Object detection device, security system, object detection method, and program

Abstract

To provide an object detection device, a security system, an object detection method, and a program which allow for object detection in a non-detection area being the outside of a regular detection range of a ranging unit.SOLUTION: An object detection device 10 includes: a ranging unit 12 which obtains a distance to an object by a triangulation method on the basis of both a parallax based on shifts of pixels between images of the object captured by two cameras in a detection area of which the images can be captured by the respective cameras and baseline lengths of the two cameras; and an object detector 131 which three-dimensionally recognizes a position and shape of the object on the basis of the distance obtained by the ranging unit 12. If the ranging unit 12 cannot obtain the parallax though an image of the object captured by at least one camera, the object detector 131 estimates the existence of an object in a non-detection area which is located between the ranging unit and the detection area and in which any object cannot be detected in principle.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to an object detection device, a security system, an object detection method, and a program.

Background Art

[0002] Conventionally, an object detection device that detects the intrusion of an object such as a person into a detection area has been disclosed. Among such object detection devices, there are those that use a distance measurement sensor such as a stereo camera that detects the intrusion of an object such as a person into a detection area based on a parallax image obtained using two cameras.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the case of a distance measurement sensor that measures and detects the distance to an object by triangulation based on the parallax and the baseline length based on the pixel shift between two images obtained by imaging a predetermined detection area, such as a stereo camera, an object in the immediate vicinity of the distance measurement sensor is imaged by only one camera, or even if it is imaged by two cameras, the captured images cannot be recognized as images of the same object. Therefore, parallax cannot be calculated from the pixel shift, and the immediate vicinity becomes a non-detection area where objects cannot be detected in principle.

[0005] The present invention has been made in view of the above, and an object thereof is to enable object detection in a non-detection area that is originally outside the detection range of a distance measurement sensor.

Means for Solving the Problems

[0006] To solve the above-mentioned problems and achieve the objective, the present invention comprises a distance measuring unit that obtains the distance to an object by triangulation based on the parallax based on the pixel displacement of the image of the object captured by each of the two imaging means in a detection area that can be captured by the two imaging means, and the baseline lengths of the two imaging means; and an object detection unit that grasps the position and shape of the object in three dimensions based on the distance obtained by the distance measuring unit. The object detection unit infers that if an image of the object has been captured by at least one of the imaging means by the distance measuring unit but the parallax has not been obtained, the object exists in a non-detection area located between the distance measuring unit and the detection area where, in principle, the object cannot be detected. Furthermore, if the image of the object presumed to be present in the non-detection area is far from the edge of the image captured by the imaging means, the object is determined to be a flying object. It is characterized by the following: [Effects of the Invention]

[0007] According to the present invention, by inferring the presence of an object in a non-detection area located between the distance measuring unit and the detection area detected by the distance measuring unit, it becomes possible to detect an object in a non-detection area that is originally outside the detection range of the distance measuring unit. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a diagram illustrating a schematic configuration of a security system using the object detection device according to the present invention. [Figure 2] Figure 2 is a perspective view showing an example of an object detection device. [Figure 3] Figure 3 shows an example of the installation of an object detection device. [Figure 4] Figure 4 is a block diagram showing an example of the configuration of an object detection device. [Figure 5] Figure 5 shows an example where an image of a moving object is obtained using only one camera. [Figure 6] Figure 6 shows another example where only one camera can capture an image of a moving object. [Figure 7] Figure 7 is a flowchart showing the processing flow in the object detection unit of the object detection device. [Modes for carrying out the invention]

[0009] Embodiments of the object detection device, security system, object detection method, and program will be described in detail below with reference to the attached drawings. However, this embodiment does not limit the present invention.

[0010] Figure 1 is a diagram illustrating the schematic configuration of a security system 1 using the object detection device 10 according to the present invention. As shown in Figure 1, the security system 1 comprises an object detection device 10 installed on the object to be protected 20, a security device 40, and a monitoring center 30 located remotely from the object to be protected 20.

[0011] The object detection device 10 and the security device 40, which constitute the security system, are connected by a network 50, for example, a LAN (Local Area Network), and are able to communicate with each other. The communication method between the object detection device 10 and the security device 40 may be wireless communication or wired communication.

[0012] Furthermore, the security device 40 is connected via a network to a fire sensor and a human body detection sensor (not shown), and is capable of receiving fire signals from the fire sensor and detection signals from the human body detection sensor. In addition, the security device 40 is capable of communicating with a remote monitoring center 30 regarding the security target 20 via an external network 60, such as the Internet. Here, the security target 20 is assumed to be a building such as a store, office, tenant building, apartment building, or house.

[0013] The security device 40 has at least two operating modes: "security mode" and "security deactivation mode." It notifies the object detection device 10 of the operating mode set in response to an operation by the administrator or other person of the security target 20. The operating modes will be described later.

[0014] The object detection device 10 transmits the detection information of the object in the security target 20 to the security device 40 via the network 50. Note that the object detection device 10 may not transmit the detection information of the object in the security target 20 according to the operation mode of the security device 40.

[0015] When the security device 40 receives the detection information of the object from the object detection device 10, it reports the abnormality of the security target 20 including the received detection information of the object to the monitoring center 30 via the external network 60 according to the operation mode. In addition, when the security device 40 receives a fire signal from the fire sensor, a detection signal from the human detection sensor, etc., it reports the received signal as an abnormality to the monitoring center 30 according to the operation mode.

[0016] Next, the operation modes of the security device 40 will be briefly described.

[0017] The "security mode" is an operation mode for securing the unmanned security target 20. When the security device 40 set to the "security mode" receives the detection information of the object from the object detection device 10, the fire signal from the fire sensor, the detection signal from the human detection sensor, etc., it reports the detection information of the object and the received signal as an abnormality to the monitoring center 30. Note that the security device 40 may perform an operation to leave a history of the operation together with the operation of reporting the abnormality, or may notify the abnormality to the security target 20 by means of a notification sound, light, etc.

[0018] The "security release mode" is an operation mode when there are relevant persons inside the security target 20 and security is not required. Even when the security device 40 set to the "security release mode" receives the detection information of the object from the object detection device 10, the detection signal from the human detection sensor, etc., it does not determine that there is an abnormality in the security target 20 and does not perform an operation of reporting the abnormality to the monitoring center 30. This is because even if the presence of an object is detected by the object detection device 10 or the like, it is determined that a relevant person has been detected.

[0019] When an abnormality is reported from the security device 40 at the monitoring center 30, the content of the report and the like are displayed on a monitor. For example, based on the display on these monitors, a monitor checks for the occurrence of an abnormality in the security target 20. When the monitor recognizes that an abnormality has occurred within the security target 20, the monitor instructs a standby security guard to go to the security target 20 or makes a report to relevant organizations such as the police and fire department as necessary. Also, at the monitoring center 30, the content of the abnormality reported from the security device 40 and the like are recorded in a recording device. Further, the monitoring center 30 may be configured to transmit a notification of the abnormality to a mobile terminal (such as a mobile phone) of a user (contractor, etc.) of the security system 1.

[0020] Next, the configuration of the object detection device 10 will be described.

[0021] FIG. 2 is a perspective view showing an example of the object detection device 10, and FIG. 3 is a diagram showing an installation example of the object detection device 10. As shown in FIG. 3, the object detection device 10 is installed on the ceiling 90 of the security target 20. Note that the object detection device 10 shown in FIG. 2 shows the installation posture when installed on the ceiling 90.

[0022] The object detection device 10 includes a distance measurement unit 12, which is, for example, a stereo camera composed of two cameras and capable of measuring distance. A stereo camera measures the distance to an object using two cameras in the same way as the human eye. Specifically, it takes pictures with the left and right cameras and calculates the pixel displacement between the two images. Next, based on the pixel displacement, the difference in angles (parallax) from the left and right cameras to the object is calculated. Next, based on the distance between the two cameras (baseline length) and the parallax, the distance to the object is calculated by triangulation. Then, based on the distance to the object, the depth information of the entire screen is grasped.

[0023] As shown in FIG. 2, the object detection device 10 including the distance measurement unit 12, which is a stereo camera, includes two cameras 12a and 12b for imaging the security target 20 inside a protective cover 11, which is a light-transmitting member.

[0024] A detection area A1 is formed around the object detection device 10, which is within the range that the two cameras 12a and 12b can capture, and is an area in which the distance measuring unit 12 can detect a moving object O, as shown in Figure 3.

[0025] As shown in Figure 3, the object detection device 10 is a security sensor installed on the ceiling 90, which monitors the target of security 20 from above.

[0026] The object detection device 10, which is a stereo camera, requires parallax between its two cameras. However, considering the distance between the two cameras (baseline length), a moving object O cannot be imaged by both cameras unless it is at a certain distance from the camera. In other words, as shown in Figure 3, if the object detection device 10 is installed in a position where it is viewed from above the ceiling 90, for example, if an insect flies within 50 cm of the object detection device, the insect will only be imaged by one camera, making it impossible to obtain parallax and thus impossible to measure the distance. Even if it is possible to image the insect with both cameras, the characteristics of the images of the insect captured will be significantly different, making it impossible to detect the pixel misalignment and thus impossible to obtain parallax. In other words, as shown in Figure 3, the object detection device 10 treats the immediate vicinity as a non-detection area A0, where detection is impossible in principle.

[0027] Figure 4 is a block diagram showing an example of the configuration of the object detection device 10. As shown in Figure 4, the object detection device 10 includes a distance measuring unit 12, a control unit 13, a storage unit 14, and a notification unit 15.

[0028] The distance measuring unit 12 is, for example, a stereo camera. The distance measuring unit 12 is equipped with two cameras 12a and 12b. The two cameras 12a and 12b are imaging means such as CMOS (Complementary Metal Oxide Semiconductor) sensors or CCD (Charge Coupled Device) sensors, and they capture images of a subject and obtain image data according to the control of the control unit 13.

[0029] The control unit 13 is a computer configuration that includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory) for storing programs, and a RAM (Random Access Memory) used as a work area. The control unit 13 functions as an object detection unit 131 that performs control operations and calculation processing for the detection of moving objects O by the object detection device 10, depending on the operating mode of the security device 40, for example, by having the CPU execute a program stored in the ROM.

[0030] The program executed by the control unit 13 of the object detection device 10 in this embodiment is provided pre-loaded into ROM or the like.

[0031] The program executed by the control unit 13 of the object detection device 10 in this embodiment may be configured to be provided as an installable or executable file recorded on a computer-readable recording medium such as a CD-ROM, flexible disk (FD), CD-R, or DVD (Digital Versatile Disc).

[0032] Furthermore, the program executed by the control unit 13 of the object detection device 10 in this embodiment may be stored on a computer connected to a network such as the Internet and provided by downloading it via the network. Alternatively, the program executed by the control unit 13 of the object detection device 10 in this embodiment may be provided or distributed via a network such as the Internet.

[0033] The object detection unit 131 of the control unit 13 grasps the shape, position, etc. of objects in the detection area A1 in three dimensions based on the distance to the object obtained from the distance measuring unit 12, and determines whether or not there is a moving object O such as an intruder. More specifically, when the parallax of the images captured by the two cameras 12a and 12b is obtained, the object detection unit 131 calculates the distance to the object, grasps the shape, position, etc. of the object in three dimensions, and then detects the moving object O. Then, if the object detection unit 131 determines that there is a moving object O, it outputs detection information to the security device 40.

[0034] In addition, if the distance measuring unit 12 has obtained at least one image of the moving object O but no parallax has been obtained, the object detection unit 131 infers that the moving object O is located in a non-detection area A0, which is between the distance measuring unit 12 and the detection area A1 and in principle cannot detect objects. More specifically, if the moving object O is in the non-detection area A0, only one of the two cameras 12a and 12b will capture an image of the moving object O, or even if both cameras 12a and 12b capture an image of the moving object O, the difference between the two images will be so large that they cannot be recognized as images of the same moving object O, and therefore the parallax cannot be calculated from the pixel shift. Therefore, the object detection unit 131 infers that the moving object O is in the non-detection area A0 if only one image of the moving object O is obtained, or if different images of the moving object O are obtained on both sides.

[0035] Here, Figure 5 shows an example where an image of the moving object O is obtained from only one camera. Figure 5 shows the case where the head of the intruder, the moving object O, has entered the non-detection area A0. In this case, only one of the two cameras 12a and 12b will obtain an image of the head of the moving object O, and parallax cannot be obtained. Furthermore, as shown in 12a of Figure 5, if the image of the moving object O reaches the edge of the imaging range, it is inferred that the moving object O is present, and the moving object O is judged to be an intruder, and detection information is output.

[0036] On the other hand, Figure 6 shows another example where an image of the moving object O is obtained from only one camera. Figure 6 shows the case where the moving object O, which is a flying object such as an insect (e.g., a moth), enters the non-detection area A0. In this case, an image of the moving object O is obtained from only one of the two cameras 12a and 12b, and parallax cannot be acquired. Furthermore, as shown in 12a of Figure 6, if the image of the moving object O is far from the edge of the imaging range, the presence of the moving object O is inferred, and the moving object O is judged to be a flying object such as an insect (e.g., a moth), and detection information is not output.

[0037] The object detection unit 131 obtains an image of the moving object O from at least one camera. After determining that the moving object O is an intruder because the image of the moving object O reaches the edge of the imaging range, the unit continues to monitor whether the image of the moving object O reaches the edge of the imaging range. If it moves away even temporarily, the unit determines that the moving object O is a flying object such as an insect. Thereafter, the unit may continue to determine that the moving object O is a flying object such as an insect even if it reaches the edge of the imaging range again. In this case, detection information is output to the alarm device 40 when the object is determined to be an intruder, and the previously output detection information is canceled when the object is determined to be a flying object such as an insect.

[0038] The memory unit 14 is a memory unit that allows information to be rewritten, such as a FlashROM. The memory unit 14 stores various information used for the detection process of the moving object O.

[0039] The notification unit 15 includes a speaker 15a that emits a notification sound based on the output of the control unit 13, and a display unit 15b such as an LED that emits light based on the output of the control unit 13.

[0040] The operation of the object detection device 10 will be described below. Figure 7 is a flowchart showing the processing flow in the object detection unit 131 of the object detection device 10.

[0041] The object detection unit 131 determines whether an image of the moving object O is captured in at least one of the two cameras 12a and 12b (step S1).

[0042] The object detection unit 131 infers that the moving object O is present in the non-detection area A0 if the distance measuring unit 12 captures an image of the moving object O in at least one of the two cameras 12a and 12b (Yes in step S1) and no parallax is obtained (step S2).

[0043] Furthermore, if the object detection unit 131 detects that neither of the two cameras 12a and 12b has captured an image of the moving object O as measured by the distance measuring unit 12, it will terminate without outputting detection information (No. in step S1).

[0044] Next, if the image of the moving object O captured by at least one of the two cameras 12a and 12b reaches the edge of the imaging range (Yes in step S3), the object detection unit 131 determines that the moving object O is an intruder (step S4) and outputs detection information to the security device 40 (step S5).

[0045] Furthermore, if the image of the moving object O captured by at least one of the two cameras 12a and 12b is far from the edge of the imaging range (No. in step S3), the object detection unit 131 determines that the moving object O is a flying object such as an insect (for example, a moth) (step S6) and terminates without outputting detection information.

[0046] Thus, according to this embodiment, by inferring the presence of an object in a non-detection area located between the distance measuring unit and the detection area detected by the distance measuring unit, it becomes possible to detect an object in a non-detection area that is originally outside the detection range of the distance measuring unit.

[0047] Furthermore, if the moving object O is identified as a flying object such as an insect (e.g., a moth), false alarms can be prevented by either not outputting detection information or canceling previously outputted detection information.

[0048] In addition, the object detection device 10 performs the processing shown in the flowchart of Figure 7, and at the same time, it detects the moving object O and performs a process to determine whether the moving object O is an intruder.

[0049] Furthermore, the security device 40 may not immediately report the detection information received from the object detection device 10 to the monitoring center 30, but may only report it after a predetermined period of time has elapsed. With this configuration, if information is received from the object detection device 10 indicating that the detection information is a false alarm within the predetermined time, a false alarm can be prevented from being sent to the monitoring center.

[0050] Furthermore, among the functions of the embodiment described above, the object detection unit 131 may be configured to receive distance information from the distance measuring unit 12 of the object detection device 10 to the object, in addition to being provided with the security device.

[0051] Furthermore, each function (object detection unit 131) of the embodiment described above can be realized by one or more processing circuits. Here, "processing circuit" in this specification includes processors programmed to execute each function by software, such as processors implemented by electronic circuits, and devices such as ASICs (Application Specific Integrated Circuits), DSPs (Digital Signal Processors), FPGAs (Field Programmable Gate Arrays), and conventional circuit modules designed to execute each function described above. [Explanation of symbols]

[0052] 1. Security System 10 Object detection device 12 Ranging section 12a, 12b Imaging means 40 Security devices 131 Object detection unit A1 Detection Area A0 Non-detection area

Claims

1. A distance measuring unit that obtains the distance to an object by triangulation based on the parallax resulting from the pixel misalignment of the image of the object captured by each of the two imaging means in a detection area that can be captured by the two imaging means, and the baseline lengths of the two imaging means, An object detection unit that grasps the position and shape of the object in three dimensions based on the distance obtained by the distance measuring unit, Equipped with, If the object detection unit has captured an image of the object with at least one of the imaging means by the distance measuring unit but has not obtained the parallax, the object detection unit infers that the object is located in a non-detection area between the distance measuring unit and the detection area where, in principle, the object cannot be detected. If the image of the object inferred to be in the non-detection area is far from the edge of the image captured by the imaging means, the object is determined to be a flying object. An object detection device characterized by the following features.

2. The object detection unit, if an image of the object has been captured by at least one of the imaging means by the distance measuring unit but no parallax has been obtained, infers that the object is present in the undetected area, and if the image of the object inferred to be present in the undetected area reaches the edge of the image captured by the imaging means, determines that the object is an intruder. The object detection device according to claim 1, characterized in that it is a feature of the present invention.

3. The object detection unit, if an image of the object has been captured by at least one of the imaging means by the distance measuring unit but the parallax has not been obtained, infers that the object is in the undetected area, and if the image of the object inferred to be in the undetected area moves even temporarily away from the edge of the image captured by the imaging means, determines that the object is a flying object. The object detection device according to claim 2, characterized in that it is a feature of the present invention.

4. If the object detection unit determines that the object is a flying object, it will incorrectly report that the object is an intruder. The object detection device according to claim 3.

5. A distance measuring unit that obtains the distance to an object by triangulation based on the parallax resulting from the pixel misalignment of the image of the object captured by each of the two imaging means in a detection area that can be captured by the two imaging means, and the baseline lengths of the two imaging means, A security device that grasps the position and shape of an object in three dimensions based on the distance obtained by the distance measuring unit, A security system equipped with, If the distance measuring unit captures an image of the object with at least one of the imaging means, but no parallax is obtained, the security device infers that the object is located in a non-detection area between the distance measuring unit and the detection area, where in principle the object cannot be detected. If the image of the object inferred to be in the non-detection area is far from the edge of the image captured by the imaging means, the security device determines that the object is a flying object. A security system characterized by the following features.

6. An object detection method in a system in which the position and shape of an object are determined in three dimensions based on the distance obtained by a distance measuring unit, which uses the parallax based on the pixel displacement of images of an object captured by each imaging unit in a detection area that can be captured by two imaging means and the baseline lengths of the two imaging means to obtain the distance to the object by triangulation, If the distance measuring unit captures an image of the object with at least one of the imaging means, but the parallax is not obtained, it is inferred that the object is located in a non-detection area between the distance measuring unit and the detection area, where in principle the object cannot be detected. If the image of the object inferred to be in the non-detection area is far from the edge of the image captured by the imaging means, the object is determined to be a flying object. A method for detecting an object characterized by the following features.

7. Computers, The system functions as an object detection unit that determines the position and shape of an object in three dimensions based on the distance obtained by a distance measuring unit, which uses the parallax based on the pixel displacement of the images of an object captured by each imaging unit within a detection area that can be captured by the two imaging units, and the baseline lengths of the two imaging units to obtain the distance to the object using triangulation. If the object detection unit has captured an image of the object with at least one of the imaging means by the distance measuring unit but has not obtained the parallax, the object detection unit infers that the object is located in a non-detection area between the distance measuring unit and the detection area where, in principle, the object cannot be detected. If the image of the object inferred to be in the non-detection area is far from the edge of the image captured by the imaging means, the object is determined to be a flying object. program.

Images