JP4997191B2 - Device for assisting parking - Google Patents

Description

  The present invention relates to an apparatus for assisting parking, and more specifically, to an apparatus for determining whether a parking space for parking a vehicle is available.

The following Patent Document 1 describes a technique for detecting a parking space where parking can be performed using a millimeter wave radar.
JP 2003-344533 A

  In order for a vehicle to search for a vacant parking space, it is necessary to determine whether or not there is a vehicle in each parking space. In the above technique, this determination is performed using a millimeter wave radar. However, since the millimeter wave radar detects a three-dimensional object, it is difficult to detect a boundary (white line or the like) of a parking space. An apparatus for automatically guiding a vehicle to the parking space based on the detection of the boundary of the parking space has been proposed. When such automatic guidance is taken into consideration, in order to detect the boundary line, An imaging device needs to be provided separately.

  It is also conceivable to detect a three-dimensional object in the parking space using two or more imaging devices. If an imaging device is used, the boundary of a parking space can be detected. However, this method requires two or more imaging devices, which may increase the cost.

  Therefore, there is a demand for an apparatus that can detect the boundary of the parking space and determine whether the parking space is vacant at a reduced cost.

  According to one aspect of the present invention, an apparatus mounted on a vehicle and used to determine whether a parking space for parking the vehicle is vacant is an imaging unit (10) that captures an image of the periphery of the vehicle. A first recognizing means (23) for recognizing the boundary line of the parking space from the captured image, and a second recognizing means for recognizing the license plate of the parked vehicle from the captured image. (25) and whether the parking space having the boundary line is vacant based on whether the license plate is recognized by the second recognition means after the boundary line is recognized by the first recognition means. Determination means (27) for determining

  According to this invention, by recognizing the license plate, it can be more reliably determined whether or not the parking space is vacant. Moreover, it can be determined whether the parking space is vacant with the detection of the boundary line of the parking space. Since the boundary line and the parking space can be recognized from the captured image, only one imaging device is required, and thus the cost can be suppressed.

  According to one embodiment of the present invention, the image processing apparatus further includes third recognition means for recognizing a predetermined feature different from the license plate for the parked vehicle from the captured image. After the boundary line is recognized by the first recognition means, it is determined whether or not the parking space having the boundary line is vacant based on whether or not the feature portion is recognized by the third recognition means.

  According to the present invention, it is possible to determine whether or not a parking space is vacant by recognizing a vehicle feature, for example, an emblem. Since determination is performed using not only the license plate but also the characteristic portion, the accuracy of the determination can be improved.

  Other features and advantages of the present invention will be apparent from the detailed description that follows.

  Next, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of an apparatus for determining whether a parking space is installed in a vehicle and parking for the vehicle according to an embodiment of the present invention.

  The imaging device 10 is a device for imaging the periphery of a vehicle. In this embodiment, the imaging device 10 is attached to a lower part of a door mirror on the left side of the vehicle so as to image the left side of the vehicle toward the top of the vehicle. It is a camera. The imaging device 10 can be realized by, for example, a CMOS camera and a CCD camera.

  Here, referring to FIG. 2A, the vehicle 30 is traveling on the road 31 in the direction of the arrow 33, and parking spaces are arranged on the left side in the traveling direction. . In this figure, three parking spaces Pa to Pc are shown. Vehicles Va and Vc are parked in the parking spaces Pa and Pc, and the parking space Pb is vacant. The white line 35 is a boundary line that separates these parking spaces from the road 31, and the white lines 37a to 37d are boundary lines that separate adjacent parking spaces. The boundary lines of the first parking space Pa are 35, 37a and 37b, the boundary lines of the second parking space Pb are 35, 37b and 37c, and the boundary line of the third parking space Pc is 35 37c and 37d. Hereinafter, the symbol 37 is used when generically referring to the boundary lines that separate the parking spaces, and the symbol P is used when generically referring to parking spaces.

  A region 39 shows an example of a region imaged by the imaging device 10 mounted on the vehicle 30. Referring to FIG. 2 (b), an image captured by the imaging device 10 when the vehicle 30 is positioned as shown in FIG. 2 (a) is shown. In the image, the boundary lines 35 and 37a and a part of the vehicle Va parked in the parking space Pa are imaged.

  Returning to FIG. 1, the control unit 12 can be realized in an electronic control unit (ECU) mounted on a vehicle. The ECU is a computer including a central processing unit (CPU) and a memory. The control unit 12 includes an image processing unit 21, a determination unit 27, and a vehicle guidance unit 29.

  The image processing unit 21 receives an image captured by the imaging device 10. The image processing unit 21 further includes a boundary line recognition unit 23 and a license plate recognition unit 24. In one embodiment, the image processing unit 21 can further include a feature recognition unit 25. The boundary line recognition unit 23 recognizes a boundary line 37 (FIG. 2) that separates the parking spaces in the received image.

  Any method can be used to recognize the boundary line 37. An example of a method for recognizing the boundary line 37 will be described with reference to FIG. In the figure, an example of an image in which a white line 37a (see FIG. 2) is captured is shown. As indicated by an arrow in the image, the captured image is scanned from left to right for each pixel row. In the scanning, a point 41 changing from a non-white pixel to a white pixel in each pixel row is detected, and a point 43 changing from a white pixel to a non-white pixel is detected. Further, the center point 45 of the line segment from the point 41 to the point 43 is calculated. When such processing is performed for all the pixel rows, a line 47 passing through the center point 45 of each pixel row is obtained, which indicates a white line 37a. Thus, if the line 47 indicating the white line 37a is detected, it can be determined that the boundary line has been recognized. In one embodiment, a point 49 where the line 47 intersects with a non-white pixel is detected. A point 49 indicates one end of the entrance of the parking space, and is used for the vehicle guide unit 29 described later.

  In addition, as shown in FIG. 2, when the boundary line 35 which divides between the road 31 and the parking space P is drawn on the road surface, this boundary line 35 is imaged on the image. Therefore, according to the above white line detection method, the boundary line 35 is also detected. The distinction between the boundary line 35 and the boundary line 37 can be made by examining the angle of the detected line 47 on the image. For example, when the X axis and the Y axis with the origin O at the lower left of the image are set for the captured image, as shown in FIG. If the angle with respect to the Y axis is equal to or greater than a predetermined value, it can be determined that the boundary line 35 separates the road and the parking space. Then, the intersection of the detected boundary line 35 and boundary line 37 can be the point 49 described above.

  Although the case where the boundary line is a white line has been described here, the boundary line may be another color such as yellow. For example, detecting a yellow border by detecting points that change from white or yellow pixels to white and non-yellow pixels, and detecting points that change from white and non-yellow pixels to white or yellow pixels. it can.

  Next, the license plate recognition unit 24 in FIG. 1 recognizes the license plate of the parked vehicle in the image received from the imaging device 10. The feature recognition unit 25 recognizes a predetermined feature of the parked vehicle in the image received from the imaging device 10. In this embodiment, the feature is an emblem that identifies the manufacturer of the vehicle.

  Referring now to FIG. 4 (a), a number plate 51 is provided at the center of the vehicle. In a general license plate 51, as shown in FIG. 4B, letters or numbers of a certain size are arranged in each of predetermined areas 51a to 51d. In many cases, an emblem for identifying the manufacturer of the vehicle is attached to a position of the vehicle shown in FIG. An example of the emblem is shown in FIG.

  Since almost all vehicles are provided with a license plate, the presence of the vehicle can be more reliably recognized by recognizing the license plate. The license plate is attached to the center of the vehicle in the vehicle width direction, and letters of a certain size are arranged on the license plate according to a certain arrangement as shown in FIG. It is easy to do.

  As described above, since the license plate is attached at substantially the center in the vehicle width direction of the vehicle, the license plate may be recognized by examining only a predetermined processing area of the captured image. For example, the processing area can be set in advance according to the range of the height from the road surface where the license plate is attached to the vehicle.

  Moreover, since the emblem is attached to almost all vehicles, it can be recognized that the vehicle exists also by recognizing the emblem. In many cases, the emblem is attached almost at the center in the vehicle width direction of the vehicle and has a predetermined shape, so that it is easy to recognize. Even when it is difficult to recognize the license plate for some reason, the presence of the vehicle can be recognized by recognizing the emblem. Therefore, it can be determined more accurately whether a vehicle exists in the parking space.

  Furthermore, the license plate and the emblem are provided at both the front and rear of the vehicle. Therefore, even if the vehicle is parked in any direction, the number plate and emblem of the vehicle can be recognized, so that the presence of the vehicle can be detected more reliably.

  Any appropriate method can be used for license plate recognition. An example of a license plate recognition method will be described with reference to FIG. FIG. 5A shows an example of an image 55 in which the vehicle is imaged. Edge extraction processing is performed on the captured image 55 by a known method. A change in the luminance level on the captured image is large between the characters (including numbers) on the license plate 51 and the background or the vehicle body. Therefore, by extracting the edge, the characters of the license plate are extracted as indicated by reference numeral 57 in FIG.

  The image thus subjected to the edge extraction process is binarized, and a well-known pattern matching for character recognition is applied to the binarized image. A matching process is performed between the character pattern stored in a predetermined dictionary and the characters in the binarized image to recognize the characters displayed on the license plate.

  Such a recognition technique is described in, for example, Japanese Patent Laid-Open No. 63-251898. In addition, for example, the methods disclosed in JP-A-7-105340, JP-A-9-326209, JP-A-10-275299, and the like may be appropriately employed.

  The license plate recognizing unit 24 determines whether or not the characters recognized in this way are in accordance with a predetermined character arrangement for the license plate as shown in FIG. If it is determined, it is determined that the license plate has been recognized. The conditions can include arbitrary conditions such as character size, character position, license plate size (width and / or height), character color, and character background color. For example, whether or not one character of recognized hiragana (in the figure, “A” in area 51b) is arranged to the left of the recognized 4-digit number (in the figure, “12-34” in area 51a). The character recognized according to an arbitrary condition such as whether character strings are arranged in the upper stage (areas 51c and 51d in the figure) and the lower stage (areas 51a and 51a in the figure) is shown in FIG. It can be determined whether the character arrangement as in b) is followed. In addition, it is determined whether the size of the area surrounding the recognized character represents the size of the license plate, or the color of the pixel constituting the recognized character indicates the color used for the license plate. You can judge whether or not.

  The license plate has a special character string arrangement depending on the type of vehicle. As such a special plate, the one shown in FIG. 5C can be considered. Pattern matching can be applied so that these plates can also be recognized. For example, it is possible to recognize a license plate on an image by preparing a special license plate dictionary and performing pattern matching with characters in the dictionary.

  Any appropriate method can be used for the emblem recognition by the feature recognition unit 25 of FIG. 1 described above. The emblem is predetermined according to the manufacturer of the vehicle. Therefore, the emblem can be recognized by applying known pattern matching to the captured image. Note that pattern matching is often performed after a captured image is converted into a binarized image in accordance with the luminance value of each pixel. However, an exterior product such as an emblem has a binary value depending on the color of the vehicle body. It may be difficult to detect from the converted image. Therefore, the captured image may be converted into image data of hue, saturation, and luminance, and pattern matching may be performed on each image data. Such a technique is disclosed in Japanese Patent Application Laid-Open No. 2004-326382.

  Returning to FIG. 1, the determination unit 27 performs the parking based on whether or not the license plate recognition unit 24 recognizes the license plate of the vehicle after the boundary line 37 of the parking space P is recognized. It is determined whether or not the space P is free.

  This determination method will be described with reference to FIG. FIG. 6 shows the same state as FIG. The vehicle 30 is traveling on the road 31 in the direction of the arrow 33, and while traveling, images are sequentially acquired by the imaging device 10, for example, at predetermined time intervals. In each acquired image, it is determined whether the boundary line 37 can be recognized and whether the license plate can be recognized.

  In this example, it is assumed that an image is acquired at each of the positions Q0 to Q5. In the image acquired at the position Q0, the boundary line 37a is recognized. Thereafter, the license plate of the vehicle Va is recognized in the image acquired at the position Q1. Since the license plate is recognized after the boundary line 37a is recognized, the determination unit 27 determines that the parking space Pa having the boundary line 37a is not vacant.

  Next, the boundary line 37b is recognized in the image acquired at the position Q2, and then the license plate is not recognized in the image acquired at the position Q3. In the image acquired at the position Q4, the boundary line 37c is recognized. After the boundary line 37b is recognized, if the license plate is not recognized until the moving distance of the vehicle 30 reaches a predetermined value, the determination unit 27 determines that the parking space Pb having the boundary line 37b is vacant. The travel distance can be calculated based on, for example, the amount of wheel rotation detected by a known wheel pulse sensor.

  Since the predetermined value is a value for determining that no vehicle is present in the parking space P, the length W in the vehicle width direction of the parking space P (this is used for a parking space for a normal passenger car, for example) Is preferably set to a value less than half of the length W (eg, W × 3/4). Alternatively, if the license plate is not recognized before the next boundary line 37c is recognized after the boundary line 37b is recognized, it is determined that the parking space Pb between the boundary lines 37b and 37c is vacant. May be.

  After the boundary line 37c is recognized at the position Q4, the license plate of the vehicle Vc is recognized in the image acquired at the position Q5. Therefore, the determination unit 27 determines that the parking space Pc having the boundary line 37c is not vacant.

  Thus, if the license plate is recognized after recognizing the boundary line, it is determined that the parking space having the boundary line is not vacant, and after the boundary line is recognized, until the moving distance reaches a predetermined value, Alternatively, if the license plate is not recognized before the next boundary line is recognized, it is determined that the parking space having the boundary line is vacant.

  When the feature recognizing unit 25 described above is provided, the availability of the parking space may be determined based on recognition of the feature unit (emblem in this embodiment as described above) in addition to the license plate. For example, if one or both of the license plate and the emblem are recognized after the boundary line is recognized, it is determined that the parking space having the boundary line is not vacant, and the movement distance is predetermined after the boundary line is recognized. If neither the license plate nor the emblem is recognized until the value is reached or the next boundary line is recognized, it is determined that the parking space having the boundary line is vacant.

  In one embodiment, when the determination unit 27 recognizes the boundary line 37, the determination unit 27 detects the position of the intersection of the boundary line 37 and the boundary line 35 that divides the road 31 and the parking space P. In the example of FIG. 6, when the boundary lines 37a to 37c are recognized, the positions of the intersections R1 to R3 are detected. The intersection R2 of the boundary line 37b of the parking space Pb determined to be vacant is used by the vehicle guide unit 29. The positions of these intersections can be detected by the method of detecting the points 49 described above with reference to FIG.

  The vehicle guidance unit 29 in FIG. 1 automatically guides the vehicle to a parking space that is determined to be free by automatic steering control that automatically steers the steering wheel of the vehicle. Vehicle guidance can be realized by any method. Here, an example of a vehicle guidance method will be described with reference to FIG. FIG. 7A shows a method for automatically guiding the vehicle 30 to the parking space Pb determined to be vacant as described with reference to FIG. As described above, the position of the intersection R2 between the boundary lines 35 and 37b is detected.

  Positions A1 to A3 in FIG. 7A are determined as optimum positions with respect to the position of the parking space Pb (in this embodiment, the position of the intersection R2), and according to these optimum positions, FIG. ) Is previously stored in the memory. The movement locus indicates the turning angle of the steered wheel (front wheel) with respect to the moving distance of the vehicle. Therefore, if the vehicle 30 is positioned at the optimum start position A1 as shown in FIG. 7A, the vehicle guiding unit 29 moves the vehicle 30 according to the movement locus set as shown in FIG. 7B. By controlling the steering angle, the vehicle can be parked at the optimum position A3. Here, the movement trajectory includes a forward straight traveling portion having a distance a corresponding to the route 101, a right steering portion having a distance b (running while maintaining a state steered to the right), a forward straight traveling portion having a distance c, a route It consists of a left steering part (traveling while maintaining the state steered to the left) at a distance d corresponding to 102 and a backward rectilinear part at a distance e.

  The vehicle guiding unit 29 determines whether the current position of the vehicle 30 with respect to the intersection R2 detected this time and the optimum start position A1 determined with respect to the detected intersection R2 are as shown in FIG. Thus, the vehicle 30 can be guided to the optimum turn-back position A2 according to the corrected movement track, and then guided to the parking space Pb. For example, if the position of the vehicle 30 is deviated forward by Δa from the optimum start position A1, the forward rectilinear portion a of the movement locus may be shortened by Δa, and if it is deviated rearward by Δa, The forward straight portion a may be extended by Δa. Further, if the position of the vehicle 30 is shifted to the right by Δe from the optimum start position A1, the backward rectilinear portion e of the movement locus may be extended by Δe, and if it is shifted to the left by Δe, the reverse drive The straight part e may be shortened by Δe.

  Alternatively, after the determination unit 27 recognizes the boundary line 37b, if the license plate (and emblem) is not recognized before the next boundary line 37c is recognized, it is determined that the parking space Pb is vacant. In this case, the intersection R3 (FIG. 6) may be used as a reference. In this case, according to the difference between the current position of the vehicle 30 with respect to the detected intersection R3 and the optimum start position A1 determined with respect to the detected intersection R3, as shown in FIG. What is necessary is just to correct a movement locus.

  Moreover, the case where the vehicle 30 inclines with respect to the parking space Pb in the optimal start position A1 may occur. In this case, using the points R2 and R3 corresponding to both ends of the entrance of the parking space Pb, the inclination θx of the line connecting the points R2 and R3 with respect to the central axis in the vehicle length direction of the vehicle 30 is calculated. When the vehicle 30 is tilted to the left (counterclockwise) by θx with respect to the parking space Pb, the right turn amount of the vehicle is increased by the angle θx as shown in FIG. The moving locus may be corrected by extending the right steering portion b shown by Δb. When the vehicle is tilted to the right by the angle θx, the right steering portion b may be shortened by Δb. The vehicle 30 can be automatically guided to the parking space Pb according to the corrected movement locus.

  Details of the method for correcting the movement locus as described above are described in, for example, Japanese Patent Application Laid-Open No. 2001-18821.

  In this embodiment, a description has been given of a mode in which both forward and backward are realized by automatic steering control. However, only forward to the optimum turn-back position is performed by automatic steering control, and the driver steers the reverse from the turn-back position. It may be. In this case, when the driver holds the steering wheel so as to achieve a predetermined turning angle (for example, the left limit turning angle) at the optimum turning position, the desired turning position A2 to the target parking position A3 can be obtained. The vehicle can be guided along the movement trajectory. Such an automatic steering method is described in, for example, Japanese Patent Laid-Open No. 2003-54437.

  Thus, since the boundary of the parking space is recognized at the same time as determining whether the parking space is free, the vehicle can be automatically guided using the recognized boundary line.

  FIG. 8 shows a flow of a process executed by the control unit 12. The start timing of the process can be arbitrarily determined. For example, the process can be activated in response to a predetermined switch provided on the vehicle being operated by a passenger. Once the process is activated, the process is repeatedly executed at predetermined time intervals. Hereinafter, the process will be described by taking the form shown in FIG. 6 as an example.

  In step S11, an image captured by the imaging device 10 is acquired. In step S12, processing for recognizing the boundary line as described above is performed on the acquired captured image. In step S13, it is determined whether a boundary line is recognized.

  As shown in FIG. 6, the boundary line 37a is not recognized until the vehicle reaches the vicinity of the position Q0. Accordingly, the determination in step S13 is No, and the process proceeds to step S15 to check the value of the boundary line flag set to 1 when the boundary line is recognized. Since the boundary line 37a has not been recognized yet, the determination in step S15 is No and the process is exited.

  When the vehicle 30 travels to the vicinity of the position Q0, the determination in step S13 is Yes, and 1 is set in the boundary line flag in step S14. The determination in step S15 is Yes, and the process proceeds to step S16. In step S16, a process for recognizing a license plate as described above is performed on the captured image. In step S17, it is determined whether the license plate is recognized.

  Until the vehicle 30 reaches near the position Q1, the license plate of the vehicle Va is not recognized, and the determination in step S17 is No. In step S21, it is determined whether or not the movement distance of the vehicle 30 after the boundary line flag in step S14 is set to 1 (in the example of FIG. 6, after the boundary line 37a is recognized) is greater than or equal to a predetermined value. As described above, the predetermined value is preferably set to a value smaller than the distance in the vehicle width direction of the parking space P and larger than half of the distance.

  Until the vehicle 30 reaches the vicinity of the position Q1, the determination in step S21 is No, so this process is exited. When the vehicle 30 reaches near the position Q1, the determination in step S17 is Yes. Since this indicates that the vehicle Va is parked in the parking space Pa, it is determined in step S18 that the parking space Pa is not vacant, and in step S19, the boundary flag set to 1 is set to zero. Reset.

  When the vehicle 30 further advances and reaches near the position Q2, the next boundary line 37b is recognized in step S12, the determination in step S13 is Yes, and the boundary line flag is set to 1 in step S14. Thereby, determination of step S15 becomes Yes.

  In step S16, a license plate recognition process is executed. Since the vehicle is not parked in the parking space Pb, even if the vehicle 30 reaches the position Q3, the determination in step S17 is No. It progresses to step S21 and it is judged whether the moving distance from the boundary line 37b of the vehicle 30 is more than predetermined value. Until the vehicle 30 reaches the vicinity of the position Q4, the determination in step S21 is No and the process is exited. When the vehicle 30 reaches the vicinity of the position Q4, the determination in step S21 is Yes, which indicates that the license plate cannot be recognized while the vehicle 30 passes through the entrance of the parking space Pb. Therefore, in step S22, it is determined that the parking space Pb is vacant, and the boundary line flag is reset to zero in step S23.

  Thus, it is determined that the parking space Pb is vacant. As described above, instead of examining the moving distance in step S21, it may be determined that the parking space Pb is vacant when the next boundary line 37c is recognized. When a feature such as an emblem is recognized, if a license plate or emblem is recognized in step S17, the process proceeds to step S18. On the other hand, if a license plate is not recognized and no emblem is recognized in step S17. The process can proceed to step S21.

  In addition, the starting timing of the vehicle guidance process by the vehicle guidance part 29 of FIG. 1 can be determined arbitrarily. For example, vehicle guidance by automatic steering control may be started in response to a determination that a parking space is vacant. Alternatively, vehicle guidance by automatic steering control may be started simultaneously with the activation of the process of FIG. In this case, as shown in FIG. 6, the process of FIG. 8 is executed at predetermined time intervals while the vehicle 30 is advanced in the direction of the arrow 33 by the automatic steering control. If it is determined that the parking space is empty, the vehicle is guided to the optimum turn-back position A2 as described with reference to FIG.

  In the above embodiment, as illustrated in FIG. 2, the case where the parking space is provided on the left side of the vehicle has been described, but the present invention can be similarly applied to the case where the parking space is provided on the right side of the vehicle. It is. In this case, an imaging device is provided in the vehicle to image the right side of the vehicle (eg, can be attached to the lower part of the right door mirror), and the number of vehicles parked after the boundary line is recognized Based on whether the plate (and emblem) is recognized, it can be determined whether the parking space is free. The vehicle can be automatically guided toward the parking space determined to be vacant.

  For example, a parking space is provided on the left side of the vehicle, and a back parking / left mode in which the vehicle is moved backward and parked in a vacant parking space, and a parking space is provided on the right side of the vehicle to reverse the vehicle. However, the vehicle can be provided with a switch for selecting one of the back parking mode and the right mode for parking in a vacant parking space. The vehicle guidance unit 29 extracts a movement locus corresponding to the selected mode from the memory in response to any one of the switches being operated by the occupant, and the movement locus is determined to be parked. Correction may be made based on the position of the space, and the vehicle may be guided according to the corrected movement trajectory.

  As described above, specific embodiments of the present invention have been described. However, the present invention is not limited to these embodiments.

The block diagram of the parking assistance apparatus according to one Example of this invention. The figure for demonstrating the imaging area and captured image by the imaging device of a vehicle according to one Example of this invention. The figure for demonstrating the method of recognizing a boundary line according to one Example of this invention. The figure for demonstrating the number plate and the characteristic part according to one Example of this invention. The figure for demonstrating the method of recognizing a license plate according to one Example of this invention. The figure for demonstrating the method of judging whether the parking space is vacant according to one Example of this invention. The figure for demonstrating vehicle guidance according to one Example of this invention. 4 is a flowchart of a process for determining whether a parking space is available according to one embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Imaging device 12 Control part 21 Image processing part 23 Boundary line recognition part 24 License plate recognition part 25 Feature recognition part 27 Determination part

Claims (2)

A device mounted on a vehicle for determining whether a parking space for parking the vehicle is vacant,
Imaging means for capturing an image around the vehicle,
First recognition means for recognizing a boundary line of a parking space from the captured image;
Second recognition means for recognizing a license plate of a parked vehicle from the captured image;
With the movement to the imaging field of the imaging means, after the boundary is recognized by the first recognition unit, based-out on whether or not the license plate is recognized by the second recognizing means, wherein Determining means for determining whether or not the parking space at a position adjacent to the recognized boundary line is empty;
An apparatus comprising: Furthermore, it comprises a third recognition means for recognizing a predetermined feature different from the license plate for the parked vehicle from the captured image.
The determination unit further determines whether or not the feature unit has been recognized by the third recognition unit after the boundary line has been recognized by the first recognition unit in accordance with the movement to the imaging field of view of the imaging unit. based-out in or to determine whether the parking space at a position adjacent to the recognized boundary lines is free,
The apparatus of claim 1.

Images