Vehicle entry / exit management program and vehicle entry / exit management device for parking lot

Abstract

To provide a vehicle entry / exit management program and a vehicle entry / exit management device for a parking lot, with an installation cost reduced.SOLUTION: In step 701, vehicle front side image data is acquired. In step 702, the vehicle front side image data is input to determine whether entry side vehicle front side image data has been extracted using a trained vehicle front side image training control model that extracts a feature quantity using training vehicle front side image data as input data. In step 703, a vehicle number is recognized from the extracted entry side vehicle front side image data. In step 704, a bar of an entry unit is raised. In step 705, vehicle rear side image data is acquired. In step 706, the vehicle rear side image data is input to determine whether entry side vehicle rear side image data has been extracted using a trained vehicle rear side image training control model that extracts a feature quantity using training vehicle rear side image data as input data. In step 707, the bar of the entry unit is lowered.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to a vehicle entrance / exit management program and a vehicle entrance / exit management device for a parking lot. [Background technology]

[0002] In a conventional vehicle entrance / exit management device for a parking lot, a gate machine for raising / lowering a bar to stop a vehicle from passing is provided at the entrance and exit of the parking lot.

[0003] In the above-mentioned conventional vehicle entry / exit management device for parking lots, the entry gate is provided with a ticket issuing machine, and an entry loop coil is embedded near the entry gate to detect vehicles. As a result, when an operator operates the ticket issuing machine after a vehicle is detected by the entry loop coil, the bar of the entry gate rises, and after the vehicle passes through the entry gate, the bar of the entry gate lowers. On the other hand, the exit gate is provided with a payment machine, and an exit loop coil is embedded near the exit gate to detect vehicles. As a result, when an operator operates the payment machine after a vehicle is detected by the exit loop coil, the bar of the exit gate rises, and after the vehicle passes through the exit gate, the bar of the exit gate lowers. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2022-54632 A (Patent No. 7277879) Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the above-mentioned conventional parking lot vehicle entry / exit management device, the loop coil is embedded near the gate machine, which poses a problem of high equipment costs. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, the parking lot vehicle entry and exit management program of the present invention is a parking lot equipped with a gate machine having a bar for entering or exiting the parking lot, a front-of-vehicle camera installed near the gate machine for acquiring a front image of the vehicle heading towards the bar, and a rear-of-vehicle camera installed near the gate machine for acquiring a rear image of the vehicle leaving the bar, and is configured to cause a computer to execute the following steps: a front-of-vehicle image extraction procedure for extracting front-of-vehicle image data of the vehicle from an image taken by the front-of-vehicle camera, a vehicle number acquisition procedure for acquiring the vehicle number of the vehicle after extracting the front-of-vehicle image data, a bar raising procedure for raising the bar after acquiring the vehicle number, a vehicle rear image extraction procedure for extracting rear-of-vehicle image data of the vehicle from an image taken by the rear-of-vehicle camera after the bar is raised, and a bar lowering procedure for lowering the bar after extracting the rear-of-vehicle image data.

[0007] In addition, the parking lot vehicle entry and exit management device of the present invention is a parking lot equipped with a gate machine having a bar for entering or exiting, a front-of-vehicle camera installed near the gate machine for acquiring a front image of the vehicle heading towards the bar, and a rear-of-vehicle camera installed near the gate machine for acquiring a rear image of the vehicle leaving the bar, and is equipped with a front-of-vehicle image extraction means for extracting front-of-vehicle image data of the vehicle from an image from the front-of-vehicle camera, a vehicle number acquisition means for acquiring the vehicle number of the vehicle after extracting the front-of-vehicle image data, a bar raising means for raising the bar after acquiring the vehicle number, a vehicle rear image extraction means for extracting rear-of-vehicle image data of the vehicle from an image from the rear-of-vehicle camera after the bar is raised, and a bar lowering means for lowering the bar after extracting the rear-of-vehicle image data. Effect of the Invention

[0008] According to the present invention, since a loop coil is not required, the equipment cost can be reduced. [Brief description of the drawings]

[0009] [Figure 1] 1 is a diagram showing a parking lot including a first embodiment of a vehicle entry / exit management device according to the present invention; [Diagram 2] FIG. 2 is a diagram showing details of the storage unit of FIG. 1. [Diagram 3] FIG. 2 is a diagram showing details of the delivery unit of FIG. 1. [Figure 4] 2 is a detailed block circuit diagram of the vehicle entrance / exit management device of FIG. 1. [Diagram 5] 5 is a flowchart for explaining a vehicle front image learning operation of the central processing unit of FIG. 4. [Figure 6] 5 is a flowchart for explaining a vehicle rear image learning operation of the central processing unit of FIG. 4. [Figure 7] 5 is a flowchart showing details of a storage routine (main routine 1) executed by the central processing unit of FIG. 4. [Figure 8] 5 is a flowchart showing details of a delivery routine (main routine 2) executed by the central processing unit of FIG. 4. [Figure 9] 5 is a block diagram showing another example of a vehicle entry / exit management device for the parking lot in FIG. 4. [Figure 10] FIG. 2 is a diagram showing a parking lot including a vehicle entrance / exit management device according to a second embodiment of the present invention. [Figure 11] FIG. 11 is a diagram showing details of the loading / unloading unit of FIG. 10. [Figure 12] 11 is a detailed block circuit diagram of the vehicle entry / exit management device of FIG. 10. [Figure 13] 13 is a flowchart showing a loading / unloading routine (main routine) executed by the central processing unit of FIG. 12. [Figure 14] 14 is a flowchart showing details of the receiving routine of FIG. 13. [Figure 15] 14 is a flowchart showing details of the delivery routine of FIG. 13. [Figure 16] 13 is a block diagram showing another example of a vehicle entry / exit management device for the parking lot in FIG. 12. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Fig. 1 is a diagram showing a parking lot including a vehicle entrance / exit management device according to a first embodiment of the present invention. In Fig. 1, a vehicle entrance IN is exclusively for vehicle entrance, and a vehicle exit OUT is exclusively for vehicle exit.

[0011] In Fig. 1, a parking lot 10 is provided with, for example, four vehicle spaces 10-1, 10-2, 10-3, and 10-4, a storage unit 20 is provided at a vehicle entrance IN, and a storage unit 30 is provided at a vehicle exit OUT, and the storage unit 20 and the storage unit 30 are connected to a vehicle entrance / exit management device 40. A payment machine 40a is connected to the vehicle entrance / exit management device 40, which is located within the parking lot 10 or in close proximity to the parking lot 10. Furthermore, the vehicle entrance / exit management device 40 is connected to an information terminal 60 or the like via the Internet 50 as necessary. An operator can operate the vehicle entrance / exit management device 40 via the information terminal 60 or directly.

[0012] FIG. 2 is a diagram showing the receiving unit 20 of FIG.

[0013] As shown in Figure 2, the storage unit 20 is composed of a bar 21, a gate machine 22 for raising and lowering the bar 21, a vehicle front camera (e.g., a monocular camera) 24 for acquiring a forward image of the vehicle W fixed below the pole 23, and a vehicle rear camera (e.g., a monocular camera) 25 for acquiring a rear image of the vehicle W fixed above the pole 23.

[0014] FIG. 3 is a diagram showing the delivery unit 30 of FIG.

[0015] As shown in Figure 3, the exit unit 30 is composed of a bar 31, a gate machine 32 for raising and lowering the bar 31, a vehicle front camera (e.g., a monocular camera) 34 for acquiring a forward image of the vehicle W fixed below the pole 33, and a vehicle rear camera (e.g., a monocular camera) 35 for acquiring a rear image of the vehicle W fixed above the pole 33.

[0016] FIG. 4 is a detailed block circuit diagram of the vehicle entry / exit management device 40 of FIG.

[0017] In FIG. 4, the vehicle entry / exit management device 40 is configured by a computer, and in detail, includes a central processing unit (CPU) 41, a read-only memory (ROM) 42 for storing programs, a flash memory 43, a random access memory (RAM) 44 for storing temporary data, an input / output (I / O) interface 45 having a USB (registered trademark) terminal, interrupt terminals INT1, INT2, etc., an image interface 46 connected to the cameras 24, 25, 34, 35, a communication interface 47 connected to the Internet 3, a vehicle front image learning control model 48, a vehicle rear image learning control model 49, etc. Each learning control model 48, 49 is a deep learning neural network consisting of an input layer, multiple intermediate layers, and an output layer. In the neural network, learning is performed by adjusting the weights and biases of a multilayer perceptron by the error backpropagation method.

[0018] 5 shows a vehicle front image learning operation routine executed by the CPU 41 in response to an interrupt from the operator via the terminal INT1. This learning operation routine is stored in the ROM 42 or the flash memory 43.

[0019] First, in step 501, a plurality of learning vehicle front image data recognized by the operator are acquired from the vehicle front camera 24, 34 or a USB memory previously acquired, etc. The more the recognized learning vehicle front image data is, the greater the learning effect in step 502 is.

[0020] In step 502, a vehicle front image learning process is performed. That is, a vehicle front image learning control model 48 that extracts feature amounts using a large number of recognized learning vehicle front image data as input data is trained by deep learning.

[0021] Then, the routine of FIG.

[0022] 6 shows a vehicle rear image learning operation routine executed by the CPU 41 in response to an interrupt from the operator via the terminal INT2. This learning operation routine is stored in the ROM 42 or the flash memory 43.

[0023] First, in step 601, a plurality of learning vehicle rear image data recognized by the operator are acquired from the vehicle rear camera 25, 35 or a USB memory previously acquired, etc. The greater the number of recognized learning vehicle rear image data, the greater the learning effect in step 602.

[0024] In step 602, a vehicle rear image learning process is performed. That is, a vehicle rear image learning control model 49 that extracts features using a large number of recognized learning vehicle rear image data as input data is trained by deep learning.

[0025] Then, the routine of FIG.

[0026] 5 and 6 may be executed by the operator at any time thereafter to update the learning values. Alternatively, the learning operations may be executed on a cloud connected to the Internet 50, and the learning results may be introduced into the learning control model 48 via the Internet 50.

[0027] 7 is a flowchart showing details of the entrance routine (main routine 1) executed by the CPU 41. In this case, it is assumed that the vehicle W is proceeding from outside the parking lot 10 toward the entrance unit 20 in FIG.

[0028] First, in step 701, image data from the vehicle front camera 24 is acquired.

[0029] Next, in step 702, the trained vehicle front image learning control model 48, which extracts features using the recognized learning vehicle front image data as input data, is used to input image data from the vehicle front camera 24 to determine whether or not the entering-side vehicle front image data has been extracted. If the entering-side vehicle front image data has been extracted, the process proceeds to step 703, and if the entering-side vehicle front image data has not been extracted, the process returns to step 701 and the above-mentioned process is repeated.

[0030] Next, in step 703, the vehicle number is recognized from the image data of the front of the entering vehicle extracted in step 702. That is, the vehicle number recognition process includes a black and white binary data conversion procedure for converting the image data of the front of the vehicle extracted in step 702 into black and white binary data, a rectangular contour area extraction procedure for extracting the connected contours of connected black data or white data of the black and white binary data as a plurality of rectangular contour areas, a number and dot area extraction procedure for determining and extracting a plurality of number areas and dot areas from the plurality of rectangular contour areas, and a vehicle number extraction procedure for extracting the vehicle number of the vehicle from the number area and dot area. The number and dot area extraction procedure includes a Boolean image conversion procedure for converting an image within each rectangular contour area into a Boolean image, a Boolean image and a Boolean number template are compared by pixel-by-pixel Boolean operation, and when the similarity between the Boolean image and the Boolean number template is equal to or greater than a first threshold or equal to or less than a second threshold that is smaller than the first threshold, each rectangular contour area is extracted. The method has a number / dot discrimination and extraction procedure for discriminating that the contour area is a number / dot area and extracting it, and the vehicle number extraction processing procedure has an average value calculation procedure for calculating the average height of all the number / dot areas, a first selection procedure for selecting one of the number / dot areas as a first area, a second selection procedure for selecting a number / dot area within a range of a predetermined multiple of the average height on the right side of the first area as a second area, a third selection procedure for selecting a number / dot area between the first and second areas as a third area, a vehicle number candidate selection procedure for setting the first, second and third areas as vehicle number candidates when there is a predetermined number of third areas, and a vehicle number selection procedure for selecting the vehicle number candidate as the vehicle number of the vehicle when there is one vehicle number candidate, and selecting the vehicle number candidate that is most likely to satisfy a predetermined condition as the vehicle number of the vehicle when there is multiple vehicle number candidates (see Patent Document 1).

[0031] Next, in step 704, the bar 21 is raised by the gate device 22 of the storage unit 20. As a result, the vehicle W can pass under the bar 21.

[0032] When the vehicle W passes under the bar 21, image data is acquired from the vehicle rear camera 25 in step 705.

[0033] Next, in step 706, the trained vehicle rear image learning control model 49, which extracts features using the recognized learning vehicle rear image data as input data, is used to input image data from the vehicle rear camera 25 to determine whether or not the entering vehicle rear image data has been extracted. If the entering vehicle rear image data has been extracted, the process proceeds to step 707, and if the entering vehicle rear image data has not been extracted, the process returns to step 705.

[0034] Next, in step 707, the bar 21 is lowered by the gate device 22 of the storage unit 20.

[0035] Finally, in step 708, the management order creation is started in the flash memory 43 or the RAM 44. For example, the management order is composed of the entry time, the vehicle number, the image data of the front side of the vehicle at the time of entry, and the image data of the rear side of the vehicle at the time of entry for each management order number.

[0036] 8 is a flowchart showing details of the leaving routine (main routine 2) executed by the CPU 41. In this case, it is assumed that the vehicle W is proceeding from the parking lot 10 toward the leaving unit 30 in FIG.

[0037] First, in step 801, image data from the vehicle front camera 34 is acquired.

[0038] Next, in step 802, the trained vehicle front image learning control model 48, which extracts features using the learning vehicle front image data recognized by the operator as input data, is used to input image data from the vehicle front camera 34 to determine whether or not the leaving-side vehicle front image data has been extracted. If the leaving-side vehicle front image data has been extracted, the process proceeds to step 803, and if the leaving-side vehicle front image data has not been extracted, the process returns to step 801 and the above-mentioned process is repeated.

[0039] Next, in step 803, the vehicle number is recognized from the image data of the front of the leaving vehicle extracted in step 802. That is, the vehicle number recognition process includes a black and white binary data conversion procedure for converting the image data of the front of the vehicle extracted in step 802 into black and white binary data, a rectangular contour area extraction procedure for extracting the connected contours of connected black data or white data of the black and white binary data as a plurality of rectangular contour areas, a number and dot area extraction procedure for determining and extracting a plurality of number areas and dot areas from the plurality of rectangular contour areas, and a vehicle number extraction procedure for extracting the vehicle number of the vehicle from the number area and dot area. The number and dot area extraction procedure includes a Boolean image conversion procedure for converting an image within each rectangular contour area into a Boolean image, and a Boolean image and a Boolean number template are compared by pixel-by-pixel Boolean operation to extract each rectangular contour area when the similarity between the Boolean image and the Boolean number template is equal to or greater than a first threshold or equal to or less than a second threshold that is smaller than the first threshold. The method has a number / dot discrimination and extraction procedure for discriminating that the contour area is a number / dot area and extracting it, and the vehicle number extraction processing procedure has an average value calculation procedure for calculating the average height of all the number / dot areas, a first selection procedure for selecting one of the number / dot areas as a first area, a second selection procedure for selecting a number / dot area within a range of a predetermined multiple of the average height on the right side of the first area as a second area, a third selection procedure for selecting a number / dot area between the first and second areas as a third area, a vehicle number candidate selection procedure for setting the first, second and third areas as vehicle number candidates when there is a predetermined number of third areas, and a vehicle number selection procedure for selecting the vehicle number candidate as the vehicle number of the vehicle when there is one vehicle number candidate, and selecting the vehicle number candidate that is most likely to satisfy a predetermined condition as the vehicle number of the vehicle when there is multiple vehicle number candidates (see Patent Document 1).

[0040] Next, in step 804, it is determined whether or not the payment has been settled by the settlement machine 40a. Normally, the payment is settled in advance by the settlement machine 40a. If the payment has not been settled, the process proceeds to the next step 805 after waiting for the payment to be settled. Alternatively, even if the payment has not been settled, the process proceeds to the next step 805, and it is also possible to write "unsettled" at the end of creating the management order in step 809 described below.

[0041] Next, in step 805, the bar 31 is raised by the gate device 32 of the outgoing unit 30. As a result, the vehicle W can pass under the bar 31.

[0042] When the vehicle W passes under the bar 31, image data is acquired from the vehicle rear camera 35 in step 806.

[0043] Next, in step 807, the trained vehicle rear image learning control model 49, which extracts features using the recognized learning vehicle rear image data as input data, is used to input image data from the vehicle rear camera 35 to determine whether or not the leaving-side vehicle rear image data has been extracted. If the leaving-side vehicle rear image data has been extracted, the process proceeds to step 808, and if the leaving-side vehicle rear image data has not been extracted, the process returns to step 806.

[0044] Next, in step 808, the bar 31 is lowered by the gate device 32 of the outgoing unit 30.

[0045] Finally, in step 809, creation of a management order is started in the flash memory 43 or the RAM 44. For example, a management order is composed of, for each management order number, the entry time, the vehicle number, front image data of the vehicle at the time of entry, rear image data of the vehicle at the time of entry, the departure time, front image data of the vehicle at the time of departure, and rear image data of the vehicle at the time of departure.

[0046] The storage routine (main routine 1) in FIG. 7 and the storage routine (main routine 2) in FIG. 8 are both main routines, and therefore are executed in parallel by multiple processes.

[0047] The vehicle entry / exit management device of FIG. 4 can also be configured with hardware shown in FIG.

[0048] In FIG. 9, a parking lot is equipped with a gate machine 22 (32) having a bar 21 (31) for allowing a vehicle to enter or leave the parking lot, a vehicle front camera 24 (34) provided near the gate machine 22 (32) for capturing a front image of the vehicle W approaching the bar 21 (31), and a vehicle rear camera 25 (35) provided near the gate machine 22 (32) for capturing a rear image of the vehicle W leaving the bar 21 (31). In the parking lot, a vehicle entry / exit management device determines the position of the vehicle W from the image of the vehicle front camera 24 (34). The system includes a vehicle front image extraction means 901 for extracting both front image data, a vehicle number acquisition means 902 for acquiring the vehicle number of the vehicle W after extracting the vehicle front image data, a bar raising means 903 for raising the bar 21 (31) after acquiring the vehicle number, a vehicle rear image extraction means 904 for extracting vehicle rear image data of the vehicle W from the image of the vehicle rear camera 25 (35) after the bar 21 (31) is raised, and a bar lowering means 905 for lowering the bar 21 (31) after the vehicle rear image data is extracted.

[0049] Fig. 10 is a diagram showing a parking lot including a vehicle entrance / exit management device according to the second embodiment of the present invention. In Fig. 10, a vehicle entrance / exit G is a vehicle entrance / exit.

[0050] In FIG. 10, a vehicle entrance / exit G is provided with an entrance / exit unit 20', which is connected to a vehicle entrance / exit management device 40'.

[0051] FIG. 11 is a diagram showing the loading / unloading unit 20' of FIG.

[0052] As shown in FIG. 11, the entrance / exit unit 20' is composed of a bar 21', a gate machine 22' for raising and lowering the bar 21', an outward facing camera (e.g., monocular camera) 24' for acquiring a front image or a rear image of the vehicle W fixed below the pole 23', and an inward facing camera (e.g., monocular camera) 25' for acquiring a rear image or a front image of the vehicle W fixed above the pole 23'. That is, the outward facing camera 24' faces the outside of the parking lot 10, and therefore, when the vehicle W enters the parking lot, a front image of the vehicle W is acquired, and on the other hand, when the vehicle W leaves the parking lot, a rear image of the vehicle W is acquired. The inward facing camera 25' faces the inside of the parking lot 10, and therefore, when the vehicle W enters the parking lot, a rear image of the vehicle W is acquired, and on the other hand, when the vehicle W leaves the parking lot, a front image of the vehicle W is acquired.

[0053] FIG. 12 is a detailed block circuit diagram of the vehicle entry / exit control device 40' of FIG.

[0054] In FIG. 12, an outward facing camera 24' and an inward facing camera 25' are connected to an image interface 46 of a vehicle entrance / exit management device 40'.

[0055] FIG. 13 is a flowchart showing a main routine executed by the CPU 41 of FIG.

[0056] The routine of FIG. 13 is substantially executed when the vehicle W approaches the entry / exit unit 20'. That is, when the vehicle is entered, the vehicle is detected by the outward facing camera 24' and its image is acquired, whereas when the vehicle is taken out, the vehicle is detected by the inward facing camera 25' and its image is acquired. Therefore, in step 1301, it is determined whether or not an image of the vehicle is present on the outward facing camera 24', that is, whether the vehicle is entering the garage or not, whereas in step 1302, it is determined whether or not an image of the vehicle is present on the inward facing camera 25', that is, whether the vehicle is leaving the garage or not. It is not necessary to determine whether the image is the front or rear of the vehicle in steps 1301 and 1302. As a result, when it is determined that the vehicle is entering the garage, the process proceeds to step 1303 to perform entry processing, whereas when it is determined that the vehicle is leaving the garage, the process proceeds to step 1304 to perform exit processing, and when the vehicle is not present, the process returns to step 1301.

[0057] FIG. 14 is a flow chart showing the details of the receiving process step 1303 in FIG.

[0058] In step 1401, image data from the outward facing camera 24' is acquired.

[0059] Next, in step 1402, the trained vehicle front image learning control model 48, which extracts features using the recognized learning vehicle front image data as input data, is used to input image data from the outward facing camera 24' to determine whether the vehicle front image data has been extracted. If the vehicle front image data has been extracted, the process proceeds to step 1403, and if the vehicle front image data has not been extracted, the process returns to step 1401 and the above-mentioned process is repeated.

[0060] Next, in step 1403, the vehicle number is recognized from the vehicle front image data extracted in step 1402. That is, the vehicle number recognition process includes a black and white binary data conversion procedure for converting the vehicle front image data extracted in step 1402 into black and white binary data, a rectangular contour area extraction procedure for extracting connected contours of connected black data or white data of the black and white binary data as a plurality of rectangular contour areas, a number and dot area extraction procedure for determining and extracting a plurality of number areas and dot areas from the plurality of rectangular contour areas, and a vehicle number extraction procedure for extracting the vehicle number of the vehicle from the number area and dot area. The number and dot area extraction procedure includes a Boolean image conversion procedure for converting an image within each rectangular contour area into a Boolean image, and a Boolean image and a Boolean number template are compared by pixel-by-pixel Boolean operation to extract each rectangular contour area when the similarity between the Boolean image and the Boolean number template is equal to or greater than a first threshold or equal to or less than a second threshold that is smaller than the first threshold. The method has a number / dot discrimination and extraction procedure for discriminating that the contour area is a number / dot area and extracting it, and the vehicle number extraction processing procedure has an average value calculation procedure for calculating the average height of all the number / dot areas, a first selection procedure for selecting one of the number / dot areas as a first area, a second selection procedure for selecting a number / dot area within a range of a predetermined multiple of the average height on the right side of the first area as a second area, a third selection procedure for selecting a number / dot area between the first and second areas as a third area, a vehicle number candidate selection procedure for setting the first, second and third areas as vehicle number candidates when there is a predetermined number of third areas, and a vehicle number selection procedure for selecting the vehicle number candidate as the vehicle number of the vehicle when there is one vehicle number candidate, and selecting the vehicle number candidate that is most likely to satisfy a predetermined condition as the vehicle number of the vehicle when there is multiple vehicle number candidates (see Patent Document 1).

[0061] Next, in step 1404, the bar 21' is raised by the gate device 22' of the entrance / exit unit 20'. As a result, the vehicle W can pass under the bar 21'.

[0062] When the vehicle W passes under the bar 21', in step 1405, image data is acquired from the inward facing camera 25'.

[0063] Next, in step 1406, the trained vehicle rear image learning control model 49, which extracts features using the recognized learning vehicle rear image data as input data, is used to input image data from the inward facing camera 25' to determine whether vehicle rear image data has been extracted. If vehicle rear image data has been extracted, the process proceeds to step 1407, and if vehicle rear image data has not been extracted, the process returns to step 1405.

[0064] Next, in step 1407, the bar 21' is lowered by the gate device 22' of the loading / unloading unit 20'.

[0065] Finally, in step 1408, creation of a management order is started in the flash memory 43 or RAM 44. For example, a management order is composed of the entry time, the vehicle number, image data of the front of the vehicle at the time of entry, and image data of the rear of the vehicle at the time of entry for each management order number.

[0066] FIG. 15 is a flow chart showing the details of the shipping process step 1304 in FIG.

[0067] In step 1501, image data from inward facing camera 25' is acquired.

[0068] Next, in step 1502, the trained vehicle front image learning control model 48, which extracts features using the learning vehicle front image data recognized by the operator as input data, is used to input image data from the inward facing camera 25' and determine whether the vehicle front image data has been extracted. If the vehicle front image data has been extracted, the process proceeds to step 1503, and if the vehicle front image data has not been extracted, the process returns to step 1501 and the above-mentioned process is repeated.

[0069] Next, in step 1503, the vehicle number is recognized from the vehicle front image data extracted in step 1502. That is, the vehicle number recognition process includes a black and white binary data conversion procedure for converting the vehicle front image data extracted in step 1502 into black and white binary data, a rectangular contour area extraction procedure for extracting connected contours of connected black data or white data of the black and white binary data as a plurality of rectangular contour areas, a number and dot area extraction procedure for determining and extracting a plurality of number areas and dot areas from the plurality of rectangular contour areas, and a vehicle number extraction procedure for extracting the vehicle number of the vehicle from the number area and dot area. The number and dot area extraction procedure includes a Boolean image conversion procedure for converting an image within each rectangular contour area into a Boolean image, and a Boolean image and a Boolean number template are compared by pixel-by-pixel Boolean operation to extract each rectangular contour area when the similarity between the Boolean image and the Boolean number template is equal to or greater than a first threshold or equal to or less than a second threshold that is smaller than the first threshold. The method has a number / dot discrimination and extraction procedure for discriminating that the contour area is a number / dot area and extracting it, and the vehicle number extraction processing procedure has an average value calculation procedure for calculating the average height of all the number / dot areas, a first selection procedure for selecting one of the number / dot areas as a first area, a second selection procedure for selecting a number / dot area within a range of a predetermined multiple of the average height on the right side of the first area as a second area, a third selection procedure for selecting a number / dot area between the first and second areas as a third area, a vehicle number candidate selection procedure for setting the first, second and third areas as vehicle number candidates when there is a predetermined number of third areas, and a vehicle number selection procedure for selecting the vehicle number candidate as the vehicle number of the vehicle when there is one vehicle number candidate, and selecting the vehicle number candidate that is most likely to satisfy a predetermined condition as the vehicle number of the vehicle when there is multiple vehicle number candidates (see Patent Document 1).

[0070] Next, in step 1504, it is determined whether or not the payment has been settled by the settlement machine 40a. Normally, the payment is settled in advance by the settlement machine 40a. If the payment has not been settled, the process proceeds to the next step 1505 after waiting for the payment to be settled. Alternatively, even if the payment has not been settled, the process proceeds to the next step 1505, and it is also possible to write "unsettled" at the end of creating the management order in step 1509 described below.

[0071] Next, in step 1505, the bar 21' is raised by the gate device 22' of the entrance / exit unit 20'. As a result, the vehicle W can pass under the bar 21'.

[0072] When the vehicle W passes under the bar 21', in step 1506, image data is acquired from the outward facing camera 24'.

[0073] Next, in step 1507, the trained vehicle rear image learning control model 49, which extracts features using the recognized learning vehicle rear image data as input data, is used to input image data from the outward facing camera 24' to determine whether vehicle rear image data has been extracted. If vehicle rear image data has been extracted, the process proceeds to step 1508, and if vehicle rear image data has not been extracted, the process returns to step 1506.

[0074] Next, in step 1508, the bar 21' is lowered by the gate device 22' of the loading / unloading unit 20'.

[0075] Finally, in step 1509, creation of a management order is started in the flash memory 43 or the RAM 44. For example, a management order is composed of, for each management order number, the entry time, the vehicle number, the front image data of the vehicle at the time of entry, the rear image data of the vehicle at the time of entry, the departure time, the front image data of the vehicle at the time of departure, and the rear image data of the vehicle at the time of departure.

[0076] The vehicle entry / exit management device of FIG. 12 can also be configured with hardware shown in FIG.

[0077] In FIG. 16, the parking lot has a gate machine 22' having a bar 21' for entering and leaving the parking lot, an outward camera 24' provided near the gate machine 22', and an inward camera 25' provided near the gate machine 22', and the vehicle entry / exit management device includes a first vehicle front image extraction means 1601-1 for extracting first vehicle front image data of the first vehicle W1 from an image of the outward camera when a first vehicle W1 enters the parking lot, a first vehicle vehicle number acquisition means 1602-1 for acquiring a first vehicle number of the first vehicle W1 after extracting the first vehicle front image data, a first bar raising means 1603-1 for raising the bar 21' after acquiring the first vehicle number, and a first vehicle rear image extraction means for extracting first vehicle rear image data of the first vehicle W1 from an image of the inward camera 25' after the bar 21' is raised by the first bar raising means. When the second vehicle leaves the depot, the depot has a second vehicle front image extraction means 1602-1 for extracting second vehicle front image data of the second vehicle from the image of the inward facing camera 25', a second vehicle number acquisition means 1602-2 for acquiring the second vehicle number of the second vehicle after extracting the second vehicle front image data, a second bar raising means 1603-2 for raising the bar 21' after acquiring the second vehicle number, a second vehicle rear image extraction means 1604-2 for extracting second vehicle rear image data of the second vehicle from the image of the outward facing camera 24' after the bar 21' is raised by the second bar raising means 1603-2, and a second bar lowering means 1605-2 for lowering the bar 21' after acquiring the second rear image data.

[0078] Incidentally, the present invention can be applied to any modifications within the obvious scope of the above-described embodiment. [Industrial Applicability]

[0079] The vehicle entry / exit management program and device for a parking lot according to the present invention can be used in a parking lot billing system, etc. [Explanation of symbols]

[0080] 10. Parking lot 10-1, 10-2, 10-3, 10-4: Vehicle interior 20: Receiving unit 21: Bar 22: Gate machine 23: Paul 24: Front camera 25: Rear camera 30: Delivery unit 31: Bar 32: Gate machine 33: Paul 34: Front camera 35: Rear camera 20': Storage and retrieval unit 21': Bar 22': Gate machine 23': Paul 24': Exterior camera 25': Inward facing camera 40', 40: Vehicle entry / exit management device 40a: Ticket counter 41: CPU 42:ROM 43: Flash memory 44:RAM 45: I / O interface 46: Image interface 47: Communication interface 48: Vehicle forward image learning control model 49: Vehicle rear image learning control model 50: Internet 60: Information terminal IN: Vehicle entrance OUT: Vehicle exit G: Vehicle entrance / exit

Claims

1. A gate machine having a bar for entering or leaving the warehouse; a vehicle front camera provided near the gate machine for capturing a forward image of a vehicle approaching the bar; a vehicle rear camera provided near the gate machine for capturing a rear image of the vehicle moving away from the bar; A parking lot comprising: a vehicle front image extraction step of extracting vehicle front image data of the vehicle from an image of the vehicle front camera; a vehicle number acquisition step of acquiring a vehicle number of the vehicle after extracting the vehicle front image data; a bar raising step of raising the bar after obtaining the vehicle number; a vehicle rear image extraction step of extracting vehicle rear image data of the vehicle from an image of the vehicle rear camera after the bar is raised; a bar lowering step of lowering the bar after extracting the vehicle rear image data; A parking lot vehicle entry / exit management program for causing a computer to execute the above.

2. a gate machine having a bar for entering and leaving the warehouse; An outward-facing camera provided near the gate machine; an inward-facing camera provided near the gate machine; A parking lot comprising: When the first vehicle enters the warehouse, a first vehicle front image extraction step of extracting first vehicle front image data of the first vehicle from an image of the outward-facing camera; a first vehicle number acquisition step of acquiring a vehicle number of the first vehicle after extracting the first vehicle front image data; a first bar raising step of raising the bar after obtaining the first vehicle number; a first vehicle rear image extraction step of extracting first vehicle rear image data of the first vehicle from an image of the inward-facing camera after the bar is raised by the first bar raising step; a first bar lowering step of lowering the bar after extracting the first vehicle rear image data; by a computer, When the second vehicle leaves the depot, a second vehicle front image extraction step of extracting second vehicle front image data of the second vehicle from an image of the inward-facing camera; a second vehicle number acquisition step of acquiring a second vehicle number of the second vehicle after extracting the second vehicle front image data; a second bar raising step of raising the bar after obtaining the second vehicle number; a second vehicle rear image extraction step of extracting second vehicle rear image data of the second vehicle from an image of the outward facing camera after the bar is raised by the second bar raising step; a second bar lowering step of lowering the bar after acquiring the second vehicle rear image data; A parking lot vehicle entry / exit management program for causing a computer to execute the above.

3. The vehicle front image extraction procedure includes:

2. A parking lot vehicle entry / exit management program to be executed by a computer as described in claim 1, comprising a vehicle front image estimation / extraction procedure that uses a learned vehicle front learning control model with learning vehicle front image data as input data, and estimates and extracts the vehicle front image data with an image from the vehicle front camera as input data.

4. The first vehicle front image extraction procedure or the second vehicle front image extraction procedure includes:

3. A parking lot vehicle entry / exit management program to be executed by a computer as described in claim 2, comprising a vehicle front image estimation / extraction procedure that uses a learned vehicle front learning control model with learning vehicle front image data as input data, and estimates and extracts the first vehicle front image data or the second vehicle front image data with images from the outward facing camera or the inward facing camera as input data.

5. The vehicle rear image extraction procedure includes:

2. A parking lot vehicle entry / exit management program to be executed by a computer as described in claim 1, comprising a vehicle rear image estimation / extraction procedure for estimating and extracting the vehicle rear image data using a learned vehicle rear learning control model with learning vehicle rear image data as input data and an image from the vehicle rear camera as input data.

6. The first vehicle rear image extraction procedure or the second vehicle rear image extraction procedure includes:

3. A parking lot vehicle entry / exit management program for managing parking lots according to claim 2, further comprising a vehicle rear image estimation / extraction procedure for estimating and extracting the first vehicle rear image data or the second vehicle rear image data using a learned vehicle rear learning control model with learning vehicle rear image data as input data and an image from the inward-facing camera or the outward-facing camera as input data.

7. A parking lot vehicle entry / exit management program as described in claim 1 or 2, further comprising a management order creation start procedure for starting creation of a management order for the vehicle after the vehicle rear image data is extracted when the vehicle enters the parking lot.

8. The parking lot vehicle entry / exit management program of claim 7, further comprising a management order creation completion procedure for completing the creation of the management order for the vehicle after the vehicle rear image data is extracted when the vehicle leaves the parking lot.

9. A gate machine having a bar for entering or leaving the warehouse; a vehicle front camera provided near the gate machine for capturing a forward image of a vehicle approaching the bar; a vehicle rear camera provided near the gate machine for capturing a rear image of the vehicle moving away from the bar; A parking lot comprising: a vehicle front image extraction unit for extracting vehicle front image data of the vehicle from an image of the vehicle front camera; a vehicle number acquisition means for acquiring a vehicle number of the vehicle after extracting the vehicle front image data; a bar raising means for raising the bar after acquiring the vehicle number; a vehicle rear image extraction means for extracting vehicle rear image data of the vehicle from an image of the vehicle rear camera after the bar is raised; a bar lowering means for lowering the bar of the gate machine after acquiring the rear image data; A vehicle entry / exit management device for a parking lot comprising:

10. a gate machine having a bar for entering and leaving the warehouse; An outward-facing camera provided near the gate machine; an inward-facing camera provided near the gate machine; A parking lot comprising: When the first vehicle enters the warehouse, a first vehicle front image extraction means for extracting first vehicle front image data of the first vehicle from an image of the outward-facing camera; a first vehicle number acquisition means for acquiring a first vehicle number of the first vehicle after extracting the first vehicle front image data; a first bar raising means for raising the bar after acquiring the first vehicle number; a first vehicle rear image extraction means for extracting first vehicle rear image data of the first vehicle from an image of the inward facing camera after the first bar raising means raises the bar; a first bar lowering means for lowering the bar after acquiring the first rear image data; Equipped with When the second vehicle leaves the depot, a second vehicle front image extraction means for extracting second vehicle front image data of the second vehicle from an image of the inward-facing camera; a second vehicle number acquisition means for acquiring a second vehicle number of the second vehicle after extracting the second vehicle front image data; a second bar raising means for raising the bar after acquiring the second vehicle number; a second vehicle rear image extraction means for extracting second vehicle rear image data of the second vehicle from an image of the outward facing camera after the second bar raising means raises the bar; and a second bar lowering means for lowering the bar after the second rear image data is acquired.

11. The vehicle front image extraction means 10. A parking lot vehicle entry / exit management device as described in claim 9, further comprising a vehicle front image estimation / extraction means for estimating and extracting the vehicle front image data using a learned vehicle front learning control model with learning vehicle front image data as input data and an image from the vehicle front camera as input data.

12. The first vehicle front image extraction means or the second vehicle front image extraction means 11. The parking lot vehicle entry / exit management device as described in claim 10, further comprising a vehicle front image estimation / extraction means for estimating and extracting the first vehicle front image data or the second vehicle rear image data using a learned vehicle front learning control model with learning vehicle front image data as input data and an image from the outward facing camera or the inward facing camera as input data.

13. The vehicle rear image extraction means 10. A parking lot vehicle entry / exit management program to be executed by a computer as described in claim 9, further comprising a vehicle rear image estimation / extraction means for estimating and extracting the vehicle rear image data using a learned vehicle rear learning control model with learning vehicle rear image data as input data and an image from the vehicle rear camera as input data.

14. The first vehicle rear image extraction means or the second vehicle rear image extraction means 11. A parking lot vehicle entry / exit management program to be executed by a computer as described in claim 10, further comprising a vehicle rear image estimation / extraction means for estimating and extracting the first vehicle rear image data or the second vehicle rear image data using a learned vehicle rear learning control model with learning vehicle rear image data as input data, and an image from the outward facing camera or the inward facing camera as input data.

15. 11. The parking lot vehicle entry / exit management device according to claim 9 or 10, further comprising a management order creation initiation means for starting creation of a management order for the vehicle after the vehicle rear image data is extracted when the vehicle enters the parking lot.

16. 16. The parking lot vehicle entry / exit management device according to claim 15, further comprising a management order creation terminating means for terminating creation of the management order for the vehicle after the vehicle rear image data is extracted when the vehicle leaves the parking lot.

Images