Vehicle identification device

Abstract

The present invention provides a vehicle identification device that can individually detect a target vehicle and adjacent vehicles, even when the target vehicle and adjacent vehicles are close together. [Solution] The vehicle identification device estimates the shape of the vehicle contained in the first image data, and compares the shape of the vehicle with a vehicle type database that records vehicle information related to the vehicle shape and the estimation result obtained by estimating the shape of the vehicle based on the first image data. If the shape of the vehicle cannot be matched, the device moves the vehicle forward or backward, and after moving the vehicle, it acquires second image data, estimates the shape of the vehicle contained in the second image data, and estimates the shape of the vehicle with the vehicle type database and the estimation result obtained by estimating the shape of the vehicle based on the second image data.

Description

Technical Field

[0001] The present disclosure relates to a vehicle identification device.

Background Art

[0002] Patent Document 1 discloses a technique for obtaining captured images of objects in a parking lot captured by a plurality of cameras, estimating a second camera that captures a second captured image in which an object is next detected based on a first captured image of the object captured by a first camera, and using this estimation result to monitor the position of a target vehicle.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in Patent Document 1 described above, when there are a plurality of vehicles including a target vehicle and the distance between the target vehicle and an adjacent vehicle is short, there is a problem that the target vehicle and the adjacent vehicle cannot be separated and detected, and the target vehicle cannot be correctly detected.

[0005] The present disclosure has been made in view of the above, and an object thereof is to provide a vehicle identification device that can individually detect a target vehicle and an adjacent vehicle even when the distance between the target vehicle and the adjacent vehicle is short.

Means for Solving the Problems

[0006] To solve the above-mentioned problems and achieve the objective, the vehicle identification device according to the present disclosure is a vehicle identification device for identifying a vehicle in a parking lot, and comprises a processor, which acquires at least first image data captured by a camera of the vehicle, estimates the shape of the vehicle included in the first image data, and compares the shape of the vehicle with a vehicle type database that records vehicle information relating to the vehicle shape and an estimation result obtained by estimating the shape of the vehicle based on the first image data. If the shape of the vehicle cannot be matched, the processor moves the vehicle forward or backward, and after moving the vehicle, acquires second image data captured by the camera of the vehicle, estimates the shape of the vehicle included in the second image data, and estimates the shape of the vehicle with the vehicle type database and an estimation result obtained by estimating the shape of the vehicle based on the second image data. [Effects of the Invention]

[0007] According to this disclosure, even when the target vehicle and adjacent vehicles are close together, the system has the effect of being able to individually detect the target vehicle and adjacent vehicles. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram showing the functional configuration of a vehicle identification device according to one embodiment. [Figure 2] Figure 2 is a flowchart showing an overview of the processes performed by a vehicle identification device according to one embodiment. [Figure 3] Figure 3 shows an example of an image captured from a vehicle identification device according to one embodiment. [Modes for carrying out the invention]

[0009] Hereinafter, a vehicle identification device according to one embodiment of the present disclosure will be described with reference to the drawings. Note that the components in the following embodiment include those that are easily substituted or substantially identical to those of a person skilled in the art. Furthermore, the figures referenced in the following description only schematically show the shape, size, and positional relationships to the extent that the contents of the present disclosure can be understood. In other words, the present disclosure is not limited to the shapes, sizes, and positional relationships exemplified in the figures.

[0010] [Overall configuration of the vehicle identification system] Figure 1 is a block diagram showing the functional configuration of a vehicle identification device according to one embodiment. The vehicle identification device 1 shown in Figure 1 is installed in a parking lot or the like and identifies a target vehicle from among multiple vehicles parked in the parking lot.

[0011] The vehicle identification device 1 shown in Figure 1 comprises a detection device 11, a camera 12, a drive unit 13, a communication unit 14, a vehicle type database 15 (hereinafter referred to as "vehicle type DB 15"), a vehicle type matching result database 16 (hereinafter referred to as "vehicle type matching result DB"), and a control device 17.

[0012] The detection device 11, under the control of the control device 17, irradiates a laser beam onto the target object, detects the distance to the object or the shape of the object based on the information of the reflected light from the object, and outputs this detection result to the control device 17. The detection device 11 is configured using, for example, Lidar (Light Detection And Ranging).

[0013] The camera 12, under the control of the control device 17, captures an image of the target object, generates image data, and outputs this image data to the control device 17. The camera 12 is composed of, for example, a lens and an image sensor.

[0014] The drive unit 13 moves the detection device 11 and camera 12 to predetermined positions under the control of the control device 17. The drive unit 13 is configured using, for example, a motor or a drive mechanism.

[0015] The communication unit 14, under the control of the control device 17, transmits an instruction signal to the target object, the vehicle, to move in the forward and backward directions. The communication unit 14 is configured using a communication module capable of Wi-Fi (Wireless Fidelity) (registered trademark) and Bluetooth (registered trademark), etc.

[0016] The Vehicle DB15 records vehicle information that associates the vehicle name, which identifies the vehicle, with the shape and feature quantities related to each vehicle's parts. The shape of each vehicle refers to the feature quantities, 2D data, and 3D data of the vehicle that have been measured in advance with measuring instruments. The parts of each vehicle refer to the front windshield, A-pillar, headlights, front grille, and front bumper cover, etc.

[0017] The vehicle type matching result DB16 records the presence or absence of a vehicle type identification result for the target vehicle by the control device 17 (described later), and the vehicle type information used for vehicle type identification.

[0018] The control device 17 is implemented using a processor having hardware such as an FPGA (Field-Programmable Gate Array) or a CPU (Central Processing Unit), and a memory that is a temporary storage area used by the processor, which stores software (programs) capable of executing applications (hereinafter simply referred to as "apps"). In one embodiment, the control device 17 functions as a processor. The control device 17 also includes an acquisition unit 171, an estimation unit 172, a discrimination unit 173, a matching unit 174, a verification unit 175, a drive control unit 176, and a registration unit 177.

[0019] The acquisition unit 171 acquires image data generated by the camera 12 capturing images of the target vehicle, and outputs the acquired image data to the estimation unit 172.

[0020] The estimation unit 172 estimates the feature amount of the target vehicle included in the image corresponding to the image data acquired by the acquisition unit 171. Specifically, the estimation unit 172 performs well-known pattern matching processing on the image corresponding to the image data to estimate the feature amount of the vehicle that is the target vehicle included in the image.

[0021] The discrimination unit 173 discriminates the vehicle shape of the target vehicle included in the image data based on the feature amount of the vehicle estimated by the estimation unit 172.

[0022] The verification unit 174 determines whether the vehicle shape of the target vehicle included in the image data discriminated by the discrimination unit 173 can be verified with the existing data of the vehicle information recorded in the vehicle type DB 15, based on the vehicle shape of the target vehicle included in the image data discriminated by the discrimination unit 173 and the vehicle information recorded in the vehicle type DB 15.

[0023] The confirmation unit 175 acquires the detection result detected by the detection device 11 and the image data captured by the camera 12, and confirms the safety status around the target vehicle based on this detection result and image data.

[0024] The drive control unit 176 drives the target vehicle in either the forward direction (warehouse exit direction) or the backward direction via the communication unit 14.

[0025] The registration unit 177 associates the vehicle shape of the target vehicle that could not be verified with the existing data of the vehicle information recorded in the vehicle type DB 15 by the verification unit 174, the vehicle type discrimination result of the target vehicle that could not be verified by the verification unit 174, and the vehicle type information when the discrimination unit 173 used it for vehicle type discrimination, and records them in the vehicle type verification result DB 16.

[0026] 〔Processing of Vehicle Identification Device〕 Next, the processing executed by the vehicle identification device 1 will be described. FIG. 2 is a flowchart showing an overview of the processing executed by the vehicle identification device 1.

[0027] As shown in Figure 2, first, the acquisition unit 171 acquires image data (first image data) from the camera 12 (step S101).

[0028] Next, the estimation unit 172 estimates the feature quantities of the vehicle included in the image corresponding to the image data, based on the image data acquired by the acquisition unit 171 (step S102). Specifically, the estimation unit 172 performs a well-known pattern matching process on the image corresponding to the image data to estimate the feature quantities of the target vehicle included in the image. Here, the feature quantities of the vehicle refer to the shape of the vehicle's exterior and the shapes of the parts that make up the vehicle. For example, vehicle parts include the front windshield, A-pillar, headlights, front grille, and front bumper cover. The estimation unit 172 may also estimate the feature quantities of the vehicle included in the image using a trained model learned by machine learning such as well-known deep learning.

[0029] Next, the discrimination unit 173 determines the vehicle shape of the target vehicle included in the image data based on the vehicle feature quantities estimated by the estimation unit 172 (step S103).

[0030] Subsequently, the matching unit 174 determines, based on the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 and the vehicle information recorded by the vehicle type DB 15, whether the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 can be matched with the existing data of vehicle information recorded by the vehicle type DB 15 (step S104). Specifically, as shown in Figure 3(a), the matching unit 174 determines whether the vehicle shape W1 of the target vehicle O1 included in the image P1 corresponding to the image data determined by the discrimination unit 173 exists in the existing data of vehicle information recorded by the vehicle type DB 15. The matching unit 174 determines that it can be matched if the vehicle shape W1 of the target vehicle O1 included in the image P1 determined by the discrimination unit 173 exists in the existing data of vehicle information recorded by the vehicle type DB 15, and determines that it cannot be matched if it does not exist in the existing data. If the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 by the matching unit 174 is determined to be matchable with existing data of vehicle information recorded by the vehicle type DB 15 (step S104: Yes), the vehicle identification device 1 proceeds to step S105, which will be described later. On the other hand, if the vehicle shape of the target vehicle included in the image corresponding to the image data determined by the discrimination unit 173 by the matching unit 174 is determined to be not matchable with existing data of vehicle information recorded by the vehicle type DB 15 (step S104: No), the vehicle identification device 1 proceeds to step S106, which will be described later.

[0031] In step S105, the discrimination unit 173 determines the vehicle type based on the feature quantities estimated by the estimation unit 172 and the vehicle information recorded by the vehicle type DB 15. Specifically, the discrimination unit 173 determines the vehicle type that matches the feature quantities estimated by the estimation unit 172 from the vehicle information recorded by the vehicle type DB 15. After step S105, the vehicle identification device 1 terminates this process.

[0032] In step S106, the confirmation unit 175 acquires the detection result detected by the detection device 11 and the image data captured by the camera 12, and confirms the safety conditions around the target vehicle based on this detection result and image data. Specifically, the confirmation unit 175 determines whether there are any obstacles or other vehicles within a predetermined distance (for example, within 3m) in the front-rear direction of the target vehicle, based on the detection result detected by the detection device 11 and the image data captured by the camera 12. If there are no obstacles or other vehicles within a predetermined distance (for example, within 3m) in the front-rear direction of the target vehicle, it determines that it is safe. Conversely, if there are obstacles or other vehicles within a predetermined distance (for example, within 3m) in the front-rear direction of the target vehicle, the confirmation unit 175 determines that it is not safe.

[0033] Next, the drive control unit 176 drives the target vehicle either forward (towards exiting the garage) or backward via the communication unit 14 (step S107). Specifically, as shown in Figure 3, the drive control unit 176 drives the target vehicle O1 via the communication unit 14, moving the target vehicle O1 forward (towards exiting the garage) (Figure 3(a) → Figure 3(b)). This makes it possible to clearly see the degree of overlap between the target vehicle O1 and the vehicle O2 adjacent to it.

[0034] Subsequently, the acquisition unit 171 acquires image data (second image data) from the camera 12 (step S108). Specifically, as shown in Figure 3(b), the acquisition unit 171 acquires image P2 corresponding to the image data (second image data) from the camera 12.

[0035] Next, the estimation unit 172 estimates the vehicle features contained in the image corresponding to the image data, based on the image data acquired by the acquisition unit 171 from the camera 12 (step S109).

[0036] Subsequently, the estimation unit 172 determines whether or not the vehicle shape can be estimated as a feature of the vehicle contained in the image corresponding to the image data (step S110). If the estimation unit 172 determines that the vehicle shape can be estimated as a feature of the vehicle contained in the image corresponding to the image data (step S110: Yes), the vehicle identification device 1 proceeds to step S110, which will be described later. On the other hand, if the estimation unit 172 determines that the vehicle shape cannot be estimated as a feature of the vehicle contained in the image corresponding to the image data (step S110: No), the vehicle identification device 1 returns to step S106.

[0037] In step S111, the discrimination unit 173 determines the vehicle shape of the target vehicle included in the image data based on the vehicle feature quantities estimated by the estimation unit 172. Specifically, as shown in Figure 3(b), the discrimination unit 173 estimates the vehicle shape W2 of the target vehicle O1 included in the image P2 based on the vehicle feature quantities estimated by the estimation unit 172.

[0038] Next, the matching unit 174 determines, based on the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 and the vehicle information recorded by the vehicle type DB 15, whether the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 can be matched with existing data of vehicle information recorded by the vehicle type DB 15 (step S112). Specifically, the matching unit 174 determines whether the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 exists in the existing data of vehicle information recorded by the vehicle type DB 15. If the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 exists in the existing data of vehicle information recorded by the vehicle type DB 15, it determines that it can be matched. If it does not exist in the existing data, it determines that it cannot be matched. If the matching unit 174 determines that the vehicle shape of the target vehicle included in the image data determined by the discrimination unit 173 can be matched with existing data of vehicle information recorded by the vehicle type DB 15 (step S112: Yes), the vehicle identification device 1 proceeds to step S113, which will be described later. In contrast, if the vehicle shape of the target vehicle included in the image data identified by the discrimination unit 173 by the matching unit 174 is determined not to be matchable with existing data of vehicle information recorded by the vehicle type DB 15 (step S112: No), the vehicle identification device 1 proceeds to step S114, which will be described later.

[0039] In step S113, the discrimination unit 173 determines the vehicle type based on the feature quantities estimated by the estimation unit 172 and the vehicle information recorded by the vehicle type DB 15. Specifically, the discrimination unit 173 determines the vehicle type that matches the feature quantities estimated by the estimation unit 172 from the vehicle information recorded by the vehicle type DB 15. After step S113, the vehicle identification device 1 terminates this process.

[0040] In step S114, the registration unit 177 records in the vehicle type matching result DB 16 the vehicle shape of the target vehicle that the matching unit 174 could not match with the existing data of vehicle information recorded in the vehicle type DB 15, the vehicle type determination result of the target vehicle that the matching unit 174 could not match, and the vehicle type information used by the determination unit 173 to determine the vehicle type of the target vehicle. As a result, the vehicle identification device 1 can improve the accuracy of matching because the contents of the matching results are updated and stored in the vehicle type matching result DB 16 in real time. After step S114, the vehicle identification device 1 terminates this process.

[0041] According to the embodiment described above, if the vehicle shape of the target vehicle included in the image corresponding to the image data identified by the discrimination unit 173 by the matching unit 174 is determined not to be matchable with existing data of vehicle information recorded by the vehicle type DB 15, the confirmation unit 175 checks the safety conditions around the target vehicle based on the detection result detected by the detection device 11 and the image data captured by the camera 12, and the drive control unit 176 drives the target vehicle forward (towards exiting the parking lot) or backward via the communication unit 14 to move it. As a result, even if the distance between the target vehicle and adjacent vehicles is short, the target vehicle and adjacent vehicles can be detected individually.

[0042] Further effects and modifications can be readily derived by those skilled in the art. Broader aspects of the present invention are not limited to the specific details and representative embodiments expressed and described above. Accordingly, various modifications are possible without departing from the spirit or scope of the overall concept of the invention as defined by the appended claims and their equivalents.

[0043] Although some embodiments of this application have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the present invention. [Explanation of symbols]

[0044] 1. Vehicle identification device 11 Detection device 12 cameras 13 Drive unit 14 Communications Department 15 Vehicle Database 16 Vehicle Matching Results Database 17 Control device 171 Acquisition Department 172 Estimation Department 173 Discrimination part 174 Verification section 175 Verification Section 176 Drive Control Unit 177 Registration Department

Claims

[Claim 1] A vehicle identification device for identifying vehicles within a parking lot, Equipped with a processor, The aforementioned processor, At least the camera acquires first image data of the vehicle, The shape of the vehicle included in the first image data is estimated, Based on a vehicle database that records vehicle information relating to the vehicle shape and an estimation result obtained by estimating the shape of the vehicle based on the first image data, the shape of the vehicle is compared. If the shape of the vehicle cannot be matched, move the vehicle forward or backward. After the vehicle has been moved, the camera acquires a second image data of the vehicle, The shape of the vehicle included in the second image data is estimated, Based on the aforementioned vehicle database and the estimation result obtained by estimating the shape of the vehicle based on the second image data, the shape of the vehicle is estimated. Vehicle identification device.

Images