the detection apparatus 600 according to the sixth embodiment of the present invention detects the presence or absence of a person in the stay detection mode. FIG. 14 is a conceptual diagram illustrating an example of an operation when the detection apparatus 600 according to the sixth embodiment of the present invention detects the presence or absence of a person in the stay detection mode. FIG. 15 is a conceptual diagram illustrating an example of the relationship between the detection region 21 in the analysis image and the upper limit frame 24 that defines the upper limit value by the restriction unit 36.

(Embodiment 1)
Hereinafter, a detection apparatus and a detection system according to Embodiment 1 of the present invention will be described with reference to FIGS.

  FIG. 1 is a block diagram of a detection device 100 and a detection system 120 according to Embodiment 1 of the present invention. As shown in FIG. 1, the detection apparatus 100 includes an imaging unit 2, a control unit 3, a transmission unit 4 that transmits a control signal to a control target according to the presence or absence of a person, a storage unit 5, Is provided.

  The control unit 3 includes a detection unit 31, a creation unit 32, and a derivation unit 33. The detection unit 31 detects the presence or absence of a person in the detection area of the analysis image captured by the imaging unit 2. When the detection unit 31 detects that a person is present in the detection area of the analysis image, the generation unit 32 generates an identification area surrounding the detected person in the analysis image. The deriving unit 33 derives the amount of movement of the person in the stay area including the identification area stagnated in the detection area and the position change amount of the identification area in the detection area.

  The deriving unit 33 derives the amount of movement of the person within the stay area during the predetermined period. The deriving unit 33 derives the position change amount of the identification area within the detection area in a predetermined period. For convenience of explanation, the above-mentioned “predetermined period” will be described as a frame interval between two consecutive frames of the analysis image captured by the imaging unit 2, but the present invention is not limited thereto, and for example, a preset period may be used. . Further, the predetermined period for deriving the amount of movement of the person in the stay area and the predetermined period for deriving the position change amount of the identification area may be the same period or different periods.

  Note that the detection area of the analysis image may be, for example, a partial area of the image captured by the imaging unit 2 or may be the entire area of the image captured by the imaging unit 2.

  The storage unit 5 includes a stagnation threshold “I” (first threshold) as an evaluation index, a cancellation threshold “U” (second threshold), and a detection threshold “T” (third) that is smaller than the cancellation threshold “U”. And a derivation threshold “K” (fourth threshold) are stored.

  The detection unit 31 detects, as detection modes, a movement detection mode that detects the presence or absence of a person in the detection area of the analysis image, and a stay detection that detects the presence or absence of a person based on the amount of movement of the person in the stay area. Mode. When the position change amount of the identification area becomes less than the stagnation threshold “I” in the movement detection mode, the detection unit 31 fixes the position of the identification area in the detection area, sets the identification area as the stay area, The presence / absence of a person is detected in the stay detection mode for the area. That is, the stay area is an identification area that is regarded as stagnant in the detection area because the amount of change in position in the detection area is less than the stagnation threshold “I”.

  Here, as long as the detection unit 31 operates in the movement detection mode, the creation unit 32 detects by updating the position of the identification region as the person moves within the detection region of the analysis image of the detection device 100. The position of the person in the area is tracked in the identification area. That is, in the movement detection mode, the detection unit 31 detects the presence / absence of a person over the entire detection area, and continues by tracking the position of the person even when the person moves greatly within the detection area. The presence or absence of the person is detected. On the other hand, in the stay detection mode, the detection unit 31 targets a stay area that is an identification area stagnated in the detection area, that is, a stay area that is an identification area whose position in the detection area is fixed. Detect presence or absence. Therefore, in the stay detection mode, the detection unit 31 detects the slight movement of the person in the stay area even when the person is stopped at one place in the detection area (stagnation). The presence or absence of a person can be detected.

  However, the detection unit 31 does not completely switch between the movement detection mode and the stay detection mode, but basically operates in the movement detection mode. When the stay area appears in the detection area (when the stagnation of the identification area occurs), the detection unit 31 detects the presence or absence of a person in the stay detection mode only in the stay area. The area detects the presence or absence of a person in the movement detection mode. That is, when the stay area appears in the detection area, the detection unit 31 partially sets the detection mode to the stay detection mode only for the stay area, and the detection mode is completely changed from the movement detection mode to the stay detection mode. It doesn't switch.

  In addition, the detection method of the presence / absence of a person in the movement detection mode in the detection unit 31, that is, the detection method of the presence / absence of a person in the detection unit 31 used for creating the identification area in the creation unit 32 is particularly limited. Instead, a conventional human detection technique may be used. For example, a technique for detecting the presence or absence of a person using the background difference may be used, or a technique for detecting the presence or absence of a person by deriving the amount of movement of the person in the detection area 21 of the analysis image using the difference between frames. But you can.

  2A, 2B, 3A, and 3B will be described below with respect to the creation method of the identification region of the creation unit 32 and the method of deriving the amount of movement of the person and the position change amount of the identification region in the stay region of the derivation unit 33. Will be described.

  2A and 2B are conceptual diagrams illustrating an example of the identification region 22 created by the creation unit 32 and an example of the derivation unit 33 deriving the positional change amount of the identification region 22. 2A and 2B are diagrams showing the detection region 21 of the analysis image of two consecutive frames.

  As shown in FIG. 2A, when the detection unit 31 detects the presence of a person in the movement detection mode, the creation unit 32 creates a rectangular identification region 22 based on the width and height of the person. Then, the analysis image is created so as to surround the person. That is, the identification area 22 is an area created so as to include a person detected by the detection unit 31 in the detection area 21 of the analysis image.

  FIG. 2B shows the next frame of FIG. 2A, in which case the identification area 22 has moved from the identification area 22a in FIG. 2A to the identification area 22b in FIG. 2B to assume that a person is moving. To do.

  The deriving unit 33 calculates the distance L1 between the center of the identification area 22a and the center of 22b, and derives the distance L1 as the position change amount of the identification area 22.

  The detection unit 31 compares the position change amount derived by the deriving unit 33 with the stagnation threshold “I” stored in the storage unit 5 in the movement detection mode. When the position change amount is equal to or greater than the stagnation threshold “I”, the detection unit 31 detects the presence or absence of a person in the movement detection mode, and partially detects when the position change amount is less than the stagnation threshold “I”. The mode is the stay detection mode.

  The creation unit 32 updates the position of the identification region 22 as the person moves in the detection region 21 of the analysis image of the detection device 100 so that the detection unit 31 tracks the person in the detection region of the analysis image. Can do. In the movement detection mode, the detection unit 31 detects that there is a person in the detection area 21 if the identification area 22 exists in the detection area 21 in the analysis image, and the identification area 22 exists in the detection area 21. If not, it is detected that there is no person in the detection area 21. That is, in any case of FIG. 2A and FIG. 2B, the detection part 31 detects that there exists a person.

  3A and 3B are conceptual diagrams illustrating an example in which the deriving unit 33 derives the amount of movement of a person when the detection apparatus 100 detects the presence or absence of a person in the stay detection mode. 3A and 3B are diagrams showing stay areas 23a and 23b of two consecutive frames. In FIG. 3A and FIG. 3B, since the identification area is stagnant in the detection area 21, the identification areas indicated by “23 a” and “23 b” are both stay areas 23.

  As shown in FIGS. 3A and 3B, in the stay detection mode, the detection unit 31 detects the presence or absence of a person based on the amount of movement of the person in the stagnant identification area 22, that is, the stay area 23. As shown in FIGS. 3A and 3B, when the movement of a person changes from the stay area 23a in FIG. 3A to the stay area 23b in FIG. 3B, the luminance of the pixels in the stay areas 23a and 23b varies with the movement of the person. Change based on.

  The deriving unit 33 derives the amount of human movement in the stay area 23 (stay areas 23a and 23b) in a predetermined period. That is, the deriving unit 33 derives, as the amount of human movement, the number of pixels having a luminance change amount from the stay region 23a that is equal to or greater than the deriving threshold “K” stored in the storage unit 5 among the pixels in the stay region 23b. To do.

  In the stay detection mode, the detection unit 31 derives the amount of human movement by the deriving unit 33, thereby detecting the amount of human movement, the detection threshold “T” stored in the storage unit 5, and the release threshold “U”. , Compare. The detection unit 31 determines that there is a person if the amount of movement of the person is greater than or equal to the detection threshold “T” and less than the release threshold “U”. The detection unit 31 determines that there is no person if the movement amount of the person is less than the detection threshold “T”. That is, in the stay detection mode, the detection unit 31 determines that there is no person if the amount of human movement is less than the detection threshold “T”, and if the amount of human movement is equal to or greater than the detection threshold “T”. Determine that there is a person.

  Incidentally, in the present embodiment, the detection unit 31 switches from the stay detection mode to the movement detection mode if the amount of human movement is equal to or greater than the release threshold “U” that is greater than the detection threshold “T”. That is, in the stay detection mode, the detection unit 31 compares the release threshold “U”, which is larger than the detection threshold “T” used for detecting the presence or absence of a person, with the amount of movement of the person in the stay area. To do. If the movement amount of the person is less than the release threshold “U”, the detection unit 31 continuously detects the presence / absence of the person in the stay detection mode in the stay area. When the amount of movement of the person becomes equal to or greater than the release threshold “U”, the detection unit 31 releases the stay area and switches the detection mode in the stay area from the stay detection mode to the movement detection mode.

  In addition, the cancellation | release of a stay area here means deleting the stay area set in the detection area. When the stay area is canceled, the detection unit 31 may return the stay area to the identification area, or may simply cancel the setting of the stay area without returning the stay area to the identification area. In the former case, when the stay area is released, an identification area instead of the stay area appears at the same position as the stay area in the detection area. In the latter case, when the stay area is released, neither the stay area nor the identification area remains at the same position in the detection area as the stay area.

  In addition, in a state where stay areas are set at a plurality of locations in the detection area, the detection unit 31 may perform switching from the stay detection mode to the movement detection mode for each stay area or all stays. It may be performed simultaneously for the region.

  Thus, if the amount of movement of the person is equal to or greater than the release threshold “U”, the detection mode is switched from the stay detection mode to the movement detection mode, so that when the person moves from the stay area to the outside of the stay area, The detection unit 31 can detect the presence of the person without losing sight.

  The load 6 is assumed to be a lighting device having at least two operation modes of “lighting” and “lighting off”.

  If the detection unit 31 determines that there is a person when the load 6 is in a state where the load 6 is turned off, the transmission unit 4 transmits a control signal that is lit to the load 6 to be controlled outside the detection apparatus 100. To do. When the load 6 is in the “lighting” operation mode, the transmission unit 4 transmits a control signal for switching to the “light-off” operation mode to the load 6 when the detection unit 31 determines that there is no person.

  In addition, the transmission part 4 may transmit a control signal directly with respect to the load 6, and may transmit a control signal indirectly via a repeater etc.

  The communication method of the transmission unit 4 may be, for example, a wireless communication method using Wi-Fi (Wireless Fidelity, registered trademark) or a wired communication method using a signal line, but is not particularly limited.

  Hereinafter, an example of the operation of the detection apparatus 100 will be described using the flowcharts shown in FIGS.

  FIG. 4 is a flowchart in which the detection apparatus 100 switches the detection mode based on the position change amount of the identification region 22.

  Before a person enters the detection area of the analysis image captured by the imaging unit 2 of the detection apparatus 100, the detection unit 31 detects the presence or absence of a person in the movement detection mode (step S41). Therefore, when detecting the presence of a person in the detection region of the analysis image, the detection unit 31 compares the position change amount of the identification region 22 derived by the deriving unit 33 with the stagnation threshold “I”. (Step S42).

  If the position change amount of the identification area 22 is less than I (NO in step S42), the detection unit 31 proceeds to step S43, and switches the detection mode from the movement detection mode to the stay detection mode. That is, if there is an identification area 22 whose position change amount is less than the stagnation threshold “I”, the detection unit 31 uses the identification area 22 as a stay area and detects the presence or absence of a person in the stay detection mode only for the stay area. As described above, the detection mode is partially set as the stay detection mode. If the amount of change in the position of the identification region 22 is equal to or greater than the stagnation threshold “I” (YES in step S42), the detection unit 31 returns to step S41 and continuously detects the presence or absence of a person in the movement detection mode.

  In step S43, the detection unit 31 detects the presence or absence of a person in the stay detection mode. Therefore, the detection unit 31 compares the amount of human motion derived by the deriving unit 33 with the release threshold “U” (step S44).

  If the movement amount of the person is less than the release threshold “U” (YES in step S44), the detection unit 31 continues to detect the presence or absence of the person in the stay detection mode. Otherwise, that is, if the amount of human movement is equal to or greater than the release threshold “U” (NO in step S44), the detection unit 31 switches the detection mode from the stay detection mode to the movement detection mode in step S45. Return to step S41. At this time, for the areas other than the stay area in the detection area, the detection unit 31 originally detects the presence or absence of a person in the movement detection mode, so the detection mode switches from the stay detection mode to the movement detection mode. This is only within the area of stay. In other words, after setting the detection mode in the stay area to the stay detection mode in step S43, the detection unit 31 releases the stay area in step S45 and switches the detection mode in the stay area to the movement detection mode.

  FIG. 5 is a flowchart in which the detection device 100 detects the presence or absence of a person in the movement detection mode.

  In step S51, the detection unit 31 detects the presence or absence of a person based on the analysis image captured by the imaging unit 2. At this time, for example, the detection unit 31 detects the presence or absence of a person by using a background difference or deriving the amount of movement of a person in the detection region of the analysis image using a difference between frames. Then, when the detection unit 31 detects that there is a person in the detection region of the analysis image (YES in step S52), the creation unit 32 creates the identification region 22 (step S53).

  In step S52, when detecting that there is no person (NO in step S52), the detection unit 31 returns to step S51 and subsequently detects the presence or absence of a person in the movement detection mode.

  In step S <b> 54, the detection unit 31 compares the position change amount of the identification area 22 derived by the deriving unit 33 with the stagnation threshold “I”.

  If the position change amount is equal to or greater than the stagnation threshold “I” (YES in step S54), the detection unit 31 returns to step S51 and subsequently detects the presence or absence of a person in the movement detection mode. If the amount of change in position is less than the stagnation threshold “I” (NO in step S54), the detection unit 31 switches the detection mode from the movement detection mode to the stay detection mode (step S55). Detect the presence or absence. As described above, step S54 is substantially the same as step S42.

  FIG. 6 is a flowchart in which the detection apparatus 100 detects the presence or absence of a person in the stay detection mode.

  In step S <b> 61, the detection unit 31 detects the presence or absence of a person based on the analysis image in the identification area 22. At this time, the detection unit 31 compares the amount of human movement in the stagnation area derived by the deriving unit 33 with the detection threshold “T” (step S62).

  If the amount of movement of the person is less than the detection threshold “T” (NO in step S62), the detection unit 31 determines that there is no person (step S64). If the human motion amount is equal to or greater than the detection threshold “T” (YES in step S62), the detection unit 31 compares the human motion amount with the release threshold “U” (step S63).

  In step S63, if the amount of movement of the person is less than the release threshold “U” (YES in step S63), the detection unit 31 determines that there is a person (step S65) and continues to detect the person in the stay detection mode. Detect presence or absence. If the movement amount of the person is equal to or greater than the release threshold “U” (NO in step S63), the detection unit 31 switches the detection mode in the stay area from the stay detection mode to the movement detection mode, and the presence of the person in the movement detection mode. Is detected (step S66).

  When detecting the presence / absence of a person in the stay area in the stay detection mode, the detection unit 31 detects the presence / absence of another person outside the stay area and in the detection area 21 of the analysis image in the movement detection mode. To do. Therefore, when a person other than the person in the stay area enters the detection area 21 of the analysis image, the detection unit 31 can detect the presence of the person in the movement detection mode.

  In addition, when the detection area 21 of the analysis image is a partial area of the image captured by the imaging unit 2, the detection unit 31 includes the presence or absence of a person included in the analysis image even outside the detection area 21. Is detected. When the detection unit 31 detects the presence of a person outside the detection region 21, the creation unit 32 creates an identification region 22 that surrounds the person. Thereby, when a person enters the detection area 21 from outside the detection area 21, the identification area 22 enters the detection area 21 from outside the detection area 21 in the analysis image so as to track the person. Therefore, the detection unit 31 may be configured to detect the presence or absence of a person in the detection region 21 based on the presence or absence of the identification region 22 in the detection region 21 in the movement detection mode. Thereby, the detection part 31 can distinguish and detect the person in the detection area 21 and the person outside the detection area 21. Note that the detection unit 31 preferably determines that the identification area 22 exists in the detection area 21 when the identification area 22 partially overlaps the detection area 21.

  The detection system 120 includes an imaging unit 2, a detection unit 31, a creation unit 32, a derivation unit 33, and a load 6 to be controlled. The detection unit 31 detects the presence or absence of a person in the detection area of the analysis image captured by the imaging unit 2. When the detection unit 31 detects that a person is present in the detection area 21 of the analysis image, the generation unit 32 generates an identification area 22 surrounding the detected person in the analysis image. The deriving unit 33 derives the amount of movement of the person in the stay area composed of the stagnant identification area 22 in the detection area 21 and the position change amount of the identification area 22 in the detection area 21. The detection system 120 further includes a transmission unit 4 that transmits a control signal to a control target based on the detection result of the presence or absence of a person by the detection unit 31. The detection unit 31 includes a movement detection mode that detects the presence or absence of a person in the detection area 21 of the analysis image, a stay detection mode that detects the presence or absence of a person based on the amount of movement of the person in the stay area, Have In the movement detection mode, when the position change amount of the identification area 22 becomes less than the stagnation threshold “I”, the detection unit 31 fixes the position of the identification area 22 in the detection area 21 and makes the identification area 22 a stay area. The presence / absence of a person is detected in the stay detection mode for the stay area.

  With the above configuration, the detection device 100 and the detection system 120 can achieve the following effects as compared with those described in Patent Document 2.

  The image processing apparatus described in Patent Document 2 includes first contour extraction means, second contour extraction means, and third contour extraction means. The first contour extraction means detects the detection target when the amount of movement or movement of the detection target is large. The second contour extraction unit detects the detection target when the amount of movement or movement of the detection target is relatively small. The second contour extracting means detects the detection target when the detection target is stationary. However, the image processing apparatus described in Patent Document 2 always detects the presence / absence of a detection target by repeating the first to third contour extracting means. For example, even if the detection target moves greatly, the image processing apparatus described in Patent Document 2 does not always detect the detection target with the first contour extraction unit. That is, even if the detection target is moving greatly, the image processing apparatus described in Patent Document 2 first determines whether or not the detection target is stationary, and then determines whether or not the detection target is moving slightly. After that, the load on the detection process is large in order to determine whether or not there is a large movement.

  In contrast, in the detection device 100 and the detection system 120, if the position change amount of the identification region 22 is less than the stagnation threshold “I” in the movement detection mode, the detection unit 31 switches to the stay detection mode to Detect presence or absence. That is, when the amount of movement of the person is relatively small or staying, the detection device 100 and the detection system 120 switch to the stay detection mode and detect the presence or absence of the person. Therefore, since the amount of processing by the image processing apparatus described in Patent Document 2 is small, the load on the detection process is small.

  Therefore, the detection apparatus 100 and the detection system 120 have a small load on the detection process, and can accurately detect the presence or absence of a person according to a human action.

  In addition, when the detection unit 31 of the detection device 100 and the detection system 120 detects a person whose movement amount is relatively small or stays in the stay detection mode, the presence of a person based on the analysis image in the identification region 22 is detected. The presence or absence of is detected. Therefore, even if the detection unit 31 of the detection device 100 and the detection system 120 has a disturbance outside the area of the identification area 22, the amount of movement is relatively small without being affected, or there is a person who is staying. The presence or absence of can be detected. For example, when an animal such as a bird or a cat enters the detection range, or when reflected light from a switch or a mirror is generated, the detection unit 100 of the detection device 100 and the detection system 120 is not affected by the human The presence or absence can be detected.

  Further, the deriving unit 33 derives the number of pixels in which the amount of change in pixel luminance in the stay region between frames is equal to or greater than the deriving threshold “K” as the amount of human movement. Therefore, even if a pixel whose luminance change amount is less than the derived threshold “K” occurs due to disturbance in the stay region, the detection unit 31 can detect the presence or absence of a person without being affected by the pixel.

  As described above, the detection device 100 and the detection system 120 can detect the presence or absence of a person with high accuracy and efficiency even when the movement amount of the person is relatively small or stays within the detection range.

  For convenience of explanation, the detection unit 31, the creation unit 32, and the derivation unit 33 are provided in the control unit 3, and the transmission unit 4 and the storage unit 5 are externally attached to the control unit 3. Because it is a matter, it is not limited to that. For example, the transmission unit 4 and the storage unit 5 may be provided in the control unit 3. The transmission unit 4 can be omitted as appropriate. When the transmission unit 4 is omitted, the detection apparatus 100 directly controls the load 6 based on the detection result of the presence or absence of a person in the detection unit 31. Configured to do.

  Further, the load 6 is not limited to the lighting device, and may be an electric device having at least two operation modes such as an air conditioner and a cooking device.

  Further, the frame interval of the imaging unit 2, that is, the period between FIGS. 2A and 2B and the period between FIGS. 3A and 3B are, for example, 100 ms to several hundred ms, but are not limited thereto.

  The deriving unit 33 does not necessarily use the center of the identification area 22 as a reference point when deriving the position change amount of the identification area 22 in the detection area 21. That is, for example, the derivation unit 33 may be configured to calculate the distance between the representative points of the identification area 22 before and after moving and derive this distance as the position change amount of the identification area 22. The representative points here are, for example, the center of gravity and the apex of the identification region 22.

  The transmission unit 4 sends different control signals to the load 6 when the detection unit 31 determines that there is a person in the stay detection mode and when it is determined that there is a person in the movement detection mode. You may send it. That is, when the detection unit 31 detects that there is a person in the detection region 21, the final determination result (control signal) differs depending on whether the detection mode of the detection unit 31 is the stay detection mode or the movement detection mode. Also good. Specifically, when the detection unit 31 detects that there is a person in the movement detection mode, the detection apparatus 100 determines “moving state”, and when the detection unit 31 has a person in the stay detection mode. When detected, it is determined as “stay state”. And the detection apparatus 100 outputs a determination result in the form which can distinguish whether the person in the detection area | region 21 is moving (it is in a moving state), or is staying (it is in a staying state). By doing so, the load 6 can be controlled according to the movement and stay of a person.

  In addition, when the determination results of both “movement state” and “stay state” are obtained in duplicate, the detection apparatus 100 preferentially adopts the determination result of “stay state”, for example. When the detection unit 31 detects that there is no person in either the movement detection mode or the stay detection mode, the detection device 100 determines that the “absence state”.

  By the way, when an abrupt light change occurs due to switching of the operation mode to be controlled or disturbance in the detection region of the analysis image, the luminance change amount in the predetermined period T1 (first predetermined period) is the first. It may be compared with a threshold value S1. If there is a region where the amount of change in luminance during the predetermined period T1 is equal to or greater than the first threshold value S1, the deriving unit 33 uses this region (region where the amount of change in luminance is equal to or greater than the first threshold value S1) as the luminance change region. Deriving the size of the region. Then, when the size of the luminance change region is equal to or greater than the second threshold S2, the detection unit 31 determines that there is a light change. Therefore, the control unit 3 can detect the occurrence of a rapid light change in the detection region of the analysis image. Note that the rapid light change mentioned here includes a change in brightness that does not lead to overexposure of the analysis image, and a change in light when changing from a bright state (high luminance) to a dark state (low luminance). .

  Here, when the detection unit 31 determines that there is an identification region in the detection region and that there is a light change during operation in the movement detection mode, the detection unit 31 detects the movement of the detection mode at least for a predetermined period T2 (second predetermined period). The mode may be switched to the stay detection mode. That is, when the detection unit 31 determines that there is a light change during operation in the movement detection mode and the detection mode is switched from the movement detection mode to the stay detection mode, the movement detection mode is performed regardless of the position change amount of the identification region. Switches to the stay detection mode. When it is determined that there is an identification area in the detection area and there is a light change during operation in the movement detection mode, the detection unit 31 determines that the identification area is a stay area regardless of the position change amount of the identification area at least during the predetermined period T2. To do. And the detection part 31 detects the presence or absence of a person in the stay detection mode about the said stay area. Accordingly, when the detection unit 31 determines that there is a light change during operation in the movement detection mode, the detection mode of the detection unit 31 is switched from the movement detection mode to the stay detection mode.

(Embodiment 2)
Hereinafter, the detection device 200 and the detection system 220 according to the second embodiment of the present invention will be described with reference to FIGS. The detection device 200 and the detection system 220 are mainly different from the detection device 100 and the detection system 120 in that the control unit 3 further includes an image processing unit 34 that performs image processing on the analysis image. In addition, the detection device 100 and the detection system 220 also determine that there is no person when the detection unit 31 detects that there is no person continuously for a predetermined period Q in the stay detection mode. And the detection system 120 is different. In addition, about the structure similar to Embodiment 1, the same code | symbol is attached | subjected and description is abbreviate | omitted suitably.

  FIG. 7 is a block diagram of the detection device 200 and the detection system 220 according to the second embodiment. As shown in FIG. 7, the control unit 3 of the detection apparatus 200 further includes an image processing unit 34. The storage unit 5 further stores a period threshold value “Q”.

  The image processing unit 34 performs Sobel filter image processing on the analysis image captured by the imaging unit 2.

  The image processing unit 34 performs Sobel filter image processing on the analysis image, and outputs the output image after the image processing to the detection unit 31, the creation unit 32, and the derivation unit 33.

  FIG. 8 is a conceptual diagram illustrating an example of an analysis image captured from the imaging unit 2 and an output image after the Sobel filter image processing is applied to the analysis image.

  As shown in FIG. 8, by applying the Sobel filter image processing to the analysis image, it is possible to detect the contour (edge) of the person to be detected with higher accuracy.

  FIG. 9 is a flowchart in which the detection apparatus 200 detects the presence or absence of a person in the stay detection mode.

  As shown in FIG. 9, in step S91, the detection unit 31 acquires an output image obtained by applying the Sobel filter image processing to the analysis image by the image processing unit 34, and the human motion amount in the stay region is obtained. Detect presence or absence.

  In step S <b> 92, the detection unit 31 compares the amount of human motion in the stay area derived by the deriving unit 33 with the detection threshold “T”. If the amount of movement of the person is equal to or greater than the detection threshold “T” (YES in step S92), the detection unit 31 compares the amount of movement of the person with the release threshold “U” (step S95). If the amount of movement of the person is less than the detection threshold “T” (NO in step S92), the detection unit 31 starts time counting and whether or not the predetermined period Q has elapsed, that is, the elapsed time is the period threshold “Q”. Is determined (step S93).

  In step S93, when the human motion amount is continuously less than the detection threshold “T” in step S93 (YES in step S93), the detection unit 31 determines that there is no person (step S94). Otherwise, the detection unit 31 resets the time count and continuously detects the presence or absence of a person in the stay detection mode (NO in step S93).

  In step S95, if the amount of movement of the person is less than the release threshold “U” (YES in step S95), the detection unit 31 determines that there is a person (step S96) and continues to the person in the stay detection mode. To track. If the movement amount of the person is U or more (NO in step S95), the detection unit 31 switches the detection mode from the stay detection mode to the movement detection mode (step S97).

  That is, in the detection device 200 and the detection system 220, the detection unit 31 detects the presence or absence of a person using an output image obtained by applying filter image processing to an analysis image at least in the stay detection mode. In this configuration, it is preferable to use differential filter image processing for the filter image processing. The detection device 200 and the detection system 220 can detect the presence or absence of a person with higher accuracy by applying the Sobel filter image processing to the analysis image by the image processing unit 34.

  The detection unit 31 is described as detecting the presence / absence of a person based on an output image to which filter image processing is applied when detecting the presence / absence of a person in the stay detection mode. However, the present invention is not limited thereto. . The detection unit 31 may also use an output image to which filter image processing is applied when detecting the presence or absence of a person in the movement detection mode.

  Further, the filter image processing is not limited to Sobel filter image processing. The filter image processing uses, for example, a primary differential filter (for example, edge detection, line detection, etc.), a secondary differential filter (for example, Laplacian filter, etc.), image enhancement (for example, low frequency enhancement, high frequency enhancement, etc.), etc. May be.

  Further, the image processing unit 34 is not limited to that provided in the control unit 3. For example, it may be provided separately from the control unit 3.

  In the stay detection mode, the detection unit 31 preferably determines that there is no person when a state where the amount of movement of the person in the stay area is less than the detection threshold “T” continues for a predetermined period Q. That is, in the stay detection mode, the presence or absence of a person in the stay detection mode is more accurately determined by determining that there is no person when the detection unit 31 detects that there is no person continuously for a predetermined period Q. Highly detectable.

  Here, when it is determined that there is no person in the stay area in the stay detection mode, the detection unit 31 preferably switches from the stay detection mode to the movement detection mode. In this case, in the stay detection mode, when a state where the amount of movement of the person in the stay area is less than the detection threshold “T” continues for a predetermined period Q, the detection unit 31 determines that there is no person and stays Switch from detection mode to movement detection mode. The switching from the stay detection mode to the movement detection mode is as described in the first embodiment, and the detection unit 31 releases the stay area and switches the detection mode in the stay area from the stay detection mode to the movement detection mode. .

  Thus, if the state where the amount of human movement is less than the detection threshold “T” continues for a predetermined period Q, the movement amount of the person in the stay area is changed from the stay detection mode to the movement detection mode. Even if it is not equal to or greater than U ″, the detection unit 31 can end the stay detection mode. In short, even if the amount of movement of the person in the stay area does not exceed the release threshold “U”, the stay area is released when a certain period of time has passed without detecting the presence of a person in the stay area. . As a result, since the stay area determined to have no person is released and switched to the movement detection mode, the detection unit 31 detects the presence or absence of a person in the stay detection mode indefinitely for the stay area in which no person exists. Can be avoided.

  It should be noted that the control unit 3 has an image processing unit 34 that performs image processing on the analysis image, and the presence of a person when the detection unit 31 detects that there is no person continuously over a predetermined period Q in the stay detection mode. The point of judging that there is no is not limited to the one that is adopted together. That is, the control unit 3 has an image processing unit 34 that performs image processing on the analysis image, and the presence of a person when the detection unit 31 detects that there is no person continuously over a predetermined period Q in the stay detection mode. Only one of the points determined to be absent may be adopted.

(Embodiment 3)
Hereinafter, the detection apparatus 300 and the detection system 320 according to Embodiment 3 of the present invention will be described. The detection device 300 and the detection system 320 differ from the detection device 200 and the detection system 220 in the manner of deriving the amount of human movement. In the detection device 300 and the detection system 320, the number of pixels in which the luminance change amount of the pixel in the stay region is greater than or equal to the derivation threshold “K” in the stay detection mode and the total number of pixels in the stay region The ratio is derived as the amount of human movement. The deriving unit 33 also includes the number of pixels whose luminance change amount in the detection area 21 of the analysis image is greater than or equal to the derivation threshold “K” and the number of pixels in the detection area 21 of the analysis image in the movement detection mode. Is derived as the amount of human movement. In addition, about the structure similar to Embodiment 2, the same code | symbol is attached | subjected and description is abbreviate | omitted suitably.

  In the stay detection mode, the deriving unit 33 of the detection device 300 and the detection system 320 calculates the number of pixels in which the luminance change amount of the pixel in the stay region is equal to or greater than the derivation threshold “K” and the total number of pixels in the stay region. The ratio is derived as the amount of human movement. Furthermore, in the movement detection mode, the deriving unit 33 determines the number of pixels whose luminance change amount in the analysis image detection area 21 is equal to or greater than the derivation threshold “K”, and all the pixels in the analysis image detection area 21. The ratio to the number is derived as the amount of human movement.

  The detection unit 31 of the detection device 300 and the detection system 320 compares the amount of human movement derived by the deriving unit 33 with the threshold value S of movement detection stored in the storage unit 5 in the movement detection mode. Determine if it exists.

  The detection unit 31 of the detection device 300 and the detection system 320 compares the amount of movement of the person derived by the deriving unit 33 with the detection threshold “T” in the stay detection mode, and determines the presence or absence of the person.

  The relationship between the movement detection threshold value S and the detection threshold value “T” is not particularly limited. For example, when the detection of a person in the stay detection mode is strictly set, the detection threshold “T” may be set to a value smaller than the movement detection threshold S.

  In the stay detection mode, the deriving unit 33 of the detection device 300 and the detection system 320 calculates the number of pixels in which the luminance change amount of the pixel in the stay region is equal to or greater than the derivation threshold “K” and the total number of pixels in the stay region. The amount of human motion is derived based on the ratio. Further, in the movement detection mode, the deriving unit 33 is based on the ratio of the number of pixels in which the luminance change amount of the pixels in the detection area 21 is equal to or greater than the derivation threshold “K” and the total number of pixels in the detection area 21. To derive the amount of human movement. Thereby, even if the number of pixels in the identification area (stay area) and the detection area 21 changes as the detection area 21 of the analysis image is enlarged or reduced, the detection unit 31 moves with the same detection threshold “T”. The presence or absence of a person can be detected using the detection threshold S.

  The deriving unit 33 only needs to apply the method of deriving the amount of human motion described above in the stay detection mode, and it is not essential to derive the amount of human motion in the movement detection mode. That is, the deriving unit 33 may derive the amount of human movement in the stay area by deriving the amount of human movement by the method described above, and it is essential to derive the amount of human movement in the detection area 21. Not.

(Embodiment 4)
Hereinafter, the detection apparatus 400 and the detection system 420 according to Embodiment 4 of the present invention will be described. The detection device 400 and the detection system 420 differ from the detection device 300 and the detection system 320 in the manner of deriving the amount of human movement. In the detection device 400 and the detection system 420, the deriving unit 33 derives the size of the pixel block area in which the amount of change in the luminance of the pixel in the stay region is equal to or greater than the derivation threshold “K” as the amount of human movement in the stay detection mode. To do. Further, the deriving unit 33 determines the number of pixels whose luminance change amount in the analysis image detection area 21 is equal to or larger than the derivation threshold “K” in the movement detection mode and the area of the pixel block in the analysis image detection area 21. Is derived as the amount of human movement. In addition, about the structure similar to Embodiment 2, the same code | symbol is attached | subjected and description is abbreviate | omitted suitably.

  The deriving unit 33 of the detection device 400 and the detection system 420 sets pixels adjacent to each other vertically, horizontally, or diagonally as one lump. That is, the deriving unit 33 sets a plurality of pixels that are continuous in one of the vertical, horizontal, and diagonal paths as one lump in the analysis image.

  In the stay detection mode, the deriving unit 33 of the detection device 400 and the detection system 420 uses, as the amount of human movement, the size of an area of a pixel block whose luminance change amount in the stay region is equal to or greater than the derivation threshold “K”. To derive. Further, in the movement detection mode, the deriving unit 33 derives the size of the pixel block area in which the amount of change in pixel luminance in the detection region 21 of the analysis image is equal to or greater than the deriving threshold “K” as the amount of human motion. To do.

  The deriving unit 33 of the detection device 400 and the detection system 420 uses the size of the block of pixels as a reference instead of the number of pixels for pixels whose luminance change amount is equal to or greater than the deriving threshold “K”. The amount of human movement is derived. Thereby, the deriving unit 33 can make it difficult to be influenced by pixels in which the luminance changes due to disturbance in a discrete manner. Therefore, the detection device 400 and the detection system 420 can detect the presence or absence of a person with higher accuracy.

  The deriving unit 33 only needs to apply the method of deriving the amount of human motion described above in the stay detection mode, and it is not essential to derive the amount of human motion in the movement detection mode. That is, the deriving unit 33 may derive the amount of human movement in the stay area by deriving the amount of human movement by the method described above, and it is essential to derive the amount of human movement in the detection area 21. Not.

(Embodiment 5)
Hereinafter, the detection apparatus 500 and the detection system 520 according to Embodiment 5 of the present invention will be described. The detection device 500 and the detection system 520 are different from the detection device 200 and the detection system 220 in the way of deriving the amount of human movement. In the detection device 500 and the detection system 520, the deriving unit 33 derives the standard deviation of the amount of change in luminance of the pixel in the stay region as the amount of human movement in the stay detection mode. The deriving unit 33 derives the standard deviation of the amount of change in luminance of the pixels in the detection area 21 of the analysis image in the movement detection mode as the amount of human movement. In addition, about the structure similar to Embodiment 2, the same code | symbol is attached | subjected and description is abbreviate | omitted suitably.

  In the stay detection mode, the deriving unit 33 of the detection device 500 and the detection system 520 derives the amount of movement of the person based on the standard deviation of the amount of change in pixel luminance within the stay region. Further, in the movement detection mode, the deriving unit 33 derives the amount of human motion based on the standard deviation of the amount of change in luminance of the pixels in the detection area 21 of the analysis image. Thereby, the detection part 31 can detect the presence or absence of a person by the difference between frames, reducing the variation in the variation | change_quantity of the brightness | luminance of a pixel.

  Therefore, the detection device 500 and the detection system 520 can detect the presence or absence of a person with higher accuracy.

  In order to reduce the variation in the amount of change in luminance of the pixels, in addition to the standard deviation described in this embodiment, for example, the slope (variance) of the amount of change during a predetermined period with the standard deviation is used as the amount of human movement. It may be derived. That is, the deriving unit 33 may derive the amount of human movement based on the slope of the change amount of the standard deviation of the change amount of the luminance of the pixel in the stay region.

  In short, in the detection device 500 and the detection system 520, in the stay detection mode, the deriving unit 33 determines the magnitude of variation from the representative value (average value or median value) of the amount of change in the luminance of the pixels in the stay region. What is necessary is just to derive | lead-out as a motion amount. Further, in the movement detection mode, the deriving unit 33 derives the magnitude of the variation from the representative value (average value or median value) of the luminance change amount of the pixel in the detection area 21 of the analysis image as the amount of human motion. do it.

  Further, the deriving unit 33 may be configured to derive the amount of human movement by a method other than the above. For example, the deriving unit 33 obtains an average value (weighted average value) by weighting a certain pixel with the previous luminance (average value) and the current luminance, and further calculates the weighted average value and the previous change amount. Based on this, the cumulative change amount is obtained. The deriving unit 33 can derive the accumulated change amount of the luminance thus obtained as the amount of human movement.

  The deriving unit 33 only needs to apply the method of deriving the amount of human motion described above in the stay detection mode, and it is not essential to derive the amount of human motion in the movement detection mode. That is, the deriving unit 33 may derive the amount of human movement in the stay area by deriving the amount of human movement by the method described above, and it is essential to derive the amount of human movement in the detection area 21. Not.

(Embodiment 6)
Hereinafter, a detection apparatus 600 and a detection system 620 according to Embodiment 6 of the present invention will be described with reference to FIGS. The detection device 600 and the detection system 620 are mainly different from the detection device 100 and the detection system 120 in that the detection device 600 and the detection system 120 further include a changing unit 35 that changes the range that becomes the detection region 21 in the analysis image. In addition, about the structure similar to Embodiment 1, the same code | symbol is attached | subjected and description is abbreviate | omitted suitably.

  When the detection mode of the detection unit 31 is switched from the movement detection mode to the stay detection mode, the change unit 35 changes the range to be the detection region 21 in the analysis image. The changing unit 35 changes the range to be the detection region 21 in the analysis image by switching the range to be the detection region 21 in the analysis image between two types of a normal area and a staying area. That is, when the detection unit 31 switches from the movement detection mode to the stay detection mode, the change unit 35 changes the range that becomes the detection region in the analysis image from the normal area to the stay time area. Therefore, if the detection mode of the detection unit 31 is the stay detection mode, the range that becomes the detection region 21 in the analysis image is a normal area. On the other hand, if the detection mode of the detection part 31 is a movement detection mode, the range used as the detection area 21 of an analysis image will be a staying area.

  In the sixth embodiment, as described in the second embodiment, in the stay detection mode, the detection unit 31 continues the state where the amount of human movement in the stay area is less than the detection threshold “T” for a predetermined period Q. Sometimes it is determined that there is no person. Furthermore, the detection part 31 switches from stay detection mode to movement detection mode, when it determines with there being no person in a stay area in stay detection mode. That is, in the stay detection mode, when the state where the amount of movement of the person in the stay area is less than the detection threshold “T” continues for a predetermined period Q, the detection unit 31 determines that there is no person and the stay detection. Switch from mode to movement detection mode.

  Hereinafter, an example of the operation of the detection apparatus 600 will be described with reference to the flowchart shown in FIG. FIG. 11 is a flowchart illustrating an operation when the detection device 600 switches the detection mode based on the position change amount of the identification region 22.

  In step S111, the detection unit 31 starts operating in the movement detection mode. At this time, the detection unit 31 applies the normal area as a range to be the detection area 21 in the analysis image (step S112), and detects the presence or absence of a person in the detection area 21 in the movement detection mode. First, the detection device 600 determines that there is no person in the detection area 21, that is, the “absence state” (step S113).

  Thereafter, while the detection unit 31 detects that there is no person in the detection area 21, that is, no person has entered the detection area 21 (NO in step S114), the detection apparatus 600 “ The judgment result of “absence state” is maintained. When the detection unit 31 detects that there is a person in the detection area 21, that is, that a person has entered the detection area 21 (YES in step S114), the detection apparatus 600 determines that it is “moving state” ( Step S115). When the determination is made that “moving state”, the detection unit 31 compares the position change amount of the identification area 22 derived by the deriving unit 33 with the stagnation threshold “I” (step S116).

  If the position change amount of the identification area 22 is less than I (YES in step S116), the detection unit 31 proceeds to step S117, and switches the detection mode from the movement detection mode to the stay detection mode. If the amount of change in the position of the identification region 22 is equal to or greater than the stagnation threshold “I” (NO in step S116), the detection unit 31 returns to step S114 and continues to detect the presence or absence of a person in the movement detection mode.

  When the detection mode of the detection unit 31 is switched to the stay detection mode in step S117, the change unit 35 changes the range to be the detection area 21 in the analysis image from the normal area to the stay time area. Therefore, when the detection mode is switched to the stay detection mode, the detection unit 31 applies the stay area as a range to be the detection area 21 in the analysis image (step S118), and the person in the identification area 22 in the stay detection mode. Is detected (step S119).

  In step S119, the detection unit 31 compares the amount of human movement in the stagnation area derived by the deriving unit 33 with the detection threshold “T”. When the amount of movement of the person is equal to or greater than the detection threshold “T” and the detection unit 31 detects that there is a person in the stay area (YES in step S119), the detection device 600 determines that the “stay state” (step). S120). If it is determined that the “stay state”, the process returns to step S119, and the detection unit 31 continues to detect the presence or absence of a person in the stay detection mode.

  If the movement amount of the person is less than the detection threshold “T” (NO in step S119), the detection unit 31 determines whether the predetermined period Q has elapsed since the movement amount of the person is less than the detection threshold “T”. Is determined (step S121). If the predetermined period Q has not elapsed (NO in step S121), the process returns to step S119, and the detection unit 31 continues to detect the presence or absence of a person in the stay detection mode. If the predetermined period Q elapses with the human motion amount being less than the detection threshold “T” (YES in step S121), the process proceeds to step S122, and the detection unit 31 switches the detection mode from the stay detection mode to the movement detection mode. .

  When the detection mode of the detection unit 31 is switched to the movement detection mode in step S122, the changing unit 35 changes the range to be the detection region 21 in the analysis image from the staying area to the normal area. Therefore, when the detection mode is switched to the movement detection mode, the detection unit 31 applies the normal area as a range to be the detection region 21 in the analysis image (step S123), returns to step S114, and the detection unit 31 The presence / absence of a person is detected in the movement detection mode.

  Hereinafter, an operation in which the changing unit 35 changes the range to be the detection region 21 in the analysis image will be described with reference to FIGS. 12A, 12B, and 12C.

  12A, 12B, and 12C show analysis images of three consecutive frames when a person 70 enters the detection area 21 from outside the detection area 21 and stops (stagnates) on the spot. FIG. 12A, 12B, and 12C assume a case where the person 70 in the office is imaged from above by the imaging unit 2 attached to the ceiling 80 (see FIG. 13) of the office.

  In the state of FIG. 12A, the person 70 exists in a position outside the detection area 21 in the analysis image (that is, in the field of view of the imaging unit 2). Therefore, the detection unit 31 operates in the movement detection mode to detect the presence of the person 70 in the analysis image, and the creation unit 32 creates the identification area 22 so as to surround the person 70. Here, since the person 70 does not exist in the detection area 21, the identification area 22 is created outside the detection area 21 in the analysis image. At this time, since the detection unit 31 operates in the movement detection mode, the normal area 211 in the analysis image is used as the detection region 21. That is, the detection area 21 in FIG. 12A is the normal area 211.

  In the state of FIG. 12B, the person 70 has moved into the detection area 21 in the analysis image. At this time, the identification area 22 is located at the end of the detection area 21, and a part of the identification area 22 overlaps the detection area 21. At this time, since the detection unit 31 operates in the movement detection mode, the normal area 211 in the analysis image is used as the detection region 21. That is, the detection area 21 in FIG. 12B is the normal area 211.

  In the state of FIG. 12C, the person 70 exists in the same position as the state of FIG. 12B. Therefore, the detection unit 31 fixes the position of the identification area 22 to make the identification area 22 a stay area 23 and switches the detection mode from the movement detection mode to the stay detection mode. Here, the detection unit 31 operates in the stay detection mode and detects the presence of the person 70 in the stay area 23. At this time, in response to the detection mode of the detection unit 31 being switched to the stay detection mode, the changing unit 35 changes the range to be the detection region 21 in the analysis image from the normal area 211 to the stay-time area 212. Therefore, the stay time area 212 in the analysis image is used as the detection area 21. That is, the detection area 21 in FIG.

  Here, the stay area 212 is set wider than the normal area 211. That is, when the detection mode of the detection unit 31 is switched to the stay detection mode, the change unit 35 changes the range to be the detection region 21 in the analysis image from the normal area 211 to the stay area 212, thereby detecting the detection region 21. Will be expanded. The stay area 212 includes the entire normal area 211 and is slightly larger than the normal area 211 so that the entire stay area 23 set in the analysis image in the state of FIG. 12C is included, for example. It is preferable that

  Hereinafter, the relationship between the normal area 211 and the stay area 212 will be described with reference to FIGS. 13 and 14. 13 and 14 assume the case where the person 70 in the office is imaged from above by the imaging unit 2 attached to the ceiling 80 of the office, as in FIG. 12C. In FIG. 13 and FIG. 14, the normal area 211, the staying area 212, and the staying area 23 are shown to exist in the real space. This shows the correspondence between the real space and the analysis image. It is an expression to make it easier, and these do not exist in real space. That is, the normal area 211, the staying area 212, and the staying area 23 are only virtually shown in FIGS. 13 and 14 and are actually set in the analysis image.

  In the example of FIG. 13, the person 70 is sitting at a place corresponding to the end of the normal area 211 (stagnation). In this state, the stay area 23 is set so as to surround the person 70. However, since only the foot of the person 70 is included in the range of the normal area 211, the portion corresponding to the upper half of the person 70 in the stay area 23 is projected from the normal area 211.

  On the other hand, the stay area 212 that is slightly larger than the normal area 211 includes substantially the entire stay area 23. That is, the staying area 212 includes substantially the entire body including the upper body of the person 70. Therefore, the detection unit 31 enlarges the detection area 21 and changes the detection area 21 from the normal area 211 to the staying area 212, so that even the person 70 existing at the end of the normal area 211 can be substantially entirely It can be accommodated in the stay area 23. Therefore, the detection unit 31 operates in the stay detection mode to detect the presence of the person 70 in the stay area 23 by detecting the fine movement of the person 70 even at the end of the normal area 211. Presence or absence can be detected reliably.

  Specifically, the staying area 212 is set according to the height of the top of the person 70 who is sitting so that the whole body of the person 70 sitting at the end of the normal area 211 can be accommodated. It is preferable. For example, if the height of the top of the person 70 in a sitting state (height from the floor) is 1.5 m and the normal area 211 is a square area with a side of 5 m on the floor, the area 212 during stay is Is preferably a square area with a side of about 10 m on the floor.

  In the example of FIG. 14, in addition to the person 701 sitting at a place corresponding to the end of the normal area 211 as in FIG. 13, the person 702 moving outside the normal area 211 and within the area 212 during the stay. Is assumed. In this case, the stay area 23 is set so as to surround the person 701. Therefore, the detection unit 31 enlarges the detection area 21 and changes the detection area 21 from the normal area 211 to the staying area 212, so that the person 701 existing at the end of the normal area 211 can also move substantially the whole body. It can be accommodated in the stay area 23. Therefore, the detection unit 31 operates in the stay detection mode, so that the presence of the person 701 in the stay area 23 can be detected by detecting the fine movement of the person 701 even in the person 701 existing at the end of the normal area 211. Presence or absence can be detected reliably.

  On the other hand, if the detection unit 31 always uses the stay area 212 as the detection area 21, the person 702 moves outside the normal area 211 as in the example of FIG. 70 (702) is present. Therefore, when the detection unit 31 is operating in the movement detection mode, the normal area 211 is set as the detection area 21, so that the person 702 moving outside the normal area 211 exists in the detection area 21. It is configured not to detect that there is. Therefore, the detection unit 31 operates in the movement detection mode, so that the presence of the person 70 in the detection area 21 exists only when the person 70 exists in the normal area 211 that is originally set as the detection area 21. It can be detected that there is.

  Moreover, it is preferable that the change part 35 makes the area 212 into which the normal area 211 was expanded according to the position of the stay area 23 in the detection area 21 as the area 212 at the time of stay. That is, when the detection unit 31 switches from the movement detection mode to the stay detection mode and the range to be the detection region 21 in the analysis image is changed from the normal area 211 to the stay area 212, the changing unit 35 changes the stay region 23. You may employ | adopt the area 212 at the time of stay according to the position. In other words, the stay area 212 is not limited to a predetermined area, and may be changed according to the position of the stay area 23 in the detection area 21.

  As a specific example, when a part of the stay area 23 protrudes from the normal area 211 when there is a person 70 at the end of the normal area 211, only the part where the stay area 23 protrudes from the analysis image is detected. The changing unit 35 sets the stay time area 212 so that the area 21 is enlarged. That is, in the analysis image, an area obtained by adding the protruding portion of the stay area 23 to the normal area 211 becomes the stay time area 212. As described above, the detection region 21 is enlarged only for the portion where the stay region 23 exists, so that the detection region 21 is not enlarged to a portion other than the stay region 23 in the analysis image. Therefore, even when the detection unit 31 is operating in the stay detection mode, the part other than the stay area 23 where the person 70 is staying in the analysis image is detected only when the person 70 exists in the normal area 211. It can be detected that there is a person 70 in the area 21.

  In addition, when the change part 35 makes the area 212 where the normal area 211 expanded according to the position of the stay area 23 in the detection area 21 into the area 212 at the time of stay, the stay area 23 of the analysis image protruded The detection area 21 may be enlarged in addition to the portion. For example, in the analysis image, when a part of the stay area 23 protrudes rightward from the normal area 211 when a person 70 exists at the right end of the normal area 211, the detection area 21 is enlarged only to the right. In this way, the changing unit 35 may set the stay area 212. That is, in this case, the changing unit 35 moves only the right side of the normal area 211 to the right, and expands the normal area 211 only to the right to the detection area 21 in the analysis image 21. Change the range.

  As described above, the enlargement of the detection area 21 is limited according to the position of the stay area 23 in the detection area 21, so that the detection area 21 can be changed (enlarged) for a range unrelated to the stay area 23. The detection unit 31 can detect the presence or absence of a person without being affected by (1).

  The detection device 600 further includes a limiting unit 36 that sets an upper limit value of the amount of change in the range that becomes the detection area 21 due to the change from the normal area 211 to the staying area 212. That is, as described above, when the changing unit 35 expands the detection area 21 by changing the range to be the detection area 21 in the analysis image from the normal area 211 to the on-stay area 212, the limiting unit 36 An upper limit value of the enlargement width of the detection area 21 is set.

  FIG. 15 is a conceptual diagram showing an example of the relationship between the detection area 21 (normal area 211) in the analysis image and the upper limit frame 24 that defines the upper limit value by the restriction unit 36.

  When the detection unit 31 is switched from the movement detection mode to the stay detection mode, the change unit 35 changes the range to be the detection region 21 in the analysis image from the normal area 211 to the stay time area 212. At this time, the detection area 21 due to the change from the normal area 211 to the stay time area 212 so that the stay area 212 that becomes the detection area 21 after the change falls within the range of the upper limit frame 24 set by the restriction unit 36. The upper limit value of the amount of change in the range is set.

  As a result, the detection region 21 is not enlarged beyond the upper limit set by the restriction unit 36, and the detection unit 31 can avoid the detection region 21 to be detected by the person 70 from becoming too wide.

  Moreover, it is preferable that the upper limit value set by the limiter 36 is not a fixed value but a variable value. In this case, in the example of FIG. 15, the upper limit frame 24 that defines the upper limit value is variable. The change of the upper limit value may be performed by the restriction unit 36 according to an operation input to the detection device 600, or may be automatically performed by the restriction unit 36.

  Further, the upper limit value set by the restriction unit 36 may be set only for a change in the detection area 21 in a specific direction. For example, the restricting unit 36 sets the range to be the detection area 21 by the change from the normal area 211 to the staying area 212 so that the change amount (enlargement amount) is limited only for the rightward expansion of the detection area 21. The upper limit value of the change amount may be set only for a specific direction. Further, as described above, when the upper limit value set by the limiter 36 is a variable value, the upper limit value may be changed for each direction of change of the detection region 21.

  Further, when the change of the detection region 21 is restricted by the restriction unit 36, the detection unit 31 adjusts the sensitivity of detection of the presence / absence of the person 70 by switching the detection threshold “I”, for example. It may be configured. For example, the index indicating the size of the stay area 23 in the analysis image is “x1”, the index indicating the degree to which the change of the detection area 21 is restricted by the restriction unit 36 is “x2”, and the sensitivity reference value is “y1”. In this case, the sensitivity y2 is expressed by y2 = (x2 / x1) × y1.

  Moreover, the change part 35 should just be the structure which changes the range used as the detection area 21 of an analysis image from the normal area 211 to the area 212 at the time of a stay, when the detection part 31 switches from a movement detection mode to a stay detection mode. The configuration is not limited to enlarging the detection area 21. For example, the change unit 35 may be configured to reduce the detection area 21 when the detection unit 31 is switched from the movement detection mode to the stay detection mode. In this case, the stay area 212 is smaller than the normal area 211. In this case, the stay area 212 is set, for example, at a position centered on the stay area 23. Further, the change unit 35 may be configured to shift the position of the detection region 21 in the analysis image when the detection unit 31 is switched from the movement detection mode to the stay detection mode. In this case, the normal area 211 and the staying area 212 have the same size and differ only in the position in the analysis image.

  Note that the detection device 600 and the detection system 620 only need to include the changing unit 35 that changes the range to be the detection region 21 in the analysis image, and the provision of the limiting unit 36 is not an essential configuration.

(Program and detection method)
The control unit 3 of the detection apparatus as described above can be realized by causing a computer to execute an appropriate program. That is, this program is a program for causing a computer to function as a detection device, and is provided, for example, by being recorded on a recording medium or via an electric communication line. The detection device here includes a detection unit 31, a creation unit 32, and a derivation unit 33. The detection unit 31 detects the presence or absence of a person in the detection area of the analysis image. When the detection unit 31 detects the presence of a person in the detection area, the generation unit 32 generates an identification area surrounding the detected person in the analysis image. The deriving unit 33 derives the amount of movement of the person in the stay area including the identification area stagnated in the detection area and the position change amount of the identification area in the detection area. The detection unit 31 has a movement detection mode that detects the presence or absence of a person in the detection area, and a stay detection mode that detects the presence or absence of a person based on the amount of movement of the person in the stay area. ing. In the movement detection mode, when the position change amount of the identification area becomes less than the stagnation threshold, the detection unit 31 fixes the position of the identification area in the detection area, and stays in the stay area using the identification area as a stay area. The presence / absence of a person is detected in the detection mode.

  Further, by using this type of detection device, the following detection method can be realized. That is, the detection method referred to here includes a detection stage, a creation stage, and a derivation stage. In the detection stage, the presence or absence of a person in the detection area of the analysis image is detected. In the creation stage, when it is detected in the detection stage that there is a person in the detection area, an identification area surrounding the detected person is created in the analysis image. In the derivation stage, the amount of movement of the person in the stay area composed of the identification area stagnated in the detection area and the position change amount of the identification area in the detection area are derived. The detection stage includes a movement detection mode for detecting the presence or absence of a person in the detection area, and a stay detection mode for detecting the presence or absence of a person based on the amount of movement of the person in the stay area. Yes. In the detection stage, when the position change amount of the identification area becomes less than the stagnation threshold value in the movement detection mode, the position of the identification area in the detection area is fixed, and the stay area is detected for the stay area. Detect the presence or absence of people in the mode.

(Conclusion)
As described above, the embodiments have been described as examples of the technology in the present invention. For this purpose, the accompanying drawings and detailed description are provided.

  Accordingly, among the components described in the accompanying drawings and the detailed description, not only the components essential for solving the problem, but also the components not essential for solving the problem in order to illustrate the above technique. May also be included. Therefore, it should not be immediately recognized that these non-essential components are essential as those non-essential components are described in the accompanying drawings and detailed description.

  Moreover, since the above-mentioned embodiment is for demonstrating the technique in this invention, a various change, replacement, addition, abbreviation, etc. can be performed in a claim or its equivalent range.

100, 200, 300, 400, 500, 600 Detection device 120, 220, 320, 420, 520, 620 Detection system 2 Imaging unit 21 Detection region 22, 22a, 22b Identification region 23, 23a, 23b Stay region 3 Control unit 31 Detection unit 32 Creation unit 33 Derivation unit 34 Image processing unit 4 Transmission unit 5 Storage unit 6 Load

Claims (5)

A detection unit that detects the presence or absence of a person in the detection region of the analysis image captured by the imaging unit;
The detector is
A movement detection mode for detecting the presence or absence of a person in the detection area;
And a stay detection mode for detecting the presence or absence of a person based on the amount of movement of the person in the stay area where the person is estimated to be stagnant in the detection area. The detector is
In the stay detection mode,
If the amount of movement of the person is less than the detection threshold, it is determined that there is no person,
The detection device according to claim 1, wherein it is determined that there is a person if the amount of movement of the person is equal to or greater than a detection threshold. The detector is
In the stay detection mode,
The detection device according to claim 2, wherein the movement detection mode is switched from the stay detection mode if the movement amount of the person is equal to or greater than a release threshold that is greater than the detection threshold. A detection system comprising: the detection device according to claim 1; and a load controlled based on a detection result of the detection unit. Computer
With a detection unit that detects the presence or absence of people in the detection area of the analysis image,
The detector is
A movement detection mode for detecting the presence or absence of a person in the detection area;
A program for functioning as a detection device comprising: a stay detection mode that detects the presence or absence of a person based on the amount of movement of a person in a stay area where a person is estimated to be stagnant in the detection area.