A method, apparatus and device for indoor direction recognition and navigation using OCR technology

Abstract

The application discloses a method, device and equipment for indoor direction identification and indoor navigation by using OCR technology, the method for indoor direction identification by using OCR technology comprises the following steps: acquiring key character information in a building; acquiring a two-dimensional picture of a plane where the key character information is located; acquiring a taken site picture, analyzing the site picture by using OCR technology to acquire picture character information; matching the picture character information with the key character information, acquiring a matched two-dimensional picture, and acquiring direction information of the site picture according to the matched two-dimensional picture. The method for indoor navigation by using OCR technology further comprises the following steps on the basis of the above indoor direction identification: establishing three-dimensional positioning information of a site; acquiring destination position information, generating a navigation route according to the three-dimensional positioning information of the site matched with the picture character information, in combination with the direction information and a gyroscope. The method for indoor direction identification and navigation is simple in operation, low in hardware requirement, and improves the accuracy of indoor direction identification and the intelligence of navigation.

Description

Technical Field

[0001] This invention relates to the field of computer vision, and more particularly to a method, apparatus, and device for indoor orientation and navigation using OCR technology. Background Technology

[0002] Indoor public spaces are complex environments with significant signal interference and three-dimensional structures, making GPS ineffective for accurate three-dimensional positioning. Beacon positioning and laser-plus-vision positioning both have substantial drawbacks. Currently, indoor positioning and navigation utilize technologies such as Bluetooth and UWB (Ultra-Wideband) beacons. Bluetooth requires attitude calculation for positioning; while this method is inexpensive, its signal is unstable, has poor interference resistance, and low accuracy. UWB positioning offers high accuracy, but beacons are expensive, resulting in higher maintenance costs.

[0003] Laser SLAM (Simultaneous Localization and Mapping) is also a localization method. It achieves localization and navigation by scanning the entire environment beforehand and comparing it with a 3D point cloud map. However, this method requires a lot of computation in environments where the image environment changes rapidly, such as hospitals, train stations, shopping malls, and airports, and its adaptability is slightly lower. Moreover, it lacks orientation discrimination based on the user's location and cannot provide the user with comprehensive positioning and navigation information.

[0004] Therefore, there is a need to provide a method, apparatus, and device for indoor orientation and navigation using OCR technology, to improve the accuracy of indoor positioning, navigation, and orientation identification, reduce the computational cost of physical resources, and enhance the user's indoor navigation experience. Summary of the Invention

[0005] The purpose of this invention is to provide a method, apparatus, and device for indoor orientation and navigation using OCR technology, thereby improving the accuracy of indoor positioning, navigation, and orientation identification, reducing the computational application cost of physical resources, enhancing the intelligence of indoor navigation, and optimizing the user's experience in indoor orientation, positioning, and navigation.

[0006] To achieve the above objectives, the present invention provides a method for indoor orientation determination using OCR technology, comprising:

[0007] Obtain key character information from inside the building;

[0008] Obtain a two-dimensional image of the plane containing the key character information;

[0009] Acquire images of the location, and use OCR technology to parse the images to obtain image character information;

[0010] The image character information is matched with the key character information to obtain a matching two-dimensional image, and the orientation information of the location image is analyzed based on the matching two-dimensional image.

[0011] Optionally, the key character information includes one or more of the following: door number, security checkpoint, and specific location number.

[0012] Optionally, the method for parsing the location image includes, in the following order: grayscale conversion of the location image, noise reduction of the location image, binarization of the location image, flipping and correcting the orientation of character information, obtaining the shooting orientation of the character information, correcting the character information, segmenting the character information, normalizing the data, extracting feature data, and using a classifier to identify the feature data.

[0013] Optionally, the method for obtaining the orientation information of the location image based on the matched two-dimensional image specifically involves: obtaining the outer frame of the image character information on the location image; if the outer frame is rectangular, the orientation angle is 0; if the outer frame is trapezoidal, the orientation angle is obtained using an angle formula; the angle formula is A = log a (1-R), where A is the direction angle, R = L1 / L2-1, L1 is the length of the lower base of the trapezoid, L2 is the length of the upper base of the trapezoid, and the value of a is determined by actual site measurement data and is a constant between 0 and 1.

[0014] The present invention also provides a method for indoor navigation using OCR technology, comprising:

[0015] Methods for obtaining indoor orientation using OCR technology;

[0016] Based on the building's internal structural information, create a 3D model of the site;

[0017] The key character information is matched with the location 3D map to determine the location, thereby generating location 3D positioning information;

[0018] The image character information is matched with the key character information to obtain the matching location's three-dimensional positioning information;

[0019] Obtain destination location information, generate a navigation route based on the matched three-dimensional location information, and combine the direction information with the real-time orientation information generated by the gyroscope.

[0020] Optionally, the building's internal structure information includes one or more of the following: number of building floors, building internal outline, shop location, and shop internal outline.

[0021] The present invention also provides an indoor orientation device using OCR technology, comprising:

[0022] The image construction module is used to obtain key character information inside the building and two-dimensional image information of the plane where the key character information is located;

[0023] The first image acquisition module is used to acquire information about the captured images;

[0024] The first image parsing module is used to parse the image information and obtain the direction information in the image information.

[0025] The present invention also provides an indoor navigation device utilizing OCR technology, comprising:

[0026] The 3D construction module is used to acquire 3D information, key character information, and 2D image information of the building's interior.

[0027] The second image acquisition module is used to acquire information about the captured images;

[0028] The second image parsing module is used to parse the image information and obtain the positioning and orientation information in the image information;

[0029] The navigation module is used to obtain destination location information and generate navigation routes.

[0030] The present invention also provides a device A, comprising:

[0031] The first memory is used to store computer programs;

[0032] A first processor is used to process the computer program to execute a method for obtaining indoor orientation using OCR technology.

[0033] The present invention also provides a device B, comprising:

[0034] The second memory is used to store computer programs;

[0035] A second processor is used to process the computer program to perform a method of indoor navigation using OCR technology.

[0036] This invention discloses a method for indoor orientation and navigation using OCR technology. It constructs a 3D building map and 2D image information, associates key character information with the 2D image information of the plane in which they are located, and calibrates the key character information and the building's 3D positioning information. Then, it uses OCR technology to obtain image character information from a photograph of the location, thereby acquiring 2D image information and 3D positioning information that match the location image. The 3D positioning information effectively locates the user's position, helping them with location guidance; the 2D image information, supplemented by direction recognition technology, obtains indoor direction information, effectively helping users orient themselves indoors. This indoor orientation and positioning method is computationally simple, has low hardware requirements, improves the accuracy of indoor orientation and positioning, and enhances the intelligence of navigation by combining indoor direction information with gyroscope-generated real-time navigation orientation. Attached Figure Description

[0037] Figure 1 This is a flowchart of an indoor orientation identification method using OCR technology in a specific embodiment of the present invention;

[0038] Figure 2 This is a flowchart of an indoor navigation method using OCR technology according to a specific embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of an indoor orientation detection scenario using OCR technology in a specific embodiment of the present invention.

[0040] Figure 4 This is a schematic diagram of an indoor orientation identification frame using OCR technology in a specific embodiment of the present invention.

[0041] Figure 5 This is a block diagram of an indoor orientation device utilizing OCR technology in a specific embodiment of the present invention;

[0042] Figure 6 This is a block diagram of an indoor navigation device utilizing OCR technology according to a specific embodiment of the present invention. Detailed Implementation

[0043] The method, apparatus, and device for indoor orientation and navigation using OCR technology according to the present invention will now be described in more detail with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. It should be understood that those skilled in the art can modify the invention described herein while still achieving its advantageous effects. Therefore, the following description should be understood as being of general knowledge to those skilled in the art and is not intended to limit the invention.

[0044] Example 1

[0045] Please refer to Figure 1 , Figure 1This is a flowchart of an indoor orientation method using OCR technology in a specific embodiment of the present invention.

[0046] Step S101: Obtain key character information inside the building;

[0047] Step S102: Obtain a two-dimensional image of the plane containing the key character information;

[0048] Step S103: Acquire the captured location image, and use OCR technology to parse the location image to obtain image character information;

[0049] Step S104: Match the image character information with the key character information to obtain a matching two-dimensional image, and analyze the orientation information of the location image based on the matching two-dimensional image.

[0050] Specifically, in step S101, key character information inside the building is obtained, including one or more of the following: door number, security checkpoint, and specific location number.

[0051] In step S102, a two-dimensional image of the plane where the key character information is located is obtained, that is, the planar image information of the wall, stairs or other objects where the key character is located. The key character information and the two-dimensional image information are correlated with each other, and the two-dimensional image information of the plane where the key character is located can be obtained based on the key character information.

[0052] In step S103, a captured scene image is acquired. This image can be obtained through a mobile device or robot, or other electronic device with image acquisition capabilities. OCR technology is then used to parse the scene image and obtain its character information. Specifically, the scene image processing, in the following order, includes: grayscale conversion of the scene image, noise reduction of the scene image, binarization of the scene image, flipping and correcting the character information orientation, acquiring the character information shooting orientation, correcting the character information, segmenting the character information, normalizing the data, extracting feature data, and using a classifier to recognize the feature data, thereby obtaining image character information that can be used for data matching.

[0053] In step S104, the image character information is matched with the key character information to obtain a matched two-dimensional image, and the direction information of the location image is obtained based on the matched two-dimensional image. The specific method for obtaining the direction information is as follows: The outer frame of the image character information on the location image is obtained. If the outer frame is rectangular, the direction angle is 0; if the outer frame is trapezoidal, the direction angle is obtained using an angle formula: A = log0 a(1-R), where A is the directional angle, R = L1 / L2-1, L1 is the length of the lower base of the trapezoid, L2 is the length of the upper base of the trapezoid, and the value of a is determined by actual site measurement data and is a constant between 0 and 1. The directional angle can be calculated by determining the angular relationship between the site image and the horizontal plane, and the angular relationship between the site image and the vertical plane. The directional angle relationship is identified by the location of the parallel lines of the trapezoid.

[0054] For further details, please refer to... Figure 3 and Figure 4 The actual testing method is as follows: The tester uses a stand to fix the mobile terminal and takes a picture of the text. The shooting angle is adjusted to place the key character information in the center of the picture. The center position can be marked by a rectangular frame marked by software. A line is drawn from directly below the mobile terminal to directly below the key character information to measure the angle A. Furthermore, to reduce the error of manual measurement, at least 7 pictures need to be taken for the same angle to calculate the value of R. After removing the maximum and minimum values, the average value of R for that angle is obtained. The testing accuracy of angle A is 1 degree. The test data is summarized in a table to intuitively show the correspondence between A and R. Based on the different ranges of R, the corresponding value of a is derived. For example, when R is between 0 and 0.1, a = 1 / 3; when R is between 0.1 and 0.3, a = 2 / 5.

[0055] Furthermore, if the value of R is existing data in the test table, the value of A can be directly obtained from the table. If the value of R is not in the table, the value of A can be obtained through the angle formula.

[0056] This indoor orientation method using OCR technology obtains character information from captured images, associates key character information with the two-dimensional image information of the plane in which they are located, and uses orientation recognition technology to obtain orientation information, effectively helping users to identify directions.

[0057] Example 2

[0058] Please refer to Figure 2 , Figure 2 This is a method for indoor navigation using OCR technology in a specific embodiment of the present invention.

[0059] Based on the above specific embodiments, the indoor navigation method of this embodiment further includes:

[0060] Step S201: Draw a 3D map of the site based on the building's internal structure information;

[0061] Step S202: The key character information is matched with the location 3D map to determine the location and generate location 3D positioning information;

[0062] Step S203: Match the image character information with the key character information to obtain the matched location three-dimensional positioning information;

[0063] Step S204: Obtain destination location information, generate a navigation route based on the matched three-dimensional positioning information of the location, and in combination with the direction information and the real-time orientation information generated by the gyroscope.

[0064] Specifically, in step S201, a three-dimensional site map of the building's internal structure is drawn, wherein the structural information includes one or more of the following: number of building floors, building internal outline, shop location, and shop internal outline. The buildings are mostly two-story or higher buildings, and the three-dimensional site map can clearly show the location information of shops, corridors, restrooms, elevators, etc., inside the building.

[0065] In step S202, the key character information is matched with the location in the 3D map of the site. This involves recording the key character information in the 3D map of the site and determining its location within the building, thereby generating 3D location information for the site. This allows the location in the 3D map to be identified based on the key character information.

[0066] In step S203, the image character information is matched with the key character information. Based on the matched key character information, the associated three-dimensional location information can be obtained. The three-dimensional location information can clearly and accurately display the location where the image was taken, thereby locating the user's position and obtaining the user's location information. This user location acquisition process is simple, effectively reduces the investment of physical resources, and provides accurate positioning with low signal dependence, effectively improving positioning accuracy in complex indoor environments.

[0067] In step S204, destination location information is obtained, a route map is generated based on the matched three-dimensional location information, and a navigation route is generated by combining the direction information and the real-time orientation information generated by the gyroscope. The direction information determines the user's initial orientation angle, and the gyroscope can acquire the user's real-time angle changes. The combination of these two allows for dynamic monitoring of the user's orientation, facilitating faster steering adjustments and more accurate direction finding, thus improving navigation intelligence.

[0068] The present invention provides an indoor navigation method using OCR technology. By constructing a three-dimensional image and obtaining key character information corresponding to the location within the three-dimensional image, complete three-dimensional location information is generated. After a user takes a photo, the system can process the captured location image using OCR technology to obtain image character information. By matching this information with the key character information, the system obtains the location's three-dimensional positioning information. Furthermore, the user can determine the orientation when taking the photo by analyzing the location image and the corresponding two-dimensional image containing the image character information. This helps the user solve turning problems while moving forward and improves navigation intelligence.

[0069] Please refer to Figure 5 , Figure 5 This is a block diagram of an indoor orientation device using OCR technology according to a specific embodiment of the present invention.

[0070] This embodiment provides a device for indoor orientation using OCR technology, comprising:

[0071] Image construction module 501 is used to obtain key character information inside the building and two-dimensional image information of the plane where the key character information is located;

[0072] The first image acquisition module 502 is used to acquire location image information for shooting;

[0073] The first image parsing module 503 is used to parse the location image information and obtain the direction information in the location image information.

[0074] Specifically, the image construction module 501 is used to obtain key character information inside the building and the two-dimensional image information where the key character information is located, and associate the two.

[0075] The first image acquisition module 502 is used to acquire images of the location taken by the user. The carrier can be a robot, mobile phone, tablet or other device with image acquisition function.

[0076] The first image parsing module 503 is used to parse the location image taken by the user, obtain the image character information in the location image, compare the image character information with the key character information, obtain the two-dimensional image corresponding to the image character information, and obtain the orientation when the location image was taken through certain analysis and calculation, thereby identifying the user's direction.

[0077] The device for indoor orientation using OCR technology of the present invention analyzes a location image taken by a user and obtains the directional information when the user took the location image through the cooperation between the image construction module, the first image acquisition module and the first image parsing module, thereby helping the user to identify directions.

[0078] Please refer to Figure 6 , Figure 6 This is a block diagram of an indoor navigation device utilizing OCR technology according to a specific embodiment of the present invention.

[0079] This embodiment provides a device for indoor navigation using OCR technology, including:

[0080] The 3D construction module 601 is used to acquire 3D information, key character information and 2D image information of the building interior;

[0081] The second image acquisition module 602 is used to acquire location image information for shooting;

[0082] The second image parsing module 603 is used to parse the location image information and obtain the location information and direction information in the location image information;

[0083] Navigation module 604 is used to obtain destination location information and generate navigation routes.

[0084] Specifically, the 3D construction module 601 is used to draw 3D image data of the building interior, obtain key character information and the 2D image information where the key character information is located, locate the key character information with the 3D image, and associate the key character information with the 2D image information.

[0085] The second image acquisition module 602 is used to acquire images of the location taken by the user. The carrier can be a robot, mobile phone, tablet or other device with image acquisition function.

[0086] The second image parsing module 603 is used to parse the location images taken by the user. On the one hand, it obtains the image character information in the captured location image, and obtains the corresponding three-dimensional location positioning information based on the matching of the image character information and key character information to locate the user's location. On the other hand, it obtains the two-dimensional image corresponding to the image character information, and obtains the orientation when the location image was taken through certain mathematical operations, thereby identifying the user's direction.

[0087] The navigation module 604 is used to obtain destination location information, generate a route map based on the matched three-dimensional positioning information of the location, and generate a navigation route by combining the direction information and the real-time orientation information generated by the gyroscope.

[0088] This invention relates to an indoor navigation device utilizing OCR technology. This device, which uses OCR technology to obtain indoor location and direction, employs a 3D construction module, an image acquisition module, an image analysis module, and a navigation module in close cooperation. It uses character information as a link between a user-captured photograph of a location and the user's 3D location information and 2D image information. This allows for the use of OCR technology to analyze the photograph, enabling precise indoor location positioning and orientation identification. This improves the accuracy of indoor location positioning and orientation identification, reduces computational costs, and enhances the convenience and intelligence of indoor navigation for users.

[0089] The present invention also provides a device A, comprising:

[0090] The first memory is used to store computer programs;

[0091] A first processor is used to process the computer program to execute a method for obtaining indoor orientation using OCR technology.

[0092] A device B, characterized in that it comprises:

[0093] The second memory is used to store computer programs;

[0094] The second processor is used to process the computer program to execute a method for obtaining indoor navigation using OCR technology.

[0095] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for indoor orientation determination using OCR technology, characterized in that, include: Obtain key character information from inside the building; Obtain a two-dimensional image of the plane containing the key character information; Acquire images of the location, and use OCR technology to parse the images to obtain image character information; The analysis method, in the following order, includes: grayscale conversion of the scene image, noise reduction of the scene image, binarization of the scene image, flipping and correcting the character information orientation, obtaining the shooting orientation of the character information, correcting the character information, segmenting the character information, normalizing the data, extracting feature data, and using a classifier to identify the feature data. The image character information is matched with the key character information to obtain a matching two-dimensional image, and the orientation information of the location image is analyzed based on the matching two-dimensional image. The specific analysis method is as follows: Obtain the outer frame of the image character information on the location image. If the outer frame is rectangular, the direction angle is 0. If the outer frame of the sign is trapezoidal, the direction angle is obtained using an angle formula; The angle formula is A=log a (1-R), where A is the direction angle, R=L1 / L2-1, L1 is the length of the lower base of the trapezoid, L2 is the length of the upper base of the trapezoid, and the value of a is determined by actual site measurement data and is a constant between 0 and 1.

2. The method for indoor orientation using OCR technology according to claim 1, characterized in that, The key character information includes one or more of the following: door number, security checkpoint, and specific location number.

3. A method for indoor navigation using OCR technology, characterized in that, include: The method for obtaining indoor orientation using OCR technology as described in claim 1 or 2; Based on the building's internal structural information, create a 3D model of the site; The key character information is matched with the location 3D map to determine the location, thereby generating location 3D positioning information; The image character information is matched with the key character information to obtain the matching location's three-dimensional positioning information; Obtain destination location information, and generate a navigation route based on the matched three-dimensional location information, combined with the direction information and the real-time orientation information generated by the gyroscope.

4. The method for indoor navigation using OCR technology according to claim 3, characterized in that, The building's internal structure information includes one or more of the following: number of building floors, building internal outline, shop location, and shop internal outline.

5. A device for indoor orientation using OCR technology, characterized in that, include: The image construction module is used to obtain key character information inside the building and two-dimensional image information of the plane where the key character information is located; The first image acquisition module is used to acquire image information of the shooting location; The first image parsing module is used to parse the location image information and obtain the direction information in the location image information.

6. A device for indoor navigation using OCR technology, characterized in that, include: The 3D construction module is used to acquire 3D information, key character information, and 2D image information of the building's interior. The second image acquisition module is used to acquire image information of the shooting location; The second image parsing module is used to parse the location image information and obtain the positioning and direction information in the image information; The navigation module is used to obtain destination location information and generate navigation routes.

7. A device A, characterized in that, include: The first memory is used to store computer programs; A first processor is configured to process the computer program to execute any one of the methods for obtaining indoor orientation using OCR technology as described in claims 1 to 4.

8. A device B, characterized in that, include: The second memory is used to store computer programs; A second processor is used to process the computer program to perform the indoor navigation method using OCR technology as described in claim 5 or 6.

Images