A method for determining a storage location number, a vehicle, and a computer-readable storage medium

Abstract

The embodiment of the present invention discloses a method for determining a storage location number, a vehicle, and a computer-readable storage medium, which are used to improve the accuracy of determining a target storage location number by a vehicle. The embodiment of the present invention comprises: obtaining a storage location map corresponding to a parking space and the number of single characters corresponding to the storage location number in the storage location map; segmenting the storage location map to obtain a segmentation map of the number of single characters; inputting the segmentation map of the number of single characters into a preset character recognition model to obtain a single character corresponding to each segmentation map; and determining the target storage location number corresponding to the parking space according to the single character corresponding to each segmentation map.

Description

Technical Field

[0001] The present invention relates to the field of vehicle applications, and in particular to a method for determining a storage location number, a vehicle, and a computer-readable storage medium. Background Art

[0002] When a vehicle identifies a parking space number, an optical character recognition (OCR) algorithm is usually used to identify the parking space number corresponding to the parking space in the garage. However, the OCR algorithm has defects such as poor recognition accuracy and / or insufficient scene generalization ability, resulting in a low accuracy of the parking space number finally obtained by the vehicle. Summary of the invention

[0003] The embodiment of the present invention provides a method for determining a storage location number, a vehicle, and a computer-readable storage medium, which are used to improve the accuracy of determining a target storage location number by a vehicle.

[0004] A first aspect of an embodiment of the present invention provides a method for determining a storage location number, which may include:

[0005] Obtain the parking space map corresponding to the parking space and the number of single characters corresponding to the parking space number in the parking space map;

[0006] Segment the library location map to obtain a segmentation map of the number of single characters;

[0007] Input the segmentation graphs of the number of single characters into a preset character recognition model to obtain the single characters corresponding to each segmentation graph;

[0008] According to the single character corresponding to each segmentation graph, the target storage location number corresponding to the parking space is determined.

[0009] In the embodiment of the present invention, when the vehicle is identifying the storage location number, each character corresponding to the storage location number in the storage location map can be identified one by one to determine the target storage location number corresponding to these characters. This method can improve the accuracy of the vehicle in determining the target storage location number.

[0010] Optionally, obtaining the number of single characters corresponding to the storage location number in the storage location map includes: binarizing the storage location map to obtain a binary image; scanning the column pixels of the binary image to determine the first number of times the sum of the first target column pixels is zero and the sum of the second target column pixels is not zero, and the first target column pixels are adjacent to the second target column pixels; determining the number of single characters corresponding to the storage location number in the storage location map based on the first number of times; or scanning the row pixels of the binary image to determine the second number of times the sum of the first target row pixels is zero and the sum of the second target row pixels is not zero, and the first target row pixels are adjacent to the second target row pixels; determining the number of single characters corresponding to the storage location number in the storage location map based on the second number of times.

[0011] In an embodiment of the present invention, the vehicle can determine the number of single characters corresponding to the storage location number in the storage location map based on the number of times the sum of the pixels of the first target column is zero and the sum of the pixels of the adjacent second target column is not zero, that is, there are as many single characters as there are times, thereby enabling the vehicle to accurately determine the number of single characters.

[0012] Optionally, the binarization process is performed on the storage location map to obtain a binarized image, including: binarization process is performed on the storage location map to obtain a first image; obtaining a closed area in the first image and the area of ​​each closed area; in the storage location map, blacking out closed areas whose areas are smaller than a preset area threshold to obtain a binarized image

[0013] In the embodiment of the present invention, since the vehicle processes the storage location map in a relatively simple manner, there may be black spots in the white single characters, but these black spots do not affect the determination of the number of single characters, while there may be white spots in the background outside the single characters, which may affect the determination of the number of single characters, that is, the white spots in the binarized image interfere with the determination of the number of single characters, which is not conducive to the subsequent segmentation of the storage location map. Therefore, the vehicle can obtain the closed area in the first image after the binarization process, and set the closed area with a smaller area to black to reduce the interference of the first image, and obtain a binarized image with higher clarity, which is conducive to the subsequent segmentation of the storage location map.

[0014] Optionally, the binarization processing is performed on the storage location map to obtain a binary image, including: performing a first preprocessing on the storage location map to obtain a preprocessed image, the first preprocessing including scaling according to a first preset size, and / or filtering; and binarization processing is performed on the preprocessed image to obtain a binary image.

[0015] In this embodiment, the vehicle scales the storage location map according to a first preset size to facilitate subsequent segmentation of the storage location map by the vehicle; the vehicle filters the storage location map to filter out image impurities in the storage location map to obtain a pre-processed image with higher definition.

[0016] Optionally, the segmentation map of the number of single characters is input into a preset character recognition model to obtain a single character corresponding to each segmentation map, including: performing data training using a deep learning processing method based on a public data set of preset parking spaces and / or a proprietary data set collected for storage location number scenarios to obtain a convolutional neural network (CNN) classifier; inputting the segmentation map of the number of single characters into the CNN classifier to output a single character corresponding to each segmentation map.

[0017] In the embodiment of the present invention, the vehicle recognizes the segmentation images one by one using a preset character recognition model, and can relatively accurately obtain the single character corresponding to each segmentation image.

[0018] Optionally, the number of the single characters is at least two, and obtaining the target storage location number corresponding to the parking space based on the single characters corresponding to each segmentation diagram includes: splicing the single characters corresponding to each segmentation diagram in a preset order to obtain the target storage location number corresponding to the parking space.

[0019] In the embodiment of the present invention, since the single characters corresponding to each segmentation diagram are relatively accurate, the target storage location number obtained by the vehicle according to the single characters corresponding to these segmentation diagrams is also relatively accurate.

[0020] Optionally, the method further includes: obtaining a preset map; and determining the current position of the vehicle according to the target storage location number and the preset map.

[0021] In an embodiment of the present invention, the vehicle is positioned according to the position of the target storage location number on a preset map to improve the accuracy of the vehicle in determining the current position of the vehicle.

[0022] A second aspect of an embodiment of the present invention provides a vehicle, which may include:

[0023] An acquisition module, used to acquire a parking location map corresponding to the parking space and the number of single characters corresponding to the parking location number in the parking location map;

[0024] The processing module is used to segment the parking space map to obtain the segmentation map of the single character number; input the segmentation map of the single character number into a preset character recognition model to obtain the single character corresponding to each segmentation map; and determine the target parking space number corresponding to the parking space according to the single character corresponding to each segmentation map.

[0025] Optionally, the processing module is specifically used to binarize the storage location map to obtain a binary image; scan the column pixels of the binary image to determine the first number of times that the sum of the first target column pixels is zero and the sum of the second target column pixels is not zero, and the first target column pixels are adjacent to the second target column pixels; determine the number of single characters corresponding to the storage location number in the storage location map based on the first number of times; or scan the row pixels of the binary image to determine the second number of times that the sum of the first target row pixels is zero and the sum of the second target row pixels is not zero, and the first target row pixels are adjacent to the second target row pixels; determine the number of single characters corresponding to the storage location number in the storage location map based on the second number of times.

[0026] Optionally, the processing module is specifically used to perform binarization processing on the storage location map to obtain a first image;

[0027] The acquisition module is specifically used to acquire the closed regions in the first image and the area of ​​each closed region;

[0028] The processing module is specifically used to blacken the closed areas whose areas are smaller than a preset area threshold in the storage location map to obtain a binary image.

[0029] Optionally, the processing module is specifically used to perform a first preprocessing on the storage location map to obtain a preprocessed image, wherein the first preprocessing includes scaling according to a first preset size, and / or filtering; and binarizing the preprocessed image to obtain a binarized image.

[0030] Optionally, the processing module is specifically used to perform data training using a deep learning processing method based on a public data set of preset parking spaces and / or a proprietary data set collected for parking space number scenarios to obtain a convolutional neural network (CNN) classifier; the segmentation map of the number of single characters is input into the CNN classifier, and a single character corresponding to each segmentation map is output.

[0031] Optionally, the processing module is specifically used for splicing the single characters corresponding to each segmentation image in a preset order when the number of the single characters is at least two, so as to obtain the target storage location number corresponding to the parking space.

[0032] Optionally, the acquisition module is specifically used to acquire a preset map;

[0033] The processing module is specifically used to determine the current position of the vehicle according to the target storage location number and the preset map.

[0034] A third aspect of an embodiment of the present invention provides a vehicle, which may include:

[0035] A memory storing executable program code;

[0036] and a processor coupled to the memory;

[0037] The processor calls the executable program code stored in the memory, and when the executable program code is executed by the processor, the processor implements the method as described in the first aspect of the embodiment of the present invention.

[0038] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having executable program code stored thereon. When the executable program code is executed by a processor, the method described in the first aspect of the embodiment of the present invention is implemented.

[0039] A fifth aspect of an embodiment of the present invention discloses a computer program product. When the computer program product runs on a computer, the computer is enabled to execute any one of the methods disclosed in the first aspect of the embodiment of the present invention.

[0040] A sixth aspect of an embodiment of the present invention discloses an application publishing platform, which is used to publish a computer program product. When the computer program product runs on a computer, the computer executes any one of the methods disclosed in the first aspect of the embodiment of the present invention.

[0041] It can be seen from the above technical solutions that the embodiments of the present invention have the following advantages:

[0042] In an embodiment of the present invention, a parking location map corresponding to a parking space and the number of single characters corresponding to the location number in the parking location map are obtained; the parking location map is segmented to obtain a segmentation map of the number of single characters; the segmentation map of the number of single characters is input into a preset character recognition model to obtain a single character corresponding to each segmentation map; and the target location number corresponding to the parking space is determined based on the single character corresponding to each segmentation map. That is, during the parking process, each character corresponding to the location number in the parking location map can be identified one by one to determine the target location number corresponding to these characters. This method can improve the accuracy of the vehicle in determining the target location number. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments and the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and other drawings can be obtained based on these drawings.

[0044] Figure 1a A schematic diagram of an embodiment of a method for determining a storage location number in an embodiment of the present invention;

[0045] Figure 1b A schematic diagram of an embodiment of a storage location map in an embodiment of the present invention;

[0046] Figure 1c It is another embodiment schematic diagram of the storage location map in the embodiment of the present invention;

[0047] Figure 1d A schematic diagram of an embodiment of a segmentation diagram of the number of single characters in an embodiment of the present invention;

[0048] Figure 1e It is a schematic diagram of another embodiment of the method for determining the storage location number in an embodiment of the present invention;

[0049] Figure 2 It is a schematic diagram of another embodiment of the method for determining the storage location number in an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of a vehicle in an embodiment of the present invention;

[0051] Figure 4 FIG. 2 is a schematic diagram of another embodiment of a vehicle in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The embodiment of the present invention provides a method for determining a storage location number, a vehicle, and a computer-readable storage medium, which are used to improve the accuracy of determining a target storage location number by a vehicle.

[0053] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiments in the present invention, they should all fall within the scope of protection of the present invention.

[0054] The technical solution of the present invention is further described below by way of embodiments. Figure 1a FIG. 1 is a schematic diagram of an embodiment of a method for determining a storage location number in an embodiment of the present invention, which may include:

[0055] 101. Obtain a parking location map corresponding to a parking space and the number of single characters corresponding to the parking location number in the parking location map.

[0056] It should be noted that a parking space is a parking space, which refers to the location where a vehicle is parked.

[0057] The single character corresponding to the storage location number can be any number from 0, 1, ..., 9, any letter from A, B, ..., Z, or any other special symbol (for example: -), which is not specifically limited here. Among them, at least two single characters can constitute a storage location number, and multiple single characters in the storage location number can be the same or different, which is not specifically limited here. For example, the storage location number is A32 and the storage location number is A33.

[0058] Optionally, the single character can also be any letter from a, b, ..., z.

[0059] It should be noted that, in actual situations, the maximum number of single characters corresponding to the storage location number is 7.

[0060] It is understandable that if the number of single characters is 0, the vehicle outputs the first prompt information (for example: NULL / null) to prompt that the parking space has no location number or the vehicle has not recognized the location number in the location map. At the same time, the vehicle does not perform any processing on the location map and searches for a parking space again.

[0061] It should be noted that the storage location map can be obtained by taking pictures with a camera installed on the vehicle. The storage location map can include but is not limited to the storage location number. For example, Figure 1b As shown, it is a schematic diagram of an embodiment of a storage location map in an embodiment of the present invention; or, Figure 1c FIG. 1 is a schematic diagram of another embodiment of a storage location map according to an embodiment of the present invention. Figure 1bIn the storage location map, the number of single characters corresponding to the storage location number is 3, and the storage location number in the storage location map is 218.

[0062] Optionally, the vehicle obtains the number of single characters corresponding to the storage location number in the storage location map, which may include but is not limited to the following implementation methods:

[0063] Implementation method 1: The vehicle performs binarization processing on the storage location map to obtain a binary image; the vehicle scans the column pixels of the binary image to determine the first number of times the sum of the first target column pixels is zero and the sum of the second target column pixels is not zero, and the first target column pixels are adjacent to the second target column pixels; the vehicle determines the number of single characters corresponding to the storage location number in the storage location map based on the first number of times.

[0064] Among them, the preset area threshold can be set before the vehicle leaves the factory, or can be customized by the user, and is not specifically limited here.

[0065] It should be noted that the binarization process refers to the vehicle setting the grayscale value corresponding to the pixel point in the storage location map to 0 or 255, that is, the vehicle presents the entire storage location map in black and white. Among them, when the grayscale value corresponding to the pixel point is set to 0, the pixel point is black; when the grayscale value corresponding to the pixel point is set to 255, the pixel point is white. It can be understood that in the binary image, the single character is white and the background part outside the single character is black. However, since the vehicle's processing of the storage location map is relatively simple, there may be black spots in the white single character, and there may be white spots in the background part outside the single character.

[0066] Optionally, the vehicle performs binarization processing on the storage location map to obtain a binary image, including: the vehicle performs binarization processing on the storage location map to obtain a first image; the vehicle obtains closed areas in the first image, and the area of ​​each closed area; the vehicle sets black the closed areas in the storage location map whose areas are smaller than a preset area threshold to obtain a binary image.

[0067] It should be noted that the preset area threshold may be set before the vehicle leaves the factory or may be user-defined, and is not specifically limited here.

[0068] It is understandable that, since the vehicle processes the storage location map in a relatively simple manner, there may be black dots in the white single characters, but these black dots do not affect the determination of the number of single characters, while there may be white dots in the background outside the single characters, which may affect the determination of the number of single characters, that is, the white dots in the binarized image interfere with the determination of the number of single characters, which is not conducive to the subsequent segmentation of the storage location map. Therefore, the vehicle can obtain the closed area (for example, white dots) in the first image after binarization, and set the white dots with smaller areas to black to reduce the interference of the first image, and obtain a binarized image with higher clarity, which is conducive to the subsequent segmentation of the storage location map.

[0069] It should be noted that the vehicle scans the column pixels of the binary image, which means that the vehicle can scan the row pixels of the entire column of the first target column pixels of the binary image, or scan some continuous row pixels of the first target column pixels of the binary image. Exemplary: the binary image is an image of 100 pixels*300 pixels, that is, 100 pixels are the width of the binary image, 300 pixels are the length of the binary image, and the first target column pixels are the 20th column pixels. In the process of scanning the column pixels of the binary image, the vehicle can scan the pixels of the 1st row to the 100th row of the 20th column pixel point, and calculate the sum of the 100 row pixels in the 20th column; or scan some continuous row pixels of the 20th column pixel point, and calculate the sum of the continuous row pixels in the lower 20 columns. For example: the vehicle scans the pixels of the 1st row to the 50th row of the 20th column pixel point of the binary image, and calculates the sum of the 50 row pixels in the 20th column.

[0070] Optionally, the vehicle scans some continuous rows of pixels of the first target column of the binary image, which means that the vehicle can divide the first target column of pixels into N segments for scanning, where N is an integer greater than or equal to 1. Wherein, the sum of the n segments of column pixels in the first target column of pixels is 0, and the sum of the column pixels corresponding to the n segments of column pixels in the second target column of pixels is not 0, and the first count can be determined to determine the number of single characters corresponding to the storage location number in the storage location map, where n is an integer greater than or equal to 1, and n≤N.

[0071] For example, the vehicle scans the pixels in the 1st row to the 50th row of the 20th column of the binary image, scans the pixels in the 21st row to the 70th row of the 20th column of the binary image, and scans the pixels in the 51st row to the 100th row of the 20th column of the binary image, that is, the vehicle scans the three types of column pixels of the binary image. If the vehicle calculates that the sum of the pixels in the 1st row to the 50th row in the 20th column is 0, and the sum of the pixels in the 1st row to the 50th row in the 21st column is not 0, then the first count is recorded as 1. If the vehicle calculates that the sum of the pixels from the 1st row to the 50th row in the 20th column is not 0, the sum of the pixels from the 21st row to the 70th row in the 20th column is 0 and the sum of the pixels from the 21st row to the 70th row in the 21st column is not 0, the sum of the pixels from the 51st row to the 100th row in the 20th column is 0 and the sum of the pixels from the 51st row to the 100th row in the 21st column is not 0, the first number can also be recorded as 1. Similarly, the vehicle scans each column of pixels in the binary image, and as long as there is a section of row pixels in a certain column of pixels that satisfies the first target column pixel sum of zero and the second target column pixel sum is not zero, the first number can be recorded, regardless of the number of segments of the row pixels in this certain column of pixels. That is, there are as many single characters as there are times, so that the vehicle can accurately determine the number of single characters.

[0072] It can be understood that in the binary image, if there is a closed area with an area greater than the preset area threshold, then the vehicle determines the distance between the closed area and any single character during the column pixel scanning or row pixel scanning process, and determines the single character with the shortest distance to the closed area. Then, the number of columns of the first target column pixels of the closed area and the number of columns of the first target column pixels of the single character with the shortest distance are determined, and the size of these two column numbers is determined, and the pixel corresponding to the column number with the smaller value is taken as the first target column pixel of the single character.

[0073] Exemplarily, assuming that a closed area with an area greater than a preset area threshold is closest to the first single character, if it is determined that the number of columns of the first target column pixels of the closed area is 10, and the number of columns of the first target column pixels of the first single character is 20, and 10 is less than 20, then the 10th column is determined to be the first target column pixels of the first single character; if it is determined that the number of columns of the first target column pixels of the closed area is 40, and the number of columns of the first target column pixels of the first single character is 20, and 20 is less than 40, then the 20th column is determined to be the first target column pixels of the first single character.

[0074] Optionally, the vehicle performs binarization processing on the storage location map to obtain a binary image, which may include: the vehicle performs a first preprocessing on the storage location map to obtain a preprocessed image, and the first preprocessing may include: scaling according to a first preset size, and / or filtering; binarizing the preprocessed image to obtain a binary image.

[0075] The first preset size scaling may be set before the vehicle leaves the factory, or may be user-defined, and is not specifically limited here.

[0076] It can be understood that the vehicle scales the storage location map according to the first preset size to facilitate the subsequent segmentation of the storage location map by the vehicle; the vehicle filters the storage location map to filter out image impurities in the storage location map to obtain a pre-processed image with higher definition.

[0077] Implementation method 2: The vehicle performs binarization processing on the storage location map to obtain a binary image; the vehicle scans the row pixels of the binary image to determine the second number of times that the sum of the first target row pixels is zero and the sum of the second target row pixels is not zero, and the first target row pixels are adjacent to the second target row pixels; the vehicle determines the number of single characters corresponding to the storage location number in the storage location map based on the second number.

[0078] It is understandable that the direction of the binarized image's storage location number in implementation 2 is different from that in implementation 1. The explanation of implementation 2 is similar to that of implementation 1, and will not be described in detail here.

[0079] Optionally, the vehicle obtains the parking space map corresponding to the parking space, which may include but is not limited to the following implementation methods:

[0080] Implementation method 1: The vehicle obtains an image to be processed corresponding to the parking space; the vehicle performs a second preprocessing on the image to be processed to obtain a second image; and a storage location map is determined, the storage location map including the storage location area in the second image. The second preprocessing may include but is not limited to at least one of the following: scaling according to a second preset size, filtering, contrast enhancement, and overexposure.

[0081] It should be noted that the image to be processed may be any one of several panoramic bird's-eye views, which include a storage location map. The panoramic bird's-eye view is captured by a camera installed on a vehicle.

[0082] It should be noted that the explanation of the second preset size is similar to that of the first preset size, and will not be elaborated here.

[0083] It is understandable that since the image to be processed corresponding to the parking space may have at least one of the following conditions, which may include but are not limited to low brightness, poor integrity, and poor clarity, the vehicle can obtain a parking space map with higher image quality after performing the second preprocessing on the image to be processed, that is, a parking space map with normal brightness, high integrity, and high clarity can be obtained, which is convenient for the vehicle to perform subsequent segmentation and recognition processing on the parking space map.

[0084] Implementation method 2: The vehicle obtains the image to be processed corresponding to the parking space; the vehicle performs a third preprocessing on the image to be processed to obtain a third image; and a storage location map is determined, the storage location map including the storage location area in the third image. The third preprocessing may include but is not limited to at least one of the following: cropping according to a third preset size, grayscale processing, image edge detection processing, image corrosion and closure, image fine-grained area removal, image slope correction, and image flipping back to normal processing.

[0085] It should be noted that the explanation of the third preset size is similar to that of the first preset size, and will not be elaborated here.

[0086] It is understandable that since the third image may include not only the storage location number but also the vehicle, this will increase the difficulty for the vehicle to identify the storage location number. Therefore, the vehicle can perform a third preprocessing on the third image to obtain a storage location map with higher image quality, thereby reducing the difficulty for the vehicle to identify the storage location number.

[0087] It should be noted that the above-mentioned implementation mode 1 and implementation mode 2 can also be combined with each other to form a new implementation mode, which is also within the protection scope of the present invention and will not be described in detail here.

[0088] 102. Segment the storage location map to obtain a segmentation map of the number of single characters.

[0089] It is understandable that in the vehicle's storage location map, after determining how many single characters the storage location number corresponds to, the storage location map can be segmented into that many segmented maps, so that each segmented map corresponds to one character, so that the vehicle can subsequently identify the segmented maps one by one.

[0090] It should be noted that if the number of the single character is one, then the entire image corresponding to the single character can be used as a segmentation image. For example, if the entire image corresponding to the single character is an image of 100 pixels * 300 pixels, the segmentation starts from the first column of pixels and ends at the 300th column of pixels, thereby obtaining a 100*300 segmentation image.

[0091] It should be noted that if there are multiple single characters, then the library map is evenly divided according to the number of single characters, or the segmentation starts from the first target column pixel of the first single character and ends at the first target column pixel of the adjacent single character. It can be understood that the difference between the column numbers corresponding to the two first target column pixels is the area image width corresponding to the first single character in the library map.

[0092] It is understandable that if the last single character has no subsequent adjacent single character, the above-mentioned regional image width can be used as the regional image width corresponding to the last single character, and the starting edge of the regional image is the first column of pixels of the last single character in this library bit map.

[0093] For example, Figure 1d , which is a schematic diagram of an embodiment of a segmentation diagram of the number of single characters in an embodiment of the present invention.

[0094] 103. Input the segmentation graphs of the number of single characters into a preset character recognition model to obtain a single character corresponding to each segmentation graph.

[0095] Optionally, the vehicle inputs the segmentation map of the number of single characters into a preset character recognition model to obtain a single character corresponding to each segmentation map, which may include: the vehicle performs data training using a deep learning processing method based on a public data set of preset parking spaces and / or a proprietary data set collected for storage location number scenarios to obtain a Convolutional Neural Networks (CNN) classifier; the segmentation map of the number of single characters is input into the CNN classifier, and the single character corresponding to each segmentation map is output.

[0096] It should be noted that the vehicle can use a pre-trained character recognition model (ie, a preset character recognition model) to recognize the segmentation images one by one, and can relatively accurately obtain the single character corresponding to each segmentation image.

[0097] It is understood that the deep learning processing method, also known as deep learning technology, is a deep learning algorithm. The CNN classifier is a vehicle that can identify each segmentation map as an element in the definition domain N (for example: "p") to obtain a single character corresponding to each segmentation map, where N is an integer greater than or equal to 1.

[0098] It should be noted that since the value range of a single character corresponding to each segmentation map is fixed, that is, each single character has 37 possibilities, the vehicle can be trained to obtain a 37-class CNN classifier. The 37 possibilities include numbers 0, 1..., 9, and letters A, B,..., Z. Optionally, the letters A, B,..., Z can be replaced by a, b,..., z.

[0099] It is understandable that, usually, the single character value corresponding to the storage location number is generally 0, 1..., 9, and A, B,..., H.

[0100] 104. Determine a target parking location number corresponding to the parking space according to the single character corresponding to each segmentation graph.

[0101] Optionally, the number of single characters is at least two, and the vehicle determines the target storage location number corresponding to the parking space based on the single characters corresponding to each segmentation diagram, which may include: the vehicle splices the single characters corresponding to each segmentation diagram in a preset order to obtain the target storage location number corresponding to the parking space.

[0102] It can be understood that the CNN classifier in the vehicle can identify each single character as an element within the definition domain N, which can also be called a single string (for example: "p_1",..., "p_C"). The vehicle then concatenates these elements in a preset order to obtain a long string (for example: "p_1...p_C"), and finally determines this long string as the target storage location number corresponding to the parking space.

[0103] For example, Figure 1e FIG. 2 is a schematic diagram of another embodiment of the method for determining the storage location number in an embodiment of the present invention.

[0104] It should be noted that, since the single characters corresponding to each segmentation diagram are relatively accurate, the target storage location numbers obtained by the vehicle based on the single characters corresponding to these segmentation diagrams are also relatively accurate.

[0105] Optionally, after step 104, the method may further include: the vehicle obtains a preset map; and the vehicle determines the current position of the vehicle according to the target storage location number and the preset map.

[0106] It is understandable that the vehicle can be positioned according to the position of the target storage location number on the preset map to improve the accuracy of the vehicle in determining the current position of the vehicle.

[0107] It should be noted that the vehicle can be parked in the parking space corresponding to the target storage location number, thereby improving the accuracy of the vehicle parking in the target parking space corresponding to the target storage location number.

[0108] Optionally, the vehicle may determine the current location of the vehicle according to the target storage location number and the preset map in the following implementations, including but not limited to:

[0109] Implementation method 1: If there is a preset storage location number on the preset map, the target storage location number is matched with the preset storage location number; if the matching is successful, the current position of the vehicle is determined.

[0110] Optionally, if a preset storage location number exists on the preset map for the vehicle, the target storage location number is matched with the preset storage location number; when the match is successful, the current location of the vehicle is determined, which may include: if a preset storage location number exists on the preset map for the vehicle, the target storage location number is matched with the preset storage location number to obtain a matching degree; when the matching degree is greater than a preset matching degree threshold, the current location of the vehicle is determined.

[0111] It is understandable that the preset matching degree threshold may be set before the vehicle leaves the factory or may be user-defined, and is not specifically limited here.

[0112] Exemplarily, the preset matching degree threshold is 95%. The vehicle matches the target storage location number with the preset storage location number, and obtains a matching degree of 100%, which is greater than the preset matching degree threshold of 95%. At this time, the vehicle determines the current position of the vehicle.

[0113] It should be noted that the target storage location number and the preset storage location number have a high degree of match, so that the vehicle can accurately determine the current position of the vehicle, thereby improving the initial positioning and real-time positioning capabilities of the vehicle on the map.

[0114] Optionally, if the target storage location number fails to match the preset storage location number on the map, a parking space is searched again.

[0115] It can be understood that when the matching degree of the target storage location number and the preset storage location number is less than or equal to the preset matching degree threshold, it means that the target storage location number fails to match the preset storage location number on the map.

[0116] Implementation method 2: If the preset storage location number does not exist on the preset map, the target storage location number is marked in the corresponding storage location number of the preset map, and the current position of the vehicle is determined.

[0117] In an embodiment of the present invention, a parking location map corresponding to a parking space and the number of single characters corresponding to the location number in the parking location map are obtained; the parking location map is segmented to obtain a segmentation map of the number of single characters; the segmentation map of the number of single characters is input into a preset character recognition model to obtain a single character corresponding to each segmentation map; and the target location number corresponding to the parking space is determined based on the single character corresponding to each segmentation map. That is, during the parking process, each character corresponding to the location number in the parking location map can be identified one by one to determine the target location number corresponding to these characters. This method can improve the accuracy of the vehicle in determining the target location number.

[0118] like Figure 2 FIG. 1 is a schematic diagram of another embodiment of a method for determining a storage location number in an embodiment of the present invention, which may include:

[0119] 201. Obtain a parking location map corresponding to a parking space and the number of single characters corresponding to the parking location number in the parking location map.

[0120] 202. Segment the storage location map to obtain a segmentation map of the number of single characters.

[0121] 203. Input the segmentation graphs of the number of single characters into a preset character recognition model to obtain a single character corresponding to each segmentation graph.

[0122] 204. Determine a target parking location number corresponding to the parking space according to the single character corresponding to each segmentation graph.

[0123] It should be noted that steps 201-204 are similar to steps 101-104 shown in FIG. 1 in this embodiment, and will not be described in detail here.

[0124] 205. Output the target parking space number corresponding to the parking space.

[0125] Optionally, the vehicle outputs the target parking space number corresponding to the parking space, which may include but is not limited to the following implementations:

[0126] Implementation method 1: The vehicle outputs the target parking space number corresponding to the parking space in a first form, and the first form may include but is not limited to at least one of the following: in the form of text on a display screen, in the form of animation on the display screen, and in the form of voice broadcast through a speaker.

[0127] It is understandable that the form of text, animation and voice broadcast can be set before the vehicle leaves the factory or customized by the user, and is not specifically limited here.

[0128] Implementation method 2: The vehicle sends the target parking space number corresponding to the parking space to the terminal device associated with the vehicle.

[0129] It is understandable that after receiving the target storage location number corresponding to the parking space, the terminal device associated with the vehicle can output the target storage location number corresponding to the parking space in the second form. The explanation of the second form is similar to the explanation of the first form, and will not be described in detail here.

[0130] It should be noted that the terminal devices associated with the vehicle may include general handheld screen electronic terminal devices, such as mobile phones, smart phones, portable terminals, terminals, personal digital assistants (PDA), portable multimedia players (Personal Media Player, PMP) devices, laptop computers, notebooks (NotePad), wireless broadband (Wireless Broadband, Wibro) terminals, tablet computers (Personal Computer, PC), smart PCs, sales terminals (Point of Sales, POS) and vehicle-mounted computers, etc.

[0131] The terminal devices associated with the vehicle may also include wearable devices. Wearable devices are portable electronic devices that can be worn directly on the user or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but can also realize powerful intelligent functions through software support, data interaction, and cloud interaction, such as: computing function, positioning function, and alarm function. They can also be connected to mobile phones and various terminals. Wearable devices may include but are not limited to watches supported by the wrist (such as watches, wrists, etc.), shoes supported by the feet (such as shoes, socks or other products worn on the legs), Glass supported by the head (such as glasses, helmets, headbands, etc.), and various non-mainstream product forms such as smart clothing, school bags, crutches, accessories, etc.

[0132] It is understandable that, whether it is implementation method 1 or implementation method 2, both allow users to more intuitively and quickly obtain the storage location number to confirm whether it matches the storage location number on the map.

[0133] In an embodiment of the present invention, a parking location map corresponding to a parking space and the number of single characters corresponding to the location number in the parking location map are obtained; the parking location map is segmented to obtain a segmentation map of the number of single characters; the segmentation map of the number of single characters is input into a preset character recognition model to obtain a single character corresponding to each segmentation map; the target location number corresponding to the parking space is determined based on the single characters corresponding to each segmentation map; and the target location number corresponding to the parking space is output. That is, during the parking process, each character corresponding to the location number in the parking location map can be identified one by one to determine the target location number corresponding to these characters and output them. This method can not only improve the accuracy of the vehicle in determining the target location number, but also allow the user to obtain the location number more intuitively and quickly.

[0134] like Figure 3 As shown, it is a schematic diagram of an embodiment of a vehicle in an embodiment of the present invention, which may include: an acquisition module 301 and a processing module 302;

[0135] The acquisition module 301 is used to acquire a parking space map corresponding to the parking space and the number of single characters corresponding to the parking space number in the parking space map;

[0136] The processing module 302 is used to segment the parking space map to obtain the segmented map of the single character number; identify the segmented map of the single character number to obtain the single character corresponding to each segmented map; and determine the target parking space number corresponding to the parking space according to the single character corresponding to each segmented map.

[0137] Optionally, in some embodiments of the present invention,

[0138] The processing module 302 is specifically used to perform binarization processing on the storage location map to obtain a binary image; scan the column pixels of the binary image to determine the first number of times that the sum of the first target column pixels is zero and the sum of the second target column pixels is not zero, and the first target column pixels are adjacent to the second target column pixels; based on the first number, determine the number of single characters corresponding to the storage location number in the storage location map; or scan the row pixels of the binary image to determine the second number of times that the sum of the first target row pixels is zero and the sum of the second target row pixels is not zero, and the first target row pixels are adjacent to the second target row pixels; based on the second number, determine the number of single characters corresponding to the storage location number in the storage location map.

[0139] Optionally, in some embodiments of the present invention,

[0140] The processing module 302 is specifically used to perform binarization processing on the storage location map to obtain a first image;

[0141] The acquisition module 301 is specifically used to acquire the closed regions in the first image and the area of ​​each closed region;

[0142] The processing module 302 is specifically used to blacken the closed areas whose areas are smaller than a preset area threshold in the storage location map to obtain a binary image.

[0143] Optionally, in some embodiments of the present invention,

[0144] The processing module 302 is specifically used to perform a first preprocessing on the storage location map to obtain a preprocessed image, wherein the first preprocessing includes scaling according to a first preset size and / or filtering; and binarizing the preprocessed image to obtain a binarized image.

[0145] Optionally, in some embodiments of the present invention,

[0146] The processing module 302 is specifically used to perform data training using a deep learning processing method based on a public data set of preset parking spaces and / or a proprietary data set collected for a parking space number scenario, to obtain a convolutional neural network (CNN) classifier; the segmentation map of the number of single characters is input into the CNN classifier, and the single character corresponding to each segmentation map is output.

[0147] Optionally, in some embodiments of the present invention,

[0148] The processing module 302 is specifically used for splicing the single characters corresponding to each segmentation image according to a preset order when the number of the single characters is at least two, so as to obtain the parking space number corresponding to the parking space.

[0149] Optionally, in some embodiments of the present invention,

[0150] The acquisition module 301 is specifically used to acquire a preset map;

[0151] The processing module 302 is specifically used to determine the current position of the vehicle according to the target storage location number and the preset map.

[0152] like Figure 4 As shown, it is a schematic diagram of another embodiment of a vehicle in an embodiment of the present invention, which may include: a memory 401 and a processor 402; the memory 401 and the processor 402 are coupled, and the processor 402 calls the executable program code stored in the memory 401;

[0153] In the embodiment of the present invention, the processor 402 has the following functions:

[0154] Obtain the parking space map corresponding to the parking space and the number of single characters corresponding to the parking space number in the parking space map;

[0155] Segment the library location map to obtain a segmentation map of the number of single characters;

[0156] Input the segmentation graphs of the number of single characters into a preset character recognition model to obtain the single characters corresponding to each segmentation graph;

[0157] According to the single character corresponding to each segmentation graph, the target storage location number corresponding to the parking space is determined.

[0158] Optionally, the processor 402 further has the following functions:

[0159] The storage location map is binarized to obtain a binary image; the column pixels of the binary image are scanned to determine the first number of times when the sum of the first target column pixels is zero and the sum of the second target column pixels is not zero, and the first target column pixels are adjacent to the second target column pixels; according to the first number, the number of single characters corresponding to the storage location number in the storage location map is determined; or, the row pixels of the binary image are scanned to determine the second number of times when the sum of the first target row pixels is zero and the sum of the second target row pixels is not zero, and the first target row pixels are adjacent to the second target row pixels; according to the second number, the number of single characters corresponding to the storage location number in the storage location map is determined.

[0160] Optionally, the processor 402 further has the following functions:

[0161] The storage location map is binarized to obtain a first image; closed areas in the first image and the area of ​​each closed area are obtained; closed areas in the storage location map whose areas are smaller than a preset area threshold are blackened to obtain a binary image.

[0162] Optionally, the processor 402 further has the following functions:

[0163] The first preprocessing is performed on the storage location map to obtain a preprocessed image, wherein the first preprocessing includes scaling according to a first preset size and / or filtering; and the preprocessed image is binarized to obtain a binarized image.

[0164] Optionally, the processor 402 further has the following functions:

[0165] According to the public data set of preset parking spaces and / or the proprietary data set collected for the parking space number scenario, a deep learning processing method is used to perform data training to obtain a convolutional neural network (CNN) classifier; the segmentation map of the single-character quantity is input into the CNN classifier, and the single character corresponding to each segmentation map is output.

[0166] Optionally, the processor 402 further has the following functions:

[0167] The number of the single characters is at least two, and the single characters corresponding to each segmentation image are spliced ​​in a preset order to obtain the target storage location number corresponding to the parking space.

[0168] Optionally, the processor 402 further has the following functions:

[0169] Obtain a preset map; determine the current location of the vehicle based on the target storage location number and the preset map.

[0170] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0171] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a server, or a data center to another website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive Solid State Disk (SSD)), etc.

[0172] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0173] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0174] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0175] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0176] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0177] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining a storage location number. It is characterized in that include: Obtain a parking location map corresponding to the parking space and the number of single characters corresponding to the parking location number in the parking location map; Segmenting the storage location map to obtain a segmentation map of the number of single characters; Inputting the segmentation graphs of the number of single characters into a preset character recognition model to obtain a single character corresponding to each segmentation graph; Determine the target parking space number corresponding to the parking space according to the single character corresponding to each segmentation graph; Get the number of single characters corresponding to the storage location number in the storage location map, including: Binarization is performed on the storage location map to obtain a binary image; Scan the column pixels of the binary image to determine a first number of times that the sum of the first target column pixels is zero and the sum of the second target column pixels is not zero, and the first target column pixels are adjacent to the second target column pixels; determine the number of single characters corresponding to the storage location number in the storage location map according to the first number of times; or, Scan the row pixels of the binary image to determine the second number of times that the sum of the first target row pixels is zero and the sum of the second target row pixels is not zero, and the first target row pixels are adjacent to the second target row pixels; based on the second number of times, determine the number of single characters corresponding to the storage location number in the storage location map.

2. The method according to claim 1, It is characterized in that The binarization process of the storage location map to obtain a binarized image includes: Binarizing the storage location map to obtain a first image; Acquire closed regions in the first image and the area of ​​each closed region; In the storage location map, closed areas whose areas are smaller than a preset area threshold are set to black to obtain a binary image.

3. The method according to claim 1, It is characterized in that The binarization process of the storage location map to obtain a binarized image includes: Performing a first preprocessing on the storage location map to obtain a preprocessed image, wherein the first preprocessing includes scaling according to a first preset size and / or filtering; The pre-processed image is binarized to obtain a binarized image.

4. The method according to any one of claims 1 to 3, It is characterized in that The step of inputting the segmentation graphs of the number of single characters into a preset character recognition model to obtain the single characters corresponding to each segmentation graph comprises: A convolutional neural network (CNN) classifier is obtained by performing data training using a deep learning processing method based on a public data set of preset parking spaces and / or a proprietary data set collected for parking space number scenarios; The segmentation maps of the number of single characters are input into the CNN classifier, and the single characters corresponding to each segmentation map are output.

5. The method according to any one of claims 1 to 3, It is characterized in that The number of the single characters is at least two, and obtaining the target storage location number corresponding to the parking space according to the single characters corresponding to each segmentation graph includes: The single characters corresponding to each segmentation image are spliced ​​in a preset order to obtain the target storage location number corresponding to the parking space.

6. The method according to claim 1, It is characterized in that The method further comprises: Get the preset map; The current location of the vehicle is determined based on the target storage location number and the preset map.

7. A vehicle, It is characterized in that include: An acquisition module, used to acquire a parking location map corresponding to the parking space and the number of single characters corresponding to the parking location number in the parking location map; A processing module is used to segment the parking lot map to obtain segmentation maps of the number of single characters; input the segmentation maps of the number of single characters into a preset character recognition model to obtain a single character corresponding to each segmentation map; and determine a target parking lot number corresponding to the parking lot according to the single character corresponding to each segmentation map; The acquisition module is also used to binarize the storage location map to obtain a binary image; scan the column pixels of the binary image to determine the first number of times that the sum of the first target column pixels is zero and the sum of the second target column pixels is not zero, and the first target column pixels are adjacent to the second target column pixels; determine the number of single characters corresponding to the storage location number in the storage location map based on the first number of times; or scan the row pixels of the binary image to determine the second number of times that the sum of the first target row pixels is zero and the sum of the second target row pixels is not zero, and the first target row pixels are adjacent to the second target row pixels; determine the number of single characters corresponding to the storage location number in the storage location map based on the second number of times.

8. A vehicle, It is characterized in that include: A memory storing executable program code; and a processor coupled to the memory; The processor calls the executable program code stored in the memory, and when the executable program code is executed by the processor, the processor implements the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having executable program code stored thereon, It is characterized in that When the executable program code is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images