Method and apparatus for identifying AVPS marker for positioning based on AVPS marker

Abstract

The present disclosure provides a method for identifying a coded mark, comprising: obtaining a mark image including a coded mark from an input image, and normalizing the mark image to obtain a normalized image; detecting a region of interest in the normalized image; generating a first binarized image by performing binarization to classify each pixel of the normalized image as one of a foreground and a background; generating a second binarized image by dividing the normalized image into a plurality of partitions, binarizing pixel values of each partition to generate a binarized image of each partition, and combining the binarized images of each partition; generating an integrated binarized image according to a result of comparing pixel values at the same position in the first binarized image and the second binarized image; and obtaining identification data of the encoded marker from the integrated binarized image. According to the equipment and the method provided by the invention, the ID of the code mark can be identified more accurately.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of Korean Patent Application No. 10-2023-0142111, filed on October 23, 2023, which is hereby incorporated by reference in its entirety. Technical Field

[0003] The present disclosure relates to a method and apparatus for identifying an AVPS marker for positioning based on the AVPS marker. Background Art

[0004] The descriptions in this section merely provide background information related to the present disclosure and do not necessarily constitute related art.

[0005] In order to facilitate parking, an automated valet parking system (AVPS) is being developed. The AVPS automatically operates a vehicle so that when a driver gets off the vehicle at a drop-off area in a parking facility, the vehicle moves to an empty parking space and parks. In addition, the AVPS automatically moves a parked vehicle to a boarding area upon request by the driver, thereby allowing the driver to board the vehicle.

[0006] For safe and reliable AVPS, Level 4 or higher autonomous driving is required. AVPS must not only identify other objects to prevent collisions, but also determine empty parking spaces and driving routes and automatically move, park, and leave the vehicle. To this end, positioning technology for estimating the current position of the target vehicle is important.

[0007] AVPS adopts vision-based positioning, using high-definition maps and cameras as positioning technology. AVPS uses coded markers specified in ISO 23374 for positioning. Coded markers are a type of fiducial markers and can be installed in parking facilities and recognized by vehicles. Fiducial markers are artificial markers that are objects placed in the field of view of the imaging system and appear in the resulting image, used as reference points or measurements. High-definition maps can include information about the ID, location, orientation, etc. of the coded markers installed in the parking facility. The accuracy of coded marker recognition affects the vehicle positioning performance of the vehicle in AVPS.

[0008] Conventional techniques for detecting fiducial markers include detection methods using ArUco markers, which are typically used for augmented reality (AR), camera pose estimation, camera calibration, etc.

[0009] The detection method using ArUco markers can be achieved by a method using general image processing technology or a detection method using deep learning.

[0010] In the case of general image processing technology, when used in AR, a method is used to detect and identify candidate markers using a general image processing method, and then, the camera pose and the pose of the marker are estimated, and a virtual object is displayed on the marker.

[0011] However, the general image processing method is suitable for simple environments, but is not suitable for vehicles operating in parking lots. There are various environments in parking lots, such as low brightness, reflections due to natural light (e.g., sunlight) or artificial light (e.g., headlights, parking lot lights, etc.), and shadows. Therefore, in the general image processing method, the acquired image is binarized to a constant value using a simple Otsu threshold, so the binarization performance of the acquired image deteriorates in an environment with uneven lighting.

[0012] In addition, the ArUco marker detection method using deep learning generates and learns occluded markers, allowing good recognition of markers even when they are occluded by objects. However, in the case of the method using deep learning, the markers are detected in the form of boxes, so it is difficult to recognize the markers even if they are detected in highly distorted images. Summary of the invention

[0013] One aspect of the present disclosure is to provide a method and apparatus for identifying an AVPS marker for positioning based on the AVPS marker.

[0014] The aspects of the present disclosure are not limited to those described above, and other aspects not mentioned herein will be clearly understood by those skilled in the art from the following description.

[0015] According to at least one embodiment, the present disclosure provides a device comprising at least one processor and a memory operably connected to the at least one processor, wherein the memory stores instructions, the instructions being used to cause the at least one processor to perform operations in response to the at least one processor implementing the instructions, the operations comprising: obtaining a marked image including a coded mark from an input image, and normalizing the marked image to obtain a normalized image; detecting a region of interest in the normalized image; generating a first binary image by performing binarization to classify each pixel of the normalized image as one of a foreground and a background; generating a second binary image by dividing the normalized image into a plurality of partitions, binarizing the pixel values ​​of each partition to generate a binary image of each partition, and combining the binary images of each partition; generating an integrated binary image based on a result of comparing pixel values ​​at the same position in the first binary image and the second binary image; and obtaining identification data of the coded mark from the integrated binary image.

[0016] According to another embodiment, the present disclosure provides a method for identifying a coded mark, which includes: obtaining a marker image including a coded mark from an input image, and normalizing the marker image to obtain a normalized image; detecting a region of interest in the normalized image; generating a first binary image by performing binarization to classify each pixel of the normalized image as one of a foreground and a background; generating a second binary image by dividing the normalized image into a plurality of partitions, binarizing the pixel values ​​of each partition to generate a binary image of each partition, and combining the binary images of each partition; generating an integrated binary image based on a result of comparing pixel values ​​at the same position in the first binary image and the second binary image; and obtaining identification data of the coded mark from the integrated binary image.

[0017] According to the embodiments of the present disclosure, the ID of the coded tag can be identified more accurately.

[0018] Based on the accurately identified coded tag ID, the exact vehicle position can be estimated, thereby improving the accuracy and stability of vehicle position identification.

[0019] The benefits of the present disclosure are not limited to those described above, and other benefits not mentioned herein will be clearly understood by those skilled in the art from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a block diagram showing a configuration of a vehicle positioning system 10 according to an embodiment of the present disclosure.

[0021] Figure 2 It shows the implementation Figure 1 1 is a block diagram of a configuration of a coding mark recognition device 200 of a mark recognition unit 100.

[0022] Figure 3 is a diagram showing an image 310 including a coded mark.

[0023] Figure 4 is a diagram showing a case where a normalized image 410 is generated from an image 310 including a coding mark.

[0024] Figures 5A-5D 410 is a diagram showing a search area for each position in the normalized image 410 .

[0025] Figure 6 is a diagram showing a 2-rectangular harr-like feature of the search area for each location.

[0026] Figure 7 is a diagram showing the form of the detected region of interest.

[0027] Figure 8 is a diagram showing an example of dividing a normalized image into a plurality of partitions.

[0028] Fig. 9 are diagrams showing examples of a first binarized image and a second binarized image, respectively.

[0029] Fig.10 is a diagram showing an integrated binarized image generated from a first binarized image and a second binarized image.

[0030] Fig.11 This is an example diagram used to illustrate the AVPS encoding tag ID decoding method.

[0031] Fig.12 is a diagram showing an integrated binarized image in which AVPS mark data associated with 4×4 data partitions is acquired in various forms.

[0032] Fig.13 is a diagram showing the form of detected AVPS tag data.

[0033] Fig.14 is a diagram showing a decoding algorithm for an AVPS encoded tag ID.

[0034] Fig.15 is a flow chart showing a method for identifying a coding mark according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to the attached illustrative drawings. In the following description, the same reference numerals preferably represent the same elements, although the elements are shown in different drawings. In addition, in the following description of some embodiments, for the purpose of clarity and brevity, detailed descriptions of related known components and functions will be omitted when it is considered to obscure the subject matter of the embodiments of the present disclosure.

[0036] Various ordinal numbers or letter codes (such as first, second, i), ii), a), b), etc.) are only prefixed to distinguish one component from another, but do not imply or suggest the substance, order or sequence of the components. Throughout the specification, when a component "includes" or "comprises" a component, the component is intended to further include other components, and other components are not excluded unless specifically stated to the contrary. Terms such as "unit", "module" and the like refer to one or more units for processing at least one function or operation, which can be implemented by hardware, software or a combination thereof.

[0037] The description of the embodiments of the present disclosure presented below in conjunction with the accompanying drawings is intended to illustrate exemplary embodiments of the present disclosure, and is not intended to represent the only implementation schemes that can practice the technical concepts of the present disclosure.

[0038] Figure 1 is a block diagram showing the configuration of a vehicle positioning system 10 according to an embodiment of the present disclosure. Figure 2 It shows the implementation Figure 1 1 is a block diagram of a configuration of a coding mark recognition device 200 of a mark recognition unit 100.

[0039] In the following, reference will be made to Figure 1 and Figure 2 The operation of the vehicle positioning system 10 and the coded mark recognition device 200 is described.

[0040] Reference Figure 1 , the vehicle positioning system 10 according to an embodiment of the present disclosure obtains the current position of the vehicle. The vehicle positioning system 10 can recognize the coded marker in the image, estimate the position of the coded marker using the coded marker recognition result and the mileage information of the vehicle, and then calculate the current position of the vehicle by performing map matching based on the estimated coded marker position and the AVPS parking facility marking information 30. To this end, the vehicle positioning system 10 includes all or part of the marker recognition unit 100, the mileage extraction unit 120, the marker position estimation unit 140, and the vehicle positioning unit 160.

[0041] The marker recognition unit 100 uses images acquired from the plurality of cameras 20a, 20b, 20c and 20d, which respectively capture the front, left, right and rear of the vehicle, to recognize the coded markers within the parking facility. The marker recognition unit 100 outputs a coded marker recognition result including the ID, bounding box and corner points of the AVPS coded marker in the image.

[0042] To this end, the landmark recognition unit 100 may perform landmark detection, landmark classification, and landmark tracking.

[0043] Landmark detection is the process of simultaneously detecting bounding boxes and corner points of objects belonging to categories corresponding to encoded landmarks within each input image frame using a deep learning network trained through multi-task learning, but is not limited to such.

[0044] Marker classification is the process of extracting the region of interest (ROI) corresponding to the encoded marker in each image frame using the detection results and identifying the encoded marker ID by performing decoding on the ROI.

[0045] Marker tracking is the process of tracking the position changes of a coding marker within a series of image frames. Coded marker tracking can be performed by additionally utilizing the identified coding marker ID information.

[0046] The mileage extraction unit 120 periodically calculates the longitudinal and lateral movement distance of the vehicle over time through wheel mileage and stores it in a buffer. Wheel mileage refers to the use of vehicle specification information and vehicle sensor information to estimate the position change of the vehicle over time. The vehicle specification information may include wheelbase, number of gear teeth, and wheel size. The vehicle sensor information may include wheel speed signals, wheel pulse signals, and steering angle signals received through a chassis controller area network (CAN).

[0047] The marker position estimation unit 140 estimates the position of the vehicle reference coded marker including the distance and direction to the coded marker based on the center of the vehicle. Herein, the term "vehicle reference" refers to the relative position and direction of the coded marker with the center of the vehicle as the origin.

[0048] Using the recognition result of the code mark, the actual size of the code mark and the internal / external parameters of the camera, the distance and direction of the camera reference code mark can be obtained. In this article, the term "camera reference" refers to the relative distance and direction of the code mark with the camera as the origin. Thereafter, the vehicle reference code mark position can be estimated by moving the reference point of the code mark distance and direction to the center of the vehicle using the vehicle specifications and the camera installation position.

[0049] A delay time occurs from time t1 when the marker recognition unit 100 receives the image to time t2 when the coded marker recognition result in the image is output. The longer the delay time, the lower the accuracy of the vehicle reference coded marker position estimation based on the coded marker recognition result due to the movement of the vehicle. Therefore, the marker position estimation unit 140 can use the buffered data stored in the mileage extraction unit 120 to obtain the longitudinal and lateral movement distances of the vehicle between time t1 and time t2, and use the obtained movement distances to correct the estimation result of the vehicle reference coded marker position.

[0050] The vehicle positioning unit 160 obtains the current position of the vehicle within the parking facility based on the coded marker identification information (in other words, information about the ID of the coded marker and the vehicle reference coded marker position) and the AVPS parking facility marker information 30. The vehicle positioning unit 160 may match the ID of the identified coded marker with the marker ID in the AVPS parking facility marker information 30, and may obtain the current position of the vehicle within the parking facility using the vehicle reference coded marker position.

[0051] The coding mark recognition device 200 according to the embodiment of the present disclosure includes all or part of a mark image acquisition unit 210, a ROI detection unit 220, an image binarization unit 230, a partition binarization unit 240, and a coding mark detection unit 250. In this article, the coding mark recognition device 200 can be Figure 2 It may be implemented by some components of the present invention, or may be implemented by further including other components.

[0052] The marker image acquisition unit 210 acquires an image including a coded marker, ie, a coded marker image, from an input image, and normalizes the coded marker image to acquire a normalized image.

[0053] Herein, the original coding mark image may be an image including a coding mark with a certain degree of distortion. When the image including the coding mark with distortion is normalized, the distortion may be removed.

[0054] Figure 3 is a diagram showing an image 310 including a coded mark. Figure 4 is a diagram showing a case where a normalized image 410 is generated from an image 310 including a coding mark.

[0055] The marker image acquisition unit 210 receives images obtained from each of the cameras 20a, 20b, 20c and 20d, and detects four corner points 311, 312, 313 and 314 related to the AVPS encoded marker from the input image, generates a bounding box from the four corner points 311, 312, 313 and 314, and acquires the area within the bounding box as the original encoded marker image.

[0056] The marker image acquisition unit 210 can obtain a square normalized image 410 by performing a warp on the image 310 including the original coding marker. The marker image acquisition unit 210 obtains a homography matrix representing the relationship between the four corner points 311, 312, 313, and 314 of the normalized image 410 and the corner point coordinates 411, 412, 413, and 414. Thereafter, the normalized image 410 can be generated by performing a bilinear interpolation on the image 310 including the original coding marker to generate an image including the coding marker without distortion. In this article, when the normalization process is not performed, the image 310 including the coding marker can be considered as the normalized image 410.

[0057] In this embodiment, technical details about the method of acquiring the normalized image 410 from the image 310 including the encoding mark are beyond the subject matter of the present disclosure, and thus further detailed description thereof will be omitted.

[0058] The ROI detection unit 220 detects a ROI of the normalized image 410. The ROI detection unit 220 detects a region in the normalized image 410 where the AVPS encoded marker is determined to be located as the ROI.

[0059] Herein, the ROI detection unit 220 may generate an integrated image of the normalized image 410 and detect the ROI using the integrated image. In the following description, the normalized image 410 may be the original normalized image 410 before the integrated image is generated, or may be an integrated image of the normalized image 410.

[0060] Figures 5A-5D 410 is a diagram showing a search area for each position in the normalized image 410 . Figure 6 is a diagram showing a 2-rectangular harr-like feature of the search area for each location.

[0061] The ROI detection unit 220 sets the left 1 / 3 region of the normalized image 410 as follows Figure 5A The left search area 501 shown in FIG. 4 sets the upper 1 / 3 area of ​​the normalized image 410 to be as follows: Figure 5B The upper search area 502 shown in FIG. 4 sets the right 1 / 3 area of ​​the normalized image 410 to be as follows: Figure 5C The right search area 503 shown in FIG. 5 and the lower 1 / 3 area of ​​the normalized image 510 is set as Figure 5D The lower search area 504 is shown.

[0062] The ROI detection unit 220 performs each harr-like feature search using a first harr-like feature 601 for the left search area 501 , a second harr-like feature 602 for the upper search area 502 , a third harr-like feature 603 for the right search area 503 , and a fourth harr-like feature 604 for the lower search area 504 .

[0063] The ROI detection unit 220 uses the first harr-like feature 601 to search the left search area 501 to obtain the left coordinate (left) of the white area within the left search area 501, uses the second harr-like feature 602 to search the upper search area 502 to obtain the upper coordinate (upper) of the white area within the upper search area 502, uses the third harr-like feature 603 to search the right search area 503 to obtain the right coordinate (right) of the white area within the right search area 503, and uses the fourth harr-like feature 604 to search the lower search area 504 to obtain the lower coordinate (lower) of the white area within the lower search area 504.

[0064] Figure 7is a diagram showing the form of the detected region of interest.

[0065] Assume that the horizontal axis of the normalized image 410 is the x-axis, and the vertical axis is the y-axis. Figure 7 As shown, the ROI detection unit 220 detects the ROI characterized by the left and upper coordinates (left, upper) and the right and lower coordinates (right, lower), in other words, estimates the area where the AVPS encoding mark exists. In this article, the ROI is described as being characterized by the left and upper coordinates (left, upper) and the right and lower coordinates (right, lower), but is not limited to this. According to this embodiment, the ROI can be characterized by the left and lower coordinates (left, lower) and the right and upper coordinates (right, upper). Therefore, the ROI can be characterized by the edges facing each other.

[0066] The image binarization unit 230 generates a first binarized image by performing binarization to classify each pixel of the normalized image into one of the foreground and the background. In this embodiment, the term "binarization" refers to an operation of classifying each pixel of the target image into one of the foreground and the background based on a reference value.

[0067] The image binarization unit 230 obtains a first median value as a reference value, and performs binarization on each pixel value of the normalized image according to whether each pixel value of the normalized image is greater than or equal to the first median value, wherein the first median value is the median value of pixels of the ROI in the normalized image.

[0068] In other words, the image binarization unit 230 generates a first binarized image by performing binarization to classify pixel values ​​of the normalized image greater than or equal to the first median value as foreground and pixel values ​​not greater than or equal to the first median value as background.

[0069] In this embodiment, when binarization is performed, pixels corresponding to the foreground may have their values ​​set to 255, and pixels corresponding to the background may have their values ​​set to 0, but embodiments of the present disclosure are not limited thereto.

[0070] The partition binarization unit 240 divides the normalized image into a plurality of partitions, performs binarization on a pixel value of each partition, generates a binarized image for each partition, and combines the binarized images of each partition to generate a second binarized image.

[0071] Figure 8 is a diagram showing an example of dividing a normalized image into a plurality of partitions.

[0072] like Figure 8 As shown, the normalized image can be divided into six partitions 801 , 802 , 803 , 804 , 805 and 806 .

[0073] exist Figure 8, the normalized image is divided into a plurality of partitions on the left and right sides, but the division may have various forms, such as a grid-type division.

[0074] The partition binarization unit 240 performs binarization on each partition 801 , 802 , 803 , 804 , 805 , or 806 .

[0075] The partition binarization unit 240 acquires, for each partition 801 , 802 , 803 , 804 , 805 , or 806 , each second median value that is a median value of boundary pixel values ​​of the corresponding partition 801 , 802 , 803 , 804 , 805 , and 806 as each reference value.

[0076] The partition binarization unit 240 binarizes the pixel value of each partition according to whether the pixel value of each partition 801, 802, 803, 804, 805 or 806 is greater than or equal to the second median value, and obtains a binarized image of each partition. In other words, the partition binarization unit 240 performs binarization by classifying pixels greater than or equal to each second median value among the pixel values ​​of each partition 801, 802, 803, 804, 805 or 806 as foregrounds, and classifying pixels not greater than or equal to each second median value as backgrounds.

[0077] The partition binarization unit 240 generates a second binarized image by combining the binarized images of each partition 801 , 802 , 803 , 804 , 805 , or 806 binarized using the corresponding median value at the corresponding position.

[0078] Fig. 9 are diagrams showing examples of a first binarized image and a second binarized image, respectively.

[0079] like Fig. 9 As shown, a first binarized image 901 and a second binarized image 902 having forms different from each other may be generated.

[0080] The encoding mark detection unit 250 generates an integrated binarized image according to the result of comparing the pixel values ​​of the first binarized image 901 with the pixel values ​​of the second binarized image 902, and obtains AVPS mark data from the integrated binarized image.

[0081] The encoding mark detection unit 250 generates an integrated binarized image according to the result of comparing the pixel values ​​at the same position in the first binarized image 901 and the second binarized image 902 .

[0082] The coding mark detection unit 250 determines at least one first pixel that is the foreground in the first binary image 901 and the foreground in the second binary image 902 as the foreground of the integrated binary image. In addition, the coding mark detection unit 250 determines at least one second pixel that is the background in the first binary image 901 and the background in the second binary image 902 as the background of the integrated binary image.

[0083] The encoding mark detection unit 250 performs binarization using at least one first pixel and at least one second pixel for at least one third pixel at a position that is neither the first pixel nor the second pixel in the first binarized image 901 and the second binarized image 902 .

[0084] The coding mark detection unit 250 calculates a first average value, which is an average value of pixels of the normalized image corresponding to the position of the at least one first pixel, and calculates a second average value, which is an average value of pixels of the normalized image corresponding to the position of the at least one second pixel. The coding mark detection unit 250 performs binarization on at least one third pixel using the first average value and the second average value to obtain an integrated binarized image.

[0085] The coding mark detection unit 250 calculates a third median value as a reference value, the third median value being a median value between the first average value and the second average value, and performs binarization on at least one third pixel based on the third median value to obtain an integrated binarized image. In other words, the coding mark detection unit 250 performs binarization to classify those greater than or equal to the third median value in at least one third pixel value as foreground, and those not greater than or equal to the third median value in at least one third pixel value as background.

[0086] The coding mark detection unit 250 sets at least one first pixel as a foreground, sets at least one second pixel as a background, and combines pixels of at least one third pixel classified as the foreground and the background into one to obtain an integrated binary image.

[0087] Fig.10 is a diagram showing an integrated binarized image 1001 generated from the first binarized image 901 and the second binarized image 902 .

[0088] like Fig.10 As shown, the integrated binarized image 1001 has reduced noise compared to the first binarized image 901 and the second binarized image 902 and is therefore a more reliable binarized image.

[0089] Fig.11 This is an example diagram used to illustrate the AVPS encoding tag ID decoding method.

[0090] like Fig.11 As shown, the encoding mark detection unit 250 divides the integrated binary image 1101 into 6×6 partitions of equal size, and in the 6×6 divided partitions, acquires 1-bit data for each of 16 areas corresponding to the 4×4 partitions 1102 (excluding the partitions that touch the boundaries of the integrated binary image 1101). Thus, a total of 16 bits of AVPS mark data are obtained.

[0091] For example, when the integrated binary image 1101 is formed of 48×48 pixels, the 4×4 data partitions 1102 except for the partitions contacting the boundaries of the integrated binary image 1101 are formed of 32×32 pixels, and each partition area is formed of 8×8 pixels.

[0092] The coding mark detection unit 250 divides each of the 16 data partitions into four equal parts in the form of 2×2 to generate four sub-partitions. When all of the four sub-partitions are white, the data in the corresponding data partition is distinguished as 1. When any of the four sub-partitions is black, the data in the corresponding data partition is distinguished as 0. Regarding distinguishing whether a sub-partition is white or black, when a preset number of pixels is white, the sub-partition can be distinguished as white, and when a preset number of pixels is white, the sub-partition can be distinguished as black. For example, when the integrated binary image 1101 is formed by 48×48 pixels, each sub-partition has 4×4 pixels.

[0093] Fig.12 is a diagram showing an integrated binarized image in which AVPS mark data associated with 4×4 data partitions is acquired in various forms.

[0094] When acquiring the orientation information from the AVPS tag data, the encoding tag detection unit 250 checks the direction of the AVPS tag data from the orientation information. In other words, the four orientation bits are used as the orientation information indicating the direction of the integrated binary image 1101.

[0095] In the AVPS mark data, the data located at the four edges correspond to the orientation bit. When the encoding mark detection unit 250 determines that three of the four orientation bits are black and one is white, it interprets the orientation bit as normal. When the encoding mark detection unit 250 determines that there are 2 or more or 0 white edges among the four edges, it is interpreted as an orientation bit abnormality, and the acquired AVPS mark data is determined to have no valid direction.

[0096] like Fig.12 As shown, the AVPS encoded marker can be detected when located in various orientations.

[0097] exist Fig.12In the AVPS encoded marks in the normalized image shown at 1210, 1220, 1230 and 1240 in FIG. 1 , the AVPS mark is detected in the normal direction 1210 when the white orientation bit 1201 is located at the left edge and the top edge.

[0098] Therefore, when the AVPS encoding mark is not detected in the normal direction 1220, 1230 or 1240, the encoding mark detection unit 250 rotates the AVPS encoding mark so that the white orientation bit 901 is located at the left edge and the top edge, and interprets the AVPS mark data after converting the AVPS mark data into an AVPS mark corresponding to the normal direction 1210.

[0099] Fig.13 is a diagram showing the format of detected AVPS tag data.

[0100] when Fig.13 When the color of each region in the 4×4 partition is black, the encoded mark detection unit 250 interprets the bit value of the region as 0, and when the color of each region is white, interprets the bit value of the region as 1.

[0101] exist Fig.13 In this case, the orientation bit 1201 in the left edge and the top edge is 1, and the orientation bits corresponding to the remaining edges are all 0, so it can be interpreted as AVPS tag data with valid orientation.

[0102] When the detected AVPS mark data is determined to have a valid orientation, the encoding mark detection unit 250 performs a parity check using the parity bits.

[0103] exist Fig.13 In the example, p00, p01, p02, and p03 correspond to parity bits. In addition, d00, d01, d02, d03, d04, d05, d06, and d07 correspond to data bits.

[0104] The encoding mark detection unit 250 performs a first parity check to determine whether the last bit value of the result value of the binary operation of d00+d01+d03+d04+d06 is equal to p00, performs a second parity check to determine whether the last bit value of the result value of the binary operation of d00+d02+d03+d05+d06 is equal to p01, performs a third parity check to determine whether the last bit value of the result value of the binary operation of d01+d02+d03+d07 is equal to p02, and performs a fourth parity check to determine whether the last bit value of the result value of the binary operation of d04+d05+d06+d07 is equal to p03.

[0105] When there is no abnormality as a result of performing the first to fourth parity checks, the encoding mark detection unit 250 determines that the detected AVPS mark data is valid data.

[0106] When determining that valid AVPS tag data has been acquired, the encoding tag detection unit 250 decodes the AVPS encoding tag ID using the data bits.

[0107] Fig.14 is a diagram showing the decoding algorithm of the AVPS encoded tag ID.

[0108] like Fig.14 As shown in the algorithm, the AVPS encoding tag ID is determined based on whether the values ​​of d00 to d07 are 11 or 0 respectively.

[0109] exist Fig.14 , the finally calculated m_candidate.ID represents the AVPS encoding tag ID, m_bitCode.m_bBits[d] represents the bit values ​​of d00 to d07, and d represents the index values ​​0 to 7 corresponding to d00 to d07, respectively.

[0110] Fig.15 is a flow chart showing a method for identifying a coding mark according to an embodiment of the present disclosure.

[0111] The coding mark recognition method according to an embodiment of the present disclosure is performed by the coding mark recognition device 200 .

[0112] The marker image acquisition unit 210 performs a marker image acquisition process to acquire a marker image including a coding marker from an input image, and normalizes the marker image to acquire a normalized image ( S1510 ).

[0113] The ROI detection unit 220 performs a ROI detection process to detect a ROI of the normalized image ( S1520 ).

[0114] The image binarization unit 230 performs an image binarization process by performing binarization to classify each pixel of the normalized image into one of a foreground and a background to generate a first binarized image ( S1530 ).

[0115] The partition binarization unit 240 performs a partition binarization process to divide the normalized image into a plurality of partitions, binarizes the pixel value of each partition, generates a binarized image for each partition, and combines the binarized images of each partition to generate a second binarized image (S1540).

[0116] The encoding marker detecting unit 250 compares pixel values ​​at the same position in the first binarized image and the second binarized image (S1550).

[0117] The coding mark detection unit 250 determines at least one first pixel corresponding to the foreground in the first binary image and the second binary image as the foreground of the integrated binary image, and determines at least one second pixel corresponding to the background in the first binary image and the second binary image as the background of the integrated binary image (S1560).

[0118] The encoding marker detecting unit 250 determines at least one third pixel in a position that is neither the first pixel nor the second pixel in the first binarized image and the second binarized image (S1570).

[0119] For at least one third pixel, the encoding mark detection unit 250 calculates a first average value, which is the average value of pixels of the normalized image corresponding to the position of at least one first pixel, and calculates a second average value, which is the average value of pixels of the normalized image corresponding to the position of at least one second pixel (S1580).

[0120] The encoding mark detection unit 250 calculates a third median value as a reference value, the third median value being a median value between the first average value and the second average value, and performs binarization on at least one third pixel based on the third median value to obtain an integrated binarized image (S1590).

[0121] At least some of the components described in the exemplary embodiments of the present disclosure may be implemented as hardware elements, including at least one or a combination of a digital signal processor (DSP), a processor, a controller, an application specific integrated circuit (ASIC), a programmable logic device (FPGA, etc.), and other electronic devices. In addition, at least some of the functions or processes described in the exemplary embodiments may be implemented as software, and the software may be stored in a recording medium. At least some of the components, functions, and processes described in the exemplary embodiments of the present disclosure may be implemented by a combination of hardware and software.

[0122] The method according to the exemplary embodiment of the present disclosure may be written as a program that can be executed on a computer, and may also be implemented in various recording media such as a magnetic storage medium, an optical reading medium, and a digital storage medium.

[0123] The embodiments of the various techniques described herein may be implemented by digital electronic circuits or by computer hardware, firmware, software, or a combination thereof. The embodiments may be implemented as a computer program tangibly embodied in a computer program product (i.e., an information carrier, such as a machine-readable storage device (computer-readable medium) or a radio signal) for processing or controlling the operation of a data processing device by a data processing device (e.g., a programmable processor, a computer, or multiple computers). Computer programs such as the above-mentioned computer programs may be written in any form of programming language, including compiled or interpreted languages, and may be deployed in any form as stand-alone programs or as modules, components, subroutines, or other units suitable for a computing environment. The computer program may be processed on one computer or at one site or on multiple computers distributed on multiple sites, and is developed to be interconnected through a communication network.

[0124] As examples, processors suitable for processing computer programs include general and special purpose microprocessors, and any one or more processors of any type of digital computer. Typically, the processor will receive instructions and data from a read-only memory or a random access memory or both. The elements of a computer may include at least one processor that executes instructions and one or more memory devices that store instructions and data. Typically, a computer may include one or more mass storage devices that store data, such as magnetic disks, magneto-optical disks, or optical disks, or may be coupled to a mass storage device to receive data from it and / or send data to it. Information carriers suitable for embodying computer program instructions and data include, for example, semiconductor memory devices, magnetic media (such as hard disks, floppy disks, and tapes), optical media (such as CD-ROMs (Compact Disc Read Only Memory), DVDs (Digital Video Disks)), magneto-optical media (such as optical magnetic floppy disks), ROMs (Read Only Memory), RAMs (Random Access Memory), flash memory, EPROMs (Erasable Programmable ROMs), and EEPROMs (Electrically Erasable Programmable ROMs). The processor and memory may be supplemented by or included in a dedicated logic circuit.

[0125] The processor may execute an operating system and software applications executed on the operating system. In addition, the processor device may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processor device may be described as being used as a single processor device, but those skilled in the art will appreciate that the processor device may include multiple processing elements and / or multiple types of processing elements. For example, the processor device may include multiple processors or a processor and a controller. In addition, other processing configurations (such as parallel processors) are also possible.

[0126] Additionally, non-transitory computer readable media may be any available media that can be accessed by a computer and may include computer storage media and transmission media.

[0127] This specification includes the details of many specific embodiments, but it should be understood that these details do not limit any invention or claimed content in this specification, but describe the features of specific exemplary embodiments. The features described in the context of each exemplary embodiment can be implemented as a combination in a single exemplary embodiment. On the contrary, the various features described in the context of a single exemplary embodiment can be implemented in multiple exemplary embodiments individually or in appropriate sub-combinations. In addition, features can operate in a specific combination and can be initially described as being claimed in the combination, but in some cases, one or more features can be excluded from the claimed combination, and the claimed combination can be changed to a sub-combination or a modification of the sub-combination.

[0128] Similarly, even if operations are described in a particular order on the accompanying drawings, it should not be understood that the operations need to be performed in a particular order or sequence to obtain the desired results or that all operations need to be performed. In certain circumstances, multitasking and parallel processing may be advantageous. In addition, it should not be understood as requiring the various device components in the above exemplary embodiments to be separated in all exemplary embodiments, and it should be understood that the above program components and devices can be incorporated into a single software product, or can be packaged in multiple software products.

[0129] It should be understood that the exemplary embodiments disclosed herein are illustrative only and are not intended to limit the scope of the invention. It will be apparent to those skilled in the art that various modifications may be made to the exemplary embodiments without departing from the spirit and scope of the claims and their equivalents.

[0130] Thus, it would be appreciated by those of ordinary skill that the scope of the claimed invention is not limited by the embodiments explicitly described above, but rather by the claims and their equivalents.

Claims

1. A device, characterized in that: include: at least one processor; as well as A memory operatively connected to the at least one processor, wherein the memory stores instructions for causing the at least one processor to perform operations in response to the at least one processor executing the instructions, the operations comprising: Obtaining a labeled image including encoded labels from an input image, and normalizing the labeled image to obtain a normalized image; detecting a region of interest in the normalized image; generating a first binarized image by performing binarization to classify each pixel of the normalized image into one of a foreground and a background; generating a second binarized image by dividing the normalized image into a plurality of partitions, binarizing a pixel value of each partition to generate a binarized image of each partition, and combining the binarized images of each partition; generating an integrated binary image according to a result of comparing pixel values ​​at the same position in the first binary image and the second binary image; and The identification data of the coded mark is obtained from the integrated binarized image.

2. The device according to claim 1, characterized in that The detecting the region of interest comprises: performing a Harr-like feature search on a left 1 / 3 region, an upper 1 / 3 region, a right 1 / 3 region and a lower 1 / 3 region of the normalized image respectively.

3. The device according to claim 1, characterized in that The generating the first binarized image comprises: determining a first median value, the first median value being a median value of pixels of the region of interest in the normalized image; and Binarization is performed on each pixel value of the normalized image based on the first median value.

4. The device according to claim 1, characterized in that Generating the second binarized image comprises: determining a second median value for each of the partitions, the second median value being a median value of boundary pixels of each of the partitions; and The binarized image of each partition is generated by binarizing the pixel values ​​of each partition based on the second median value.

5. The device according to claim 1, characterized in that The step of generating the integrated binary image comprises: determining that at least one first pixel corresponding to a foreground in the first binarized image and the second binarized image is a foreground of the integrated binarized image; and It is determined that at least one second pixel corresponding to the background in the first binarized image and the second binarized image is the background of the integrated binarized image.

6. The device according to claim 5, characterized in that The generating the integrated binarized image includes: using the at least one first pixel and the at least one second pixel to perform the binarization on at least one third pixel in the first binarized image and the second binarized image that is neither the first pixel nor the second pixel.

7. The device according to claim 6, characterized in that The step of generating the integrated binary image comprises: determining a first average value, the first average value being an average value of pixels in the normalized image corresponding to the position of the at least one first pixel; determining a second average value, the second average value being an average value of pixels in the normalized image corresponding to the position of the at least one second pixel; and Binarization is performed on the at least one third pixel using the first average value and the second average value.

8. The device according to claim 7, characterized in that The step of generating the integrated binary image comprises: determining a third median value, the third median value being a median value between the first mean value and the second mean value; and Binarization is performed on the at least one third pixel based on the third median value.

9. The device according to claim 1, characterized in that The obtaining of the identification data of the coding mark comprises: Dividing the integrated binary image into 6×6 partitions of equal size; and Each 4×4 partition of the 6×6 partitions that is not on a boundary of the integrated binary image is decoded.

10. The device according to claim 1, characterized in that The operations also include: The current position of the vehicle in the parking facility is obtained based on the identification data of the coded tag and the parking facility marking information of the automatic valet parking system (AVPS) by matching the identification data of the coded tag with the tag ID in the AVPS parking facility marking information.

11. A computer-implemented method for identifying a coded mark, the method comprising: Obtaining, by at least one processor, a labeled image including a coded label from an input image, and normalizing the labeled image to obtain a normalized image; detecting, by the at least one processor, a region of interest in the normalized image; generating, by the at least one processor, a first binarized image by performing binarization to classify each pixel of the normalized image as one of a foreground and a background; generating, by the at least one processor, a second binarized image by dividing the normalized image into a plurality of partitions, binarizing pixel values ​​of each partition to generate a binarized image of each partition, and combining the binarized images of each partition; generating, by the at least one processor, an integrated binary image according to a result of comparing pixel values ​​at the same position in the first binary image and the second binary image; as well as The identification data of the encoded mark is obtained from the integrated binarized image by the at least one processor.

12. The method according to claim 11, characterized in that The detecting the region of interest comprises: performing a Harr-like feature search on a left 1 / 3 region, an upper 1 / 3 region, a right 1 / 3 region and a lower 1 / 3 region of the normalized image respectively.

13. The method according to claim 11, characterized in that The generating the first binarized image comprises: determining a first median value, the first median value being a median value of pixels of the region of interest in the normalized image; and Binarization is performed on each pixel value of the normalized image based on the first median value.

14. The method according to claim 11, characterized in that Generating the second binarized image comprises: determining a second median value for each of the partitions, the second median value being a median value of boundary pixels of each of the partitions; and The binarized image of each partition is generated by binarizing the pixel values ​​of each partition based on the second median value.

15. The method according to claim 11, characterized in that The step of generating the integrated binary image comprises: determining that at least one first pixel corresponding to a foreground in the first binarized image and the second binarized image is a foreground of the integrated binarized image; and It is determined that at least one second pixel corresponding to the background in the first binarized image and the second binarized image is the background of the integrated binarized image.

16. The method according to claim 15, characterized in that The generating the integrated binarized image includes: using the at least one first pixel and the at least one second pixel to perform the binarization on at least one third pixel in the first binarized image and the second binarized image that is neither the first pixel nor the second pixel.

17. The method according to claim 16, characterized in that The step of generating the integrated binary image comprises: determining a first average value, the first average value being an average value of pixels in the normalized image corresponding to the position of the at least one first pixel; determining a second average value, the second average value being an average value of pixels in the normalized image corresponding to the position of the at least one second pixel; and Binarization is performed on the at least one third pixel using the first average value and the second average value.

18. The method according to claim 17, characterized in that The step of generating the integrated binary image comprises: determining a third median value, the third median value being a median value between the first mean value and the second mean value; and Binarization is performed on the at least one third pixel based on the third median value.

19. The method according to claim 11, characterized in that The obtaining of the identification data of the coding mark comprises: Dividing the integrated binary image into 6×6 partitions of equal size; and Each 4×4 partition of the 6×6 partitions that is not on a boundary of the integrated binary image is decoded.

20. The method according to claim 11, characterized in that Also includes: The current position of the vehicle in the parking facility is obtained based on the identification data of the coded tag and the parking facility marking information of the automatic valet parking system (AVPS) by matching the identification data of the coded tag with the tag ID in the AVPS parking facility marking information.

Images