Vehicle detection device

Abstract

To reliably sense a vehicle in a parking space.SOLUTION: A vehicle sensing device includes: an image recognition unit 10 that is allocated to each parking space in a parking lot and includes a monitoring camera 11 that picks up an image of the inside of the parking space and an image processing unit 12 therefor; a Doppler radar unit 20 that detects a mobile body entering / exiting the parking space; and a signal processing unit 30 that determines whether the mobile body is a vehicle and has parked in the parking space based on vehicle discrimination data on the mobile body output from the image recognition unit 10 and movement data on the mobile body output from the Doppler radar unit 20. Directivity of the mobile body entering / exiting the parking space is sensed by the Doppler radar unit 20. The image recognition unit 10 determines whether the mobile body is a vehicle or any other object. The signal processing unit performs parking determination in a case where the mobile body is a vehicle.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a vehicle detection device, and more particularly to a vehicle detection device in a parking lot.

Background Art

[0002] As a sensor for vehicle detection applied to a parking lot, in a flap (lock plate) type parking lot, mainly a loop coil sensor is used, and in a flapless parking lot, in many cases, ultrasonic, infrared, radar, image sensors, etc. are used in combination.

[0003] Since the loop coil sensor applied to a flap type parking lot is embedded for each parking section (compartment) of a vehicle, not only does the construction cost increase, but in the case of failure, it is necessary to dig it up for repair or replacement. Therefore, in recent years, the number of flapless parking lots has been on the increase trend.

[0004] As a vehicle detection method for a flapless parking lot, in Patent Document 1, a vehicle detection method by vehicle number recognition has been proposed. In this vehicle detection method, a vehicle number image of a vehicle parked in a parking section is periodically read by a surveillance camera, and if it is the same as the previous read image, it is determined that the same vehicle is parked.

[0005] Also, as another vehicle detection method, in Patent Document 2, a vehicle detection method by using radar and ultrasonic waves in combination has been proposed. In this vehicle detection method, after a vehicle entering a parking section is detected by an FMCW radar sensor, an ultrasonic sensor is activated, and the vehicle is detected by both sensors. Since the FMCW radar sensor is not suitable for short-distance detection due to distance resolution, an ultrasonic sensor is used in combination to also handle short-distance detection.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

[0007] One problem with the vehicle detection method using license plate recognition described in Patent Document 1 is that it can be difficult to read the license plate on the surveillance camera when a vehicle enters at night with its headlights on or during bad weather. Also, when a vehicle stops in a parking space, the license plate may not be within the field of view of the surveillance camera, making it impossible to determine whether or not the vehicle is parked.

[0008] Furthermore, the vehicle detection method using radar and ultrasound described in Patent Document 2 detects objects within the monitoring area based on reflected waves from the object. Therefore, it cannot accurately determine what the object is, and there is a possibility that it may also detect objects other than vehicles (for example, people, dogs, wheelbarrows, etc.).

[0009] Therefore, the object of the present invention is to provide a vehicle detection device that can reliably detect vehicles within a parking space. [Means for solving the problem]

[0010] To solve the above problems, the vehicle detection device of the present invention comprises an image recognition unit including a surveillance camera assigned to each parking space in a parking lot and an image processing unit therefor that captures images within the parking space; a Doppler radar unit that detects moving objects entering and exiting the parking space; and a signal processing unit that determines whether the moving object is a vehicle and has parked in the parking space based on vehicle identification data of the moving object output from the image recognition unit and movement data of the moving object output from the Doppler radar unit. The Doppler radar unit detects the direction of the moving object entering and exiting the parking space, the image recognition unit determines whether the moving object is a vehicle or another object, and the signal processing unit makes a parking decision if the moving object is a vehicle.

[0011] In the present invention, the signal processing unit is characterized in that, when the Doppler radar unit detects the moving object within the parking space, the image recognition unit acquires image data from the surveillance camera while the Doppler radar unit is detecting the object, starts image processing, and makes a parking determination if the moving object is a vehicle.

[0012] Furthermore, in the present invention, if the moving object is a vehicle, when the Doppler radar unit stops detecting the vehicle approaching the parking space, it is determined that the vehicle has stopped in the parking space, and after a predetermined time has elapsed... Charge It is characterized by the act of starting.

[0013] Furthermore, the present invention is characterized in that, when the vehicle moves away from the parking space, the image recognition unit determines that the vehicle has left the parking space when it stops detecting the vehicle.

[0014] Furthermore, the present invention is characterized in that, when the vehicle moves away from the parking space without payment being made, the image recognition unit determines that the vehicle has left the parking space without payment being made when it stops detecting the vehicle moving away from the parking space, and saves the image used to determine the moving object. [Effects of the Invention]

[0015] According to the present invention, the image recognition unit can determine what an object is, which was difficult for radar sensors and ultrasonic sensors to do, and by combining this with the direction detection of the Doppler radar unit, it is possible to reliably detect only vehicles within a parking space. [Brief explanation of the drawing]

[0016] [Figure 1] A system block diagram showing a vehicle detection device according to the present invention. [Figure 2] (a)~(d) Schematic diagrams illustrating the basic operation of the present invention. [Figure 3] A flowchart showing an example of the operation of the present invention.

Embodiments for Carrying Out the Invention

[0017] Next, embodiments of the present invention will be described with reference to the drawings, but the present invention is not limited thereto.

[0018] As shown in FIG. 1, the vehicle detection device 1 of the present invention basically includes an image recognition unit 10 including a surveillance camera 11 and its image processing unit 12, a Doppler radar unit 20, and a signal processing unit 30.

[0019] Referring also to FIG. 2, in this embodiment, the image recognition unit 10, the Doppler radar unit 20, and the signal processing unit 30 are arranged on a pole 2 erected beside a parking section (also referred to as a vehicle compartment) S for one vehicle provided in coin parking, for example.

[0020] The surveillance camera 11 of the image recognition unit 10 and the transmitting and receiving units (both not shown) of the Doppler radar unit 20 are directed toward the entrance and exit directions of the parking section S. In addition to the normal Doppler method, for example, an FSK Doppler method or an FMCW method may be applied to the Doppler radar unit 20.

[0021] The signal processing unit 30 determines whether the moving object is approaching or moving away based on the reflected wave from the moving object output from the receiving unit of the Doppler radar unit 20. Further, the signal processing unit 30 determines whether the moving object has parked as a vehicle based on the vehicle discrimination data (data indicating that the moving object is a vehicle) from the image processing unit 12.

[0022] As a preferred aspect, the vehicle detection device 1 includes an LED illumination 13 as illumination means for illuminating the parking section S for use at night or in a dark place such as an underground parking lot. Instead of the LED illumination, an infrared light may be used.

[0023] Here, referring to Figures 2(a) to (d), the basic operation of the vehicle detection device 1 from entry to exit will be explained. It should be assumed that the surveillance camera 11 is constantly capturing images of the parking space S, and that the Doppler radar unit 20 is constantly transmitting Doppler signals from its transmitter toward the parking space S.

[0024] First, as shown in Figure 2(a), when a moving object (in this case, vehicle V) enters the parking space S and the Doppler radar unit 20 detects that the moving object is approaching, the image processing unit 12 starts image processing, and the vehicle identification data is sent to the signal processing unit 30, and if the moving object is a vehicle, a parking decision is made.

[0025] Next, as shown in Figure 2(b), when the moving object enters and stops within the parking space S, the Doppler signal frequency output from the Doppler radar unit 20 becomes 0 Hz. The signal processing unit 30 determines that the moving object has stopped, and if the image recognition results in the identification of the moving object as a vehicle V, it sends a signal to the payment machine 40 to begin entering the parking space after a predetermined time has elapsed. The payment machine 40, upon receiving this, recognizes that the parking space S is occupied (full) and displays this on a display or the like (not shown).

[0026] Next, as shown in Figure 2(c), when the vehicle V reverses to exit the parking space S, the Doppler radar unit 20 transmits a Doppler signal indicating that the vehicle V is moving away to the signal processing unit 30.

[0027] At this time, the image processing unit 12 starts image processing, and the vehicle identification data is sent to the signal processing unit 30. This is because, for example, even if a person slips past a parked vehicle V, the Doppler radar unit 20 outputs a Doppler signal indicating that the moving object is moving away. If, as a result of image recognition, the moving object is determined to be a person rather than a vehicle V, the system will not send an exit signal to the payment machine 40, indicating that parking will continue.

[0028] Then, as shown in Figure 2(d), when the vehicle V has finished exiting the parking space S, the Doppler signal frequency from the Doppler radar unit 20 becomes 0 Hz, and the signal processing unit 30 informs the payment machine 40 that the vehicle has finished exiting. In the case of unpaid parking fees, an image of the parked vehicle is stored in a storage unit (not shown).

[0029] Next, we will explain the specific operation of this vehicle detection device 1 following the flowchart in Figure 3.

[0030] First, in the initial step 100, it is determined whether the Doppler radar unit 20 has detected a moving object (in this case, a vehicle V). If the result is YES and a vehicle V is detected, in the next step 101, it is determined whether the direction of movement of the vehicle V is toward the parking space S.

[0031] If the determination in step 101 is YES and vehicle V is approaching parking space S, then in step 102, a determination is made as to whether detection is currently in progress (vehicle V has already been parked in parking space S for a predetermined time and is being charged).

[0032] If a vehicle V enters an empty parking space S, it is not currently being detected (charged), so the determination in step 102 is NO, and the process proceeds to the next step 103, where it is determined whether the vehicle has been detected before. In this case, since the vehicle has not yet been detected, the image recognition unit 10 performs vehicle detection in step 104.

[0033] If the result is YES and it is determined to be a vehicle, in the next step 105, it is determined whether the stationary (parked) state has elapsed for a predetermined time. If the predetermined time has elapsed, in step 106 it will enter the detection (charging) state. If the vehicle determination in step 104 earlier determines that what approached is not a vehicle, for example, a person, the process returns to the first step 100.

[0034] Furthermore, if the determination in step 105 is NO and the vehicle has not been stopped for the predetermined time (before entering the billing state), the process returns to step 100, and steps 101 onwards are executed until the predetermined time for entering the billing state has elapsed. In this case, the determination result in step 103 is YES (because the vehicle was detected in the previous step 104), and the process moves to step 107, where it is determined whether the vehicle has been stopped for the predetermined time. If the predetermined time has elapsed, the process enters detection mode (billing state) in step 108.

[0035] If the result in step 107 is NO and the vehicle has not been stopped for the predetermined time (before it enters the billing state), the process returns to step 100, and steps 101-103 and 107 are repeated until the vehicle has been stopped for the predetermined time.

[0036] If a vehicle V, once it has entered a parking space S, moves forward within the parking space S to correct its stopping position, for example, the process proceeds through steps 101, 102, and 103 to step 107, where it is determined whether a predetermined time has elapsed since the vehicle was stopped. If the predetermined time has elapsed, the process proceeds to step 108, where the vehicle enters a detection state (charging state).

[0037] In contrast, if a vehicle V that has entered a parking space S moves, for example, in the direction away from it, to correct its parking position, step 101 will result in a NO determination and the process will proceed to step 110. Then, in step 111, it is determined whether the vehicle is currently being detected (charged), and if the result is NO, in step 112 it is determined whether the vehicle has been detected before.

[0038] In this case, since the vehicle has already been identified in step 104, the determination in step 112 is YES, and the process moves to step 113, where it is determined whether the vehicle has been stopped for a predetermined time. If the predetermined time has elapsed, the system will proceed to step 114 and enter a detection state (charging state).

[0039] If the result in step 112 is NO, or if the result in step 113 is NO, the process returns to step 100, and steps 101, 110-113 are repeated.

[0040] Next, if a vehicle V parked in parking space S reverses out, the process proceeds through a NO determination in step 101, a YES determination in step 110, a YES determination in step 111, and finally to a vehicle determination in step 115. In this vehicle determination in step 115, it is determined whether there is a vehicle in parking space S. If there is a vehicle in parking space S, the result is YES; if there is no vehicle in parking space S, the result is NO.

[0041] If the vehicle detection in step 115 is NO, and vehicle V has reversed and is no longer in parking space S, the process moves to the next step, step 116, where it is determined whether a predetermined time has elapsed since the vehicle left the parking space. If the determination is YES and the predetermined time has elapsed since the vehicle left the parking space, the detection is cleared in step 117, and the billing process ends.

[0042] If the vehicle determination in step 115 is YES (meaning vehicle V is still in parking space S), the process returns to step 100, and steps 100, 101, 110, and 111 are repeated.

[0043] Furthermore, if the determination in step 116 is NO, steps 100, 101, 110, 111, and 115 are repeatedly executed until the detection deactivation time is reached.

[0044] As described above, according to the present invention, when a moving object is detected in the parking area (detection area) S by the Doppler radar unit 20, the image recognition unit 10 acquires image data from the surveillance camera 11 while the Doppler radar unit 20 is detecting the object (approaching or moving away) and starts image processing to determine whether the moving object is a vehicle or something else. The signal processing unit 30 makes a parking determination to determine whether the moving object is parked or not if it is a vehicle.

[0045] If the moving object is a vehicle, the Doppler radar unit 20 will stop detecting the vehicle approaching the parking space S, at which point it will be determined that the vehicle has stopped in the parking space S, and detection (billing) will begin after a predetermined time has elapsed.

[0046] Furthermore, for vehicles moving away from parking space S, it may be determined that the vehicle has left parking space S once the image recognition unit 10 stops detecting the vehicle. If the moving object is not a vehicle, the system determines that there is no vehicle in parking space S and then stops the vehicle detection by the image recognition unit 10.

[0047] If a vehicle is moving away from parking space S without having paid, when the Doppler radar unit 20 detects a moving object moving away from parking space S, the image processing unit 12 acquires image data from the surveillance camera 11 while the Doppler radar unit 20 is detecting (moving away from) the object and starts image processing to determine whether the moving object is a vehicle or something else. If the moving object is a vehicle, when the image processing unit 12 stops detecting a moving object moving away from parking space S, it determines that the vehicle has left without paying and saves the image used to determine the moving object.

[0048] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the embodiments described above. Modifications or improvements made to the embodiments described above by those skilled in the art are also included in the technical scope of the present invention. [Explanation of Symbols]

[0049] 1. Vehicle detection device 10 Image Recognition Unit 11 Surveillance cameras 12 Image Processing Unit 13 LED lighting 20 Doppler radar section 30 Image Processing Unit 40 Payment machine

Claims

1. The system includes an image recognition unit that includes a surveillance camera assigned to each parking space in the parking lot and an image processing unit that captures images within the parking space, a Doppler radar unit that detects moving objects entering and exiting the parking space, and a signal processing unit that determines whether the moving object is a vehicle and has parked in the parking space based on vehicle identification data of the moving object output from the image recognition unit and movement data of the moving object output from the Doppler radar unit. A vehicle detection device characterized in that the Doppler radar unit detects the direction of the moving object entering and exiting the parking space, the image recognition unit determines whether the moving object is a vehicle or another object, and the signal processing unit makes a parking decision if the moving object is a vehicle.

2. The vehicle detection device according to claim 1, characterized in that when the Doppler radar unit detects the moving object within the parking space, the image recognition unit acquires image data from the surveillance camera while the Doppler radar unit is detecting the object, starts image processing, and determines that the moving object is a vehicle if it is parked.

3. The vehicle detection device according to claim 2, characterized in that, if the moving object is a vehicle, the device determines that the vehicle has stopped in the parking space when the Doppler radar unit stops detecting the vehicle approaching the parking space, and starts charging after a predetermined time has elapsed.

4. The vehicle detection device according to claim 3, characterized in that when the above-mentioned vehicle moves away from the above-mentioned parking space, the image recognition unit determines that the vehicle that was in the parking space has left the parking space when it stops detecting the vehicle.

5. The vehicle detection device according to claim 2, characterized in that, when the above vehicle moves away from the above parking space without payment being made, the image recognition unit determines that the vehicle has left the parking space without payment being made when it stops detecting the vehicle moving away from the above parking space, and saves the image used to determine the moving object.

Images