CN111813984B

CN111813984B - Method and device for realizing indoor positioning by using homography matrix and electronic equipment

Info

Publication number: CN111813984B
Application number: CN202010583173.6A
Authority: CN
Inventors: 刘雯; 王硕; 常家兴
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2020-06-23
Filing date: 2020-06-23
Publication date: 2022-09-30
Anticipated expiration: 2040-06-23
Also published as: CN111813984A

Abstract

The embodiment of the invention provides a method and a device for realizing indoor positioning by using a homography matrix and electronic equipment, and is applied to the technical field of computers. The positioning method comprises the following steps: aiming at each query image in the first query image and the second query image, determining an image with the highest matching degree with the query image from a plurality of fingerprint images by utilizing a pre-constructed database corresponding to a specified direction as a matching image, and determining a target fingerprint image with the highest matching degree with the query image from two reference fingerprint images corresponding to the matching image; and respectively mapping the target fingerprint image and the matched image to the query image by using the homography matrix, determining a first distance and a second distance, and determining a coordinate value of the position of the object to be positioned, which is related to a first coordinate axis, by using the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position. This scheme need not to put up the sign in advance, has avoided the destruction to the original environment of indoor area.

Description

Method and device for realizing indoor positioning by using homography matrix and electronic equipment

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method and an apparatus for implementing indoor positioning using a homography matrix, and an electronic device.

Background

Carry out position location to moving objects such as intelligent robot in the indoor environment, belong to a common indoor location demand. Currently, indoor positioning is realized by using a homography matrix, which is a common mode for indoor positioning of a moving object.

In the related art, in the method for realizing indoor positioning by using the homography matrix, fingerprint images which are collected at a plurality of specified positions indoors and contain pre-posted marks are acquired in advance, and a database is constructed by using image data of each fingerprint image; when an indoor moving object needs to be positioned, a target image which is collected at the current position of the moving object and contains an identifier is obtained, a fingerprint image with the highest matching degree with the target image is determined as a matching image based on image data in a database and image data of the target image, a homography matrix of the target image and the matching image is determined, edge line constraint information, namely contour information of the identifier, of the matching image is determined by the homography matrix, and the current position of the moving object is determined by the edge line constraint information.

However, in the related art, in order to realize indoor positioning, a mark needs to be posted indoors in advance, which causes damage to the indoor original environment and influences the beauty of the indoor original environment.

Disclosure of Invention

The embodiment of the invention aims to provide a method, a device and electronic equipment for realizing indoor positioning by using a homography matrix, which are used for solving the problem that the original indoor environment is damaged by a mark pre-posted during indoor positioning in the prior art. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a method for implementing indoor positioning by using a homography matrix, including:

acquiring a first query image and a second query image; the first query image and the second query image are respectively images of an indoor area obtained by image acquisition in the direction of the extension directions of an x axis and a y axis at the position of an object to be positioned; the x-axis and the y-axis are coordinate axes in a plane coordinate system constructed for the indoor area;

aiming at each query image in the first query image and the second query image, determining an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image by utilizing a pre-constructed database corresponding to a specified direction; the database comprises image data of the fingerprint images, each fingerprint image is an image obtained by acquiring an image towards the specified direction at a target position of the indoor area in advance, and the target position is a coordinate point in the plane coordinate system; the specified direction is the direction toward which the query image is collected;

determining a target fingerprint image with the highest matching degree with the query image from two reference fingerprint images corresponding to the matching image; each reference fingerprint image is an image acquired at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with a first coordinate axis of the second target position, and the second target position is a target position where the matching image is acquired; the first coordinate axis is a coordinate axis parallel to the specified direction;

respectively mapping the target fingerprint image and the matching image to the query image by using the homography matrix between the target fingerprint image and the query image and the homography matrix between the matching image and the query image to obtain a first mapping image and a second mapping image corresponding to the query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the query image, wherein the first pixel point is a point at a specified position in each mapping image, and the second pixel point is a point at the specified position in the query image;

determining a coordinate value of the position of the object to be positioned relative to a second coordinate axis by using the first distance, the second distance, the coordinate of the second target position and the coordinate of a third target position; the third target position is a target position where the target fingerprint image is located when being collected, and the second coordinate axis is a coordinate axis other than the first coordinate axis.

Optionally, before determining the target fingerprint image that has the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image, the method further includes:

mapping the matching image to the query image by using the homography matrix of the matching image and the query image to obtain a third mapping image corresponding to the query image;

determining a third distance between a third pixel point of the third mapping image and a second pixel point of the query image; the third pixel point is a point at the specified position in the third mapping image;

judging whether the third distance is greater than a preset threshold value, if so, determining a target fingerprint image with the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image;

if not, taking the coordinate value of the second target position relative to the second coordinate axis as the coordinate value of the position of the object to be positioned relative to the second coordinate axis.

Optionally, when the specified direction is an extending direction toward a y axis, the determining, by using the first distance, the second distance, the coordinate of the second target position, and the coordinate of the third target position, a coordinate value of the position where the object to be positioned is located with respect to a second coordinate axis includes:

obtaining a coordinate value of the position of the object to be positioned relative to the x axis by adopting a preset first coordinate calculation formula based on the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position;

the first coordinate calculation formula includes:

wherein x is _l Coordinate values about the x-axis for the location of the object to be positioned; x when the absolute value of the abscissa of the third target position is smaller than the absolute value of the abscissa of the second target position ₁ Is the abscissa, x, of the third target position ₂ Is the abscissa, t, of the second target position ₁ Is the first distance, t ₂ Is the second distance; x when the absolute value of the abscissa of the third target position is larger than the absolute value of the abscissa of the second target position ₁ Is the abscissa, x, of the second target position ₂ Is the abscissa, t, of the third target position ₁ Is the second distance, t ₂ Is the first distance;

when the specified direction is an extending direction facing to an x axis, the determining, by using the first distance, the second distance, the coordinate of the second target position, and the coordinate of the third target position, the coordinate information about the second coordinate axis of the position where the object to be positioned is located includes:

obtaining a coordinate value of the position of the object to be positioned relative to the y axis by adopting a preset second coordinate calculation formula based on the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position;

the second coordinate calculation formula includes:

wherein, y _l Coordinate values about the y-axis for the position of the object to be positioned; y when the absolute value of the abscissa of the third target position is smaller than the absolute value of the abscissa of the second target position ₁ Is the ordinate, y, of the third target position ₂ Is the ordinate, t, of the second target position ₁ Is the first distance, t ₂ Is the second distance; y when the absolute value of the abscissa of the third target position is larger than the absolute value of the abscissa of the second target position ₁ Is the ordinate, y, of the second target position ₂ Is the ordinate, t, of the third target position ₁ Is the second distance, t ₂ Is the first distance.

Optionally, the plurality of fingerprint images belong to a plurality of image groups, each fingerprint image having a unique number; for each image group, the coordinates of the target positions corresponding to all the fingerprint images in the image group on the first coordinate axis are the same, and the serial number of each fingerprint image in the image group is increased or decreased according to the coordinates of the second coordinate axis;

the determining mode of the two reference fingerprint images corresponding to the matching image comprises the following steps:

and inquiring the number of the matched image and the target image group to which the matched image belongs, determining two numbers adjacent to the number of the matched image in the target image group, and determining fingerprint images corresponding to the two numbers to obtain two reference fingerprint images.

In a second aspect, an embodiment of the present invention provides an apparatus for implementing indoor positioning using a homography matrix, including:

the image acquisition module is used for acquiring a first query image and a second query image; the first query image and the second query image are respectively images of indoor areas obtained by acquiring images towards the extending directions of an x axis and a y axis at the position of an object to be positioned; the x-axis and the y-axis are coordinate axes in a plane coordinate system constructed for the indoor area;

the image matching module is used for determining an image with the highest matching degree with the query image from the plurality of fingerprint images as a matching image by utilizing a pre-constructed database corresponding to the specified direction aiming at each query image in the first query image and the second query image; the database comprises image data of the fingerprint images, each fingerprint image is an image obtained by acquiring an image towards the specified direction at a target position of the indoor area in advance, and the target position is a coordinate point in the plane coordinate system; the specified direction is the direction toward which the query image is collected;

the image determining module is used for determining a target fingerprint image with the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image; each reference fingerprint image is an image collected at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with a first coordinate axis of the second target position, and the second target position is a target position where the matching image is collected; the first coordinate axis is a coordinate axis parallel to the specified direction;

an image mapping module, configured to map the target fingerprint image and the matching image to the query image respectively by using a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image, to obtain a first mapping image and a second mapping image corresponding to the query image, determine a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determine a second distance between the first pixel point of the second mapping image and the second pixel point of the query image, where the first pixel point is a point at an assigned position in each mapping image and the second pixel point is a point at the assigned position in the query image;

the position determining module is used for determining coordinate values of the position of the object to be positioned relative to a second coordinate axis, wherein the coordinate values are of the first distance, the second distance, the coordinate of the second target position and the coordinate of a third target position; the third target position is a target position where the target fingerprint image is located when being collected, and the second coordinate axis is a coordinate axis other than the first coordinate axis.

Optionally, an apparatus for implementing indoor positioning by using a homography matrix provided in an embodiment of the present invention further includes:

the image judging module is used for mapping the matching image to the query image by utilizing the homography matrix of the matching image and the query image before the image determining module determines the target fingerprint image with the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image, so as to obtain a third mapping image corresponding to the query image;

Optionally, the position determining module is specifically configured to:

when the designated direction is an extending direction facing to a y axis, obtaining a coordinate value of the position of the object to be positioned relative to the x axis by adopting a preset first coordinate calculation formula based on the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position;

the first coordinate calculation formula includes:

obtaining a coordinate value of the position of the object to be positioned, which is related to the y axis, by adopting a preset second coordinate calculation formula based on the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position;

the second coordinate calculation formula includes:

wherein, y _l A coordinate value about the y axis for the position of the object to be positioned; y when the absolute value of the abscissa of the third target position is smaller than the absolute value of the abscissa of the second target position ₁ Is the ordinate, y, of the third target position ₂ Is the ordinate, t, of the second target position ₁ Is the first distance, t ₂ Is the second distance; y when the absolute value of the abscissa of the third target position is larger than the absolute value of the abscissa of the second target position ₁ Is the ordinate, y, of the second target position ₂ Is the ordinate, t, of the third target position ₁ Is the second distance, t ₂ Is the first distance.

Optionally, the plurality of fingerprint images belong to a plurality of image groups, each fingerprint image having a unique number; for each image group, the coordinates of the target positions corresponding to all the fingerprint images in the image group on the first coordinate axis are the same, and the serial number of each fingerprint image in the image group is increased or decreased according to the coordinates of a second coordinate axis, wherein the second coordinate axis is a coordinate axis except the first coordinate axis;

In a third aspect, an embodiment of the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

and the processor is used for realizing the steps of the method for realizing indoor positioning by utilizing the homography matrix provided by the embodiment of the invention when executing the program stored on the memory.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements the steps of the method for implementing indoor positioning using a homography matrix, provided by an embodiment of the present invention.

The embodiment of the invention has the following beneficial effects:

in the scheme provided by the embodiment of the invention, aiming at each query image in a first query image and a second query image, a database which is constructed in advance and corresponds to a specified direction is utilized to determine an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image, and a target fingerprint image with the highest matching degree with the query image is determined from two reference fingerprint images corresponding to the matching image; respectively mapping the target fingerprint image and the matching image to the query image by utilizing a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image to obtain a first mapping image and a second mapping image corresponding to the query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the query image; and determining the coordinate value of the position of the object to be positioned relative to a second coordinate axis by using the first distance, the second distance, the coordinate of the second target position and the coordinate of a third target position. Therefore, according to the scheme, when the mobile object in the indoor area is positioned, the position difference of each query image and the position difference of the first mapping image and the second mapping image at the appointed position are determined without pasting marks in advance, and the accurate positioning of the query image is realized by utilizing the determined position difference and the coordinates of the target position of the matching image and the target fingerprint image when the matching image and the target fingerprint image are collected. Consequently, can solve among the prior art in the indoor location mark of posting in advance among this scheme, destroy the problem of indoor original environment.

Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other embodiments can be obtained according to the drawings without creative efforts.

FIG. 1 is a flow chart of a method for implementing indoor positioning using homography matrix according to the present invention;

FIG. 2 is another flowchart of a method for implementing indoor positioning using homography matrix according to the present invention;

FIG. 3 is a schematic structural diagram of an apparatus for implementing indoor positioning using homography matrix according to the present invention;

FIG. 4 is another structural diagram of an apparatus for indoor positioning using homography matrix according to the present invention;

FIG. 5 is a schematic structural diagram of an electronic device according to the present invention;

FIG. 6 is a homography matrix mapping relationship diagram;

FIG. 7 is a diagram of a coordinate transformation relationship according to the present invention;

FIG. 8 is an imaging geometry of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to solve the problem that in the prior art, an original indoor environment is damaged by a pre-posted identifier during indoor positioning, the invention provides a method and a device for realizing indoor positioning by using a homography matrix, and electronic equipment.

First, a method for implementing indoor positioning by using a homography matrix according to an embodiment of the present invention is described below.

The indoor positioning method provided by the embodiment of the invention aims at indoor positioning of a moving object, and the moving object can be an intelligent robot. For example: when the intelligent robot for guiding customers exists in a shopping mall or serving customers in the shopping mall needs to be positioned indoors, the indoor positioning method provided by the embodiment of the invention can be utilized.

The method for realizing indoor positioning by using the homography matrix provided by the embodiment of the invention can be applied to electronic equipment. In a particular application, the electronic device may be a control device of a moving object to be positioned, for example: smart phones, tablet computers, notebook computers, desktop computers, and the like; of course, it is also reasonable that the electronic device is a moving object to be positioned.

As shown in fig. 1, a method for implementing indoor positioning by using a homography matrix according to an embodiment of the present invention may include the following steps:

s101, acquiring a first query image and a second query image;

the first query image and the second query image are respectively images of indoor areas obtained by acquiring images towards the extending directions of an x axis and a y axis at the position of an object to be positioned; the x-axis and the y-axis are coordinate axes in a planar coordinate system constructed for the indoor area.

Specifically, the object to be positioned may be a moving object in an indoor area, such as an intelligent robot, and the position of the object to be positioned may be any position on the ground of the indoor area. If the execution main body is a moving object to be positioned, when the positioning is needed, an image acquisition module of the moving object, such as a camera, is utilized to acquire images towards the extending direction of the x axis to obtain a first query image, and the image acquisition module is utilized to acquire images towards the extending direction of the y axis to obtain a second query image; if the execution main body is the control device, when positioning is needed, the moving object to be positioned uses the image acquisition device of the moving object, such as a camera, to acquire an image in the extending direction of the x axis to obtain a first query image, and acquires an image in the extending direction of the y axis to obtain a second query image, and the first query image and the second query image are uploaded to the control device, so that the control device can obtain the first query image and the second query image.

S102, aiming at each query image in the first query image and the second query image, determining an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image by utilizing a pre-constructed database corresponding to a specified direction;

the database corresponding to the designated direction comprises image data of the plurality of fingerprint images, each fingerprint image is an image obtained by acquiring an image towards the designated direction at a target position of the indoor area in advance, and the target position is a coordinate point in the plane coordinate system; the specified direction is the direction toward which the query image is acquired; it will be appreciated that the origin of the planar coordinates may be any point on the ground of the indoor area, with each target location having one coordinate.

Specifically, the plurality of fingerprint images are images acquired by a predetermined acquisition device performing image acquisition toward a specified direction at each target position. The predetermined acquisition device may be a mobile device to be positioned or a mobile object of the same type as the mobile object to be positioned.

It can be understood that a plurality of fingerprint images obtained by image acquisition towards the extending direction of the x-axis are used for constructing a database corresponding to the extending direction of the x-axis, and a plurality of fingerprint images obtained by image acquisition towards the extending direction of the y-axis are used for constructing a database corresponding to the extending direction of the y-axis. Specifically, the method comprises the following steps: aiming at a first query image, determining an image with the highest matching degree with the first query image from a plurality of fingerprint images for constructing a database corresponding to the extension direction of the x axis by utilizing a database corresponding to the extension direction of the x axis, wherein the database is constructed in advance, and the image is used as a matching image corresponding to the first query image; and aiming at the second query image, determining an image with the highest matching degree with the second query image from a plurality of fingerprint images for constructing the database corresponding to the extension direction of the y axis by utilizing the database corresponding to the extension direction of the y axis, which is constructed in advance, as a matching image corresponding to the second query image.

It should be noted that the image data of the fingerprint image may be a feature matrix of the fingerprint image or the fingerprint image itself, which is reasonable; correspondingly, when the image data of the fingerprint image is the feature matrix of the fingerprint image, the implementation manner of determining the image with the highest matching degree with the query image from the plurality of fingerprint images may be: matching the feature matrix corresponding to the query image with the feature matrix corresponding to each fingerprint image to obtain a matching degree, and taking the fingerprint image corresponding to the feature matrix with the highest matching degree as a matching image; when the image data of the fingerprint image is the fingerprint image itself, the implementation manner of determining the image with the highest matching degree with the query image from the plurality of fingerprint images may be: the method comprises the steps of determining a feature matrix corresponding to an inquired image by adopting a feature extraction algorithm, matching the feature matrix corresponding to the inquired image with the feature matrix corresponding to each fingerprint image to obtain a matching degree, and using the fingerprint image corresponding to the feature matrix with the highest matching degree as a matched image, wherein the feature matrix algorithm is any algorithm capable of realizing the function of extracting the feature matrix corresponding to the image in the prior art, and in addition, the determination mode of the matching degree can adopt any mode capable of calculating the similarity of the two feature matrices, and is not limited herein.

In addition, since the processing procedures for the first query image and the second query image are similar, and the difference is only that the coordinate values of the coordinate axes of the position where the object to be positioned is located are determined by the first query image and the second query image, in the embodiment of the present invention, S102 to S105 are descriptions of the processing procedures for each of the first query image and the second query image, that is, the query images in the steps of S102 to S105 are all the query images to which the query images are directed.

S103, determining a target fingerprint image with the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image;

each reference fingerprint image is an image acquired at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with a first coordinate axis of the second target position, and the second target position is a target position where the matching image is acquired; the first coordinate axis is a coordinate axis parallel to the specified direction. Each reference fingerprint image is specifically an image acquired by a preset acquisition device at a first target position, and the second target position is a target position where the preset acquisition device is located when the matching image is acquired. Wherein, the first coordinate axis is an x axis or a y axis, specifically: in the processing process of the first query image, the designated direction is the extending direction towards the x axis, and at the moment, the first coordinate axis is the x axis; in the process of processing the second query image, the designated direction is an extending direction towards the y axis, and at this time, the first coordinate axis is the y axis.

For clarity of the solution, the following illustrates the first target position as a target position adjacent to the second target position and having the same coordinate as the first coordinate axis of the second target position.

Assuming that coordinate points exist, respectively: seven target positions of A (1, 0), B (1, 1), C (1, 2), D (2, 0), E (2, 1), F (2, 2) and G (1, 3); if the first coordinate axis is the x axis and B (1, 1) is the second target position, A (1, 0) and C (1, 2) are target positions which are adjacent to the second target position and have the same coordinate with the first coordinate axis of the second target position, and two fingerprint images corresponding to A (1, 0) and C (1, 2) are two reference fingerprint images; if the first coordinate axis is the x axis and E (2, 1) is the second target position, D (2, 0) and F (2, 2) are target positions adjacent to the second target position and having the same coordinate as the first coordinate axis of the second target position, and the two fingerprint images corresponding to D (2, 0) and F (2, 2) are the two reference fingerprint images.

There are various implementation manners for determining the two reference fingerprint images corresponding to the matching image, for example, in one implementation manner, an association relationship of each fingerprint image is pre-constructed so as to associate each fingerprint image with the two other fingerprint images, and the two fingerprint images associated with the matching image are determined from the pre-constructed association relationship as the two reference fingerprint images. For clarity of the scheme and clarity of layout, other specific implementations of determining two reference fingerprint images from a plurality of fingerprint images will be described in detail later with reference to specific embodiments.

In addition, as an example, the implementation manner of determining the target fingerprint image with the highest matching degree with the query image may be: and matching the characteristic matrix corresponding to each reference fingerprint image with the characteristic matrix corresponding to the query image to obtain the matching degree, and taking the fingerprint image corresponding to the characteristic matrix with the highest matching degree as a target fingerprint image.

S104, respectively mapping the target fingerprint image and the matching image to the query image by using a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image to obtain a first mapping image and a second mapping image corresponding to the query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the query image;

the method comprises the steps of obtaining a target fingerprint image, obtaining a matching image, mapping the target fingerprint image and the matching image to an inquiry image by utilizing a homography matrix between the target fingerprint image and the inquiry image to obtain a first mapping image corresponding to the inquiry image, and mapping the matching image to the inquiry image by utilizing the homography matrix between the matching image and the inquiry image to obtain a second mapping image.

The first pixel point is a point at a specified position in each mapping image, and the second pixel point is a point at the specified position in the query image.

Illustratively, a ransac (random Sample consensus) algorithm is adopted to obtain a homography matrix between the target fingerprint image and the query image, and a homography matrix between the matching image and the query image.

Specifically, the first pixel point may be a pixel point at any vertex position of four vertices of the mapping image, or a center pixel point, and the second pixel point may be a pixel point at any vertex position of four vertices of the query image, or a center pixel point.

S105, determining a coordinate value of the position of the object to be positioned relative to a second coordinate axis by using the first distance, the second distance, the coordinate of the second target position and the coordinate of a third target position;

and the third target position is the target position where the target fingerprint image is located when being collected.

There are various specific implementation manners for determining the coordinate value of the position of the object to be positioned with respect to the second coordinate axis by using the first distance, the second distance, the coordinate of the second target position, and the coordinate of the third target position.

For example, in an implementation manner, when the specified direction is an extending direction toward a y axis, a preset first coordinate calculation formula is adopted based on the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position to obtain a coordinate value of the position of the object to be positioned with respect to the x axis;

the first coordinate calculation formula includes:

the second coordinate calculation formula includes:

wherein, y _l Coordinate values about the y-axis for the position of the object to be positioned; when the absolute value of the abscissa of the third target position is smaller than that of the second target positionTarget absolute value of y ₁ Is the ordinate, y, of the third target position ₂ Is the ordinate, t, of the second target position ₁ Is the first distance, t ₂ Is the second distance; y when the absolute value of the abscissa of the third target position is larger than the absolute value of the abscissa of the second target position ₁ Is the ordinate, y, of the second target position ₂ Is the ordinate, t, of the third target position ₁ Is the second distance, t ₂ Is the first distance.

For the sake of clarity, the process of determining the coordinate values of the coordinate axes of the position of the object to be located by using the first query image and the second query image is respectively described:

for the first query image, the processing procedure is as follows:

determining an image with the highest matching degree with a first query image from a plurality of fingerprint images by utilizing a pre-constructed database corresponding to the designated direction as a matching image corresponding to the first query image; the designated direction is the extending direction of the x axis; the database corresponding to the designated direction comprises a plurality of fingerprint images, and each fingerprint image is an image obtained by acquiring an image towards the extending direction of the x axis at the target position of an indoor area in advance;

determining a target fingerprint image with the highest matching degree with the first query image from two reference fingerprint images corresponding to the obtained matching image; each reference fingerprint image is an image acquired at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with the x-axis of the second target position, and the second target position is the target position where the determined matching image is acquired;

respectively mapping the matching image and the target fingerprint image to the first query image by using a homography matrix between the matching image and the first query image and a homography matrix between the target fingerprint image and the first query image to obtain a first mapping image and a second mapping image corresponding to the first query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the first query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the first query image;

determining a coordinate value of the position of the object to be positioned, which is related to the y axis, by using the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position; and the third target position is the target position where the target fingerprint image is located when being collected.

For the second query image, the processing procedure is as follows:

determining an image with the highest matching degree with a second query image from a plurality of fingerprint images by utilizing a pre-constructed database corresponding to the designated direction as a matching image corresponding to the second query image; the designated direction is the extending direction of the y axis; the database corresponding to the designated direction comprises a plurality of fingerprint images, and each fingerprint image is an image obtained by acquiring an image towards the extending direction of the y axis at the target position of an indoor area in advance;

determining a target fingerprint image with the highest matching degree with the second query image from two reference fingerprint images corresponding to the obtained matching image; each reference fingerprint image is an image collected at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with the y axis of the second target position, and the second target position is the target position where the determined matching image is collected;

respectively mapping the target fingerprint image and the matching image to a second query image by using a homography matrix between the target fingerprint image and the second query image and a homography matrix between the matching image and the second query image to obtain a first mapping image and a second mapping image corresponding to the second query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the second query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the second query image;

determining a coordinate value of the position of the object to be positioned relative to the x axis by using the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position; and the third target position is the target position where the target fingerprint image is located when being collected.

In the scheme provided by the embodiment of the invention, aiming at each query image in a first query image and a second query image, a database which is constructed in advance and corresponds to a specified direction is utilized to determine an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image, and a target fingerprint image with the highest matching degree with the query image is determined from two reference fingerprint images corresponding to the matching image; respectively mapping the target fingerprint image and the matching image to the query image by utilizing a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image to obtain a first mapping image and a second mapping image corresponding to the query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the query image; and determining the coordinate value of the position of the object to be positioned relative to the second coordinate axis by using the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position. It is thus clear that this scheme need not to post the sign in advance, has avoided destroying the problem of indoor original environment.

In the following, a method for implementing indoor positioning by using a homography matrix provided in the embodiments of the present application is described with reference to another embodiment.

S201, acquiring a first query image and a second query image;

the first query image and the second query image are respectively images of an indoor area obtained by image acquisition in the direction of the extension directions of an x axis and a y axis at the position of an object to be positioned; the x-axis and the y-axis are coordinate axes in a planar coordinate system constructed for the indoor area.

S202, aiming at each query image in the first query image and the second query image, determining an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image by utilizing a pre-constructed database corresponding to a specified direction;

the database comprises image data of the plurality of fingerprint images, each fingerprint image is an image obtained by acquiring an image facing the specified direction at a target position of the indoor area in advance, and the target position is a coordinate point in the plane coordinate system; the specified direction is the direction towards which the query image is acquired.

The steps of S201 to S202 in this embodiment are the same as those of S101 to S102 in the above embodiment, and are not described herein again.

S203, mapping the matching image to the query image by using the homography matrix of the matching image and the query image to obtain a third mapping image corresponding to the query image; determining a third distance between a third pixel point of the third mapping image and a second pixel point of the query image; judging whether the third distance is larger than a preset threshold value or not, and if so, executing S204; if not, taking the coordinate value of the second target position relative to the second coordinate axis as the coordinate value of the position of the object to be positioned relative to the second coordinate axis. Wherein the third pixel point is a point at the designated position in the third mapping image;

the third pixel point of the third mapping image may be a pixel point at any vertex position of four vertices of the third mapping image, or a center pixel point, the third pixel point is a point at the specified position in the third mapping image, and the third pixel point of the third mapping image may preferentially be the center pixel point of the third mapping image.

In order to reduce the time for determining the position of the object to be positioned, before determining a target fingerprint image with the highest matching degree with the query image from two reference fingerprint images corresponding to the matching image, determining a third distance between a third pixel point of the third mapping image and a second pixel point of the target query image, comparing the third distance with a preset threshold value, and judging whether the object to be positioned is at the target position, namely judging whether the third distance is greater than the preset threshold value, if the third distance is not greater than the preset threshold value, considering that the object to be positioned is at the target position, so that a coordinate value of the second target position relative to a second coordinate axis is used as a coordinate value of the position of the object to be positioned relative to the second coordinate axis. In order to make the scheme clear and the layout clear, the following process of determining the third distance by using the homography matrix of the matching image and the query image and judging whether the third distance is greater than the preset threshold value is described in detail with reference to a specific embodiment.

S204, determining a target fingerprint image with the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image;

each reference fingerprint image is an image collected at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with a first coordinate axis of the second target position, and the second target position is a target position where the matching image is collected; the first coordinate axis is a coordinate axis perpendicular to the specified direction.

S205, respectively mapping the target fingerprint image and the matching image to the query image by using a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image to obtain a first mapping image and a second mapping image corresponding to the query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the query image;

the first pixel point is a point at a designated position in each mapping image, and the second pixel point is a point at the designated position in the query image.

S206, determining a coordinate value of the position of the object to be positioned relative to a second coordinate axis by using the first distance, the second distance, the coordinate of the second target position and the coordinate of a third target position;

the third target position is a target position where the target fingerprint image is located when being collected, and the second coordinate axis is a coordinate axis other than the first coordinate axis.

S204 to S206 in this embodiment are the same as S103 to S105 in the above embodiments, and are not described herein again.

In the scheme provided by the embodiment of the invention, aiming at each query image in a first query image and a second query image, a database which is constructed in advance and corresponds to a specified direction is utilized to determine an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image, and a target fingerprint image with the highest matching degree with the query image is determined from two reference fingerprint images corresponding to the matching image; respectively mapping the target fingerprint image and the matching image to the query image by utilizing a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image to obtain a first mapping image and a second mapping image corresponding to the query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the query image; and determining the coordinate value of the position of the object to be positioned relative to a second coordinate axis by using the first distance, the second distance, the coordinate of the second target position and the coordinate of a third target position. Therefore, the scheme can realize accurate positioning of the target query image, does not need to paste the mark in advance, and avoids damage to the original environment of the indoor area due to the pasting of the mark. In addition, in the scheme, whether the object to be positioned is located at the target position is judged firstly by utilizing the homography matrix, so that the time required for determining the position of the object to be positioned is shortened, and the positioning efficiency is improved.

Optionally, in an implementation, the plurality of fingerprint images belong to a plurality of image groups, each fingerprint image having a unique number; for each image group, the coordinates of the target positions corresponding to all the fingerprint images in the image group on the first coordinate axis are the same, and the serial number of each fingerprint image in the image group is increased or decreased according to the coordinates of a second coordinate axis, wherein the second coordinate axis is a coordinate axis except the first coordinate axis;

the determining mode of the two reference fingerprint images corresponding to the matched image comprises the following steps:

In particular, numbering of multiple fingerprint images may comprise the following steps a1-a 2:

a1, determining a target position corresponding to each fingerprint image, and grouping a plurality of fingerprint images according to the coordinates of the target position to obtain a plurality of image groups, wherein the coordinates of the target positions corresponding to all the fingerprint images in each image group on the first coordinate axis are the same;

a2, numbering each fingerprint image in each image group, so that the number of each fingerprint image is increased or decreased according to the coordinates of the second coordinate axis;

the number form of each fingerprint image may be various, for example: 1. 2, 3, 4 … … i-1, i +1, or: A. b, C, D … … U, V, W, etc., and it is reasonable that the two numbers adjacent to the matching image number i, i.e., i-1 and i +1, and the two numbers adjacent to the matching image number V, i.e., U and W.

Illustratively, the presence coordinate points are: seven target positions of A (1, 0), B (1, 1), C (1, 2), D (2, 0), E (2, 1), F (2, 2), G (1, 3); if the first coordinate axis is an x-axis, the first image group includes: the coordinate points are: fingerprint images corresponding to the target positions of a (1, 0), B (1, 1), C (1, 2), and G (1, 3), the second image group including: the coordinate points are: d (2, 0), E (2, 1) and F (2, 2); further, within the first image group, the number of each fingerprint image may be: the number of the fingerprint image corresponding to the target position of the coordinate point A (1, 0) is 1, the number of the fingerprint image corresponding to the target position of B (1, 1) is 2, the number of the fingerprint image corresponding to the target position of C (1, 2) is 3, and the number of the fingerprint image corresponding to the target position of G (1, 3) is 4; alternatively, within the first image group, the number of each fingerprint image may be: the fingerprint image 1 corresponding to the target position of G (1, 3), the fingerprint image corresponding to the target position of C (1, 2) has the number 2, the fingerprint image corresponding to the target position of B (1, 1) has the number 3, and the fingerprint image corresponding to the target position of coordinate point a (1, 0) has the number 4, which is not limited herein.

For clarity of the scheme, the homography matrix is described in detail below.

As shown in fig. 6, the mapping relationship diagram is a homography matrix, where, for a point X in world coordinates, X is the central pixel point of the matching image, and the point X is (X, y, z,1) ^T Mapped on the imaging plane I with the pixel coordinate x ═ U (U) ₁ ,V ₁ ,1) ^T The pixel coordinate mapped on the imaging plane I' is x ═ (U) ₂ ,V ₂ ,1) ^T C and C' are the optical centers of the camera at the shooting positions of the query image and the matching image, respectively, and are obtained by a camera imaging model:

wherein M is ₁ Is an internal reference matrix of the camera, the elements of which are intrinsic parameters of the camera and do not change normally, M ₂ In the case of a two-dimensional image plane, where z is 0, and M is a 3 × 3 matrix, the transformation relationship may be:

the simultaneous equations (1-2) and (1-3) can obtain that the projection pixel points between the imaging planes have the mapping relationship:

wherein the content of the first and second substances,

is a homography between the query image and the matching image.

In the following, a process of determining a coordinate value of the position of the object to be positioned with respect to the x-axis in the embodiment of the present invention is described.

Assuming that the coordinate value of the object to be positioned about the x axis is located at a position between two preset coordinate points in the plane coordinate system, at this time, the specified direction is an extending direction toward the y axis, and the first coordinate axis is the y axis. Determining homography matrix H between query image and matching image _i The central point of the query image is O ═ x, y, z in the world coordinate system ^T The central point of the matched image is P in the world coordinate system ₁ ＝(x ₁ ,y ₁ ,z ₁ ) ^T The central point of the target fingerprint image is P in the world coordinate system ₂ ＝(x ₂ ,y ₂ ,z ₂ ) ^T Due to the fixed height of the camera, y in a world coordinate system is ensured ₁ ＝y ₂ With the centers of the target fingerprint image and the matching image being equidistant from the optical center of the camera, i.e. z ₁ ＝z ₂ 。

Point P obtainable from camera imaging model ₁ P in the image coordinate system ₁ ＝(X ₁ ,Y ₁ ) ^T ，P ₂ P in the image coordinate system ₂ ＝(X ₂ ,Y ₂ ) ^T In the same way, the query image center point O ═ x, y, z) ^T Corresponding to the midpoint O of the image coordinate system ₁ The coordinate transformation relationship is shown in fig. 7 (X, Y). Wherein, P ₁ ,P ₂ The three points O are collinear and lie on a straight line parallel to the X-axis in the world coordinate system and the X-axis in the image coordinate system, so that the geometric relationship available from fig. 7 is shown in fig. 8.

Wherein L is ₁ ＝x ₁ -x is P ₁ Distance of O point in X-axis direction in world coordinate System, L ₂ ＝x-x ₂ Is O, P ₂ Distance of points in the x-axis direction in the world coordinate system,/ ₁ ＝X ₁ -X is P ₁ Distance of O in the direction of the U axis in the camera coordinate system, l ₂ ＝X-X ₂ Is P ₂ O distance in the U-axis direction in the camera coordinate system. From the triangle similarity principle, we can get:

that is, the distance ratio in the image coordinate system is equal to the distance ratio in the world coordinate system, and the camera coordinate system obtains the pixel coordinate system of the final image only through translation and compression transformation, so that the distance ratio relationship is also satisfied in the pixel coordinate system. P in camera coordinate system ₁ ＝(X ₁ ,Y ₁ ) ^T 、P ₂ ＝(X ₂ ,Y ₂ ) ^T And O ₁ P in the corresponding pixel coordinate system ═ X, Y ₁ ＝(U ₁ ,V ₁ ) ^T 、P ₂ ＝(U ₂ ,V ₂ ) ^T And O ₁ (U, V) and (t) ₁ ＝U ₁ -U is P ₁ O is the distance in the direction of the U axis in the pixel coordinate system,

is O, P ₂ The distance in the direction of the U axis in the pixel coordinate system, then there is a relationship:

from the formulas (1-6), the difference ratio of the pixels in the pixel coordinate system can reflect the difference ratio of the distances in the world coordinate system, and the point P ₁ ,P ₂ Abscissa x of ₁ ,x ₂ And its transformation to P in the pixel coordinate system ₁ ,P ₂ Pixel difference t of point in X-axis direction ₁ ,t ₂ The coordinate value of the x axis to be positioned in the world coordinate system is:

in addition, the process of determining the coordinate value of the position of the object to be positioned with respect to the y axis is similar to the above process, and is not described herein again.

For example, the following describes in detail a process of determining a third distance by using the homography matrix of the matching image and the query image, and determining whether the third distance is greater than a preset threshold:

the abscissa of the center point of the query image is taken as X _o Taking the abscissa of the center point of the matching image as X _i And respectively taking the abscissa corresponding to the two reference fingerprint images adjacent to the matched image as X _i-1 And X _i+1 Wherein i-1 and i +1 are numbers of two reference fingerprint images respectively, and a preset threshold is taken as T, and the specific steps comprise the following steps B1-B5:

b1, calculating x _o And x _i The difference between: t is t _i ＝|X _o -X _i |；

B2, determination of t _i >If not, taking the coordinate value of the second coordinate axis of the second target position corresponding to the matching image as the coordinate value of the position of the object to be positioned on the second coordinate axis;

B3，t _i >if the comparison result is not true, comparing the best matching graph with the target query image in the two reference fingerprint images respectivelyA high fingerprint image as a target fingerprint image;

b4, calculating x according to the target fingerprint image determined in the step B3, assuming that the number of the target fingerprint image is i-1 _O And x _i-1 The difference of (c): t is t _i-1 ＝|X _o -X _i-1 |；

B5, calculating the position of the object to be positioned:

wherein L is the coordinate of the object to be positioned, L _i Is the coordinate of the second target position corresponding to the matching image,/ _i-1 And the coordinates of a third target position corresponding to the target fingerprint image.

With respect to the foregoing method embodiment, an apparatus for implementing indoor positioning by using a homography matrix is provided in the embodiments of the present application, and as shown in fig. 3, the apparatus may include:

an image obtaining module 310, configured to obtain a first query image and a second query image; the first query image and the second query image are respectively images of indoor areas obtained by acquiring images towards the extending directions of an x axis and a y axis at the position of an object to be positioned; the x-axis and the y-axis are coordinate axes in a plane coordinate system constructed for the indoor area;

an image matching module 320, configured to determine, for each query image of the first query image and the second query image, an image with a highest matching degree with the query image from the plurality of fingerprint images as a matching image, by using a pre-established database corresponding to a specified direction; the database comprises image data of the fingerprint images, each fingerprint image is an image obtained by acquiring an image towards the specified direction at a target position of the indoor area in advance, and the target position is a coordinate point in the plane coordinate system; the specified direction is the direction toward which the query image is collected;

an image determining module 330, configured to determine, from the two reference fingerprint images corresponding to the matching image, a target fingerprint image with a highest matching degree with the query image; each reference fingerprint image is an image acquired at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with a first coordinate axis of the second target position, and the second target position is a target position where the matching image is acquired; the first coordinate axis is a coordinate axis parallel to the specified direction;

an image mapping module 340, configured to map the target fingerprint image and the matching image to the query image by using a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image, respectively, to obtain a first mapping image and a second mapping image corresponding to the query image, determine a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determine a second distance between a first pixel point of the second mapping image and a second pixel point of the query image, where the first pixel point is a point at a specified position in each mapping image and the second pixel point is a point at the specified position in the query image;

a position determining module 350, configured to determine, according to the first distance, the second distance, the coordinate of the second target position, and the coordinate of the third target position, a coordinate value of the position where the object to be positioned is located with respect to a second coordinate axis; the third target position is a target position where the target fingerprint image is located when being collected, and the second coordinate axis is a coordinate axis other than the first coordinate axis.

Optionally, on the basis of including the image obtaining module 310, the image matching module 320, the image determining module 330, the image mapping module 340 and the position determining module 350, as shown in fig. 4, an apparatus for implementing indoor positioning by using a homography matrix provided in an embodiment of the present application may further include:

an image determining module 360, configured to map the matching image to the query image by using a homography matrix of the matching image and the query image before the image determining module 330 determines a target fingerprint image with a highest matching degree with the query image from two reference fingerprint images corresponding to the matching image, so as to obtain a third mapping image corresponding to the query image; determining a third distance between a third pixel point of the third mapping image and a second pixel point of the query image; the third pixel point is a point at the specified position in the third mapping image; judging whether the third distance is greater than a preset threshold value, if so, determining a target fingerprint image with the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image; if not, taking the coordinate value of the second target position relative to the second coordinate axis as the coordinate value of the position of the object to be positioned relative to the second coordinate axis.

Optionally, the position determining module 350 is specifically configured to:

when the designated direction is an extension direction towards a y axis, obtaining a coordinate value of the position of the object to be positioned, which is related to the x axis, by adopting a preset first coordinate calculation formula based on the first distance, the second distance, the coordinate of the second target position and the coordinate of the third target position;

the first coordinate calculation formula includes:

the second coordinate calculation formula includes:

wherein, y _l A coordinate value about the y axis for the position of the object to be positioned; y when the absolute value of the abscissa of the third target position is smaller than the absolute value of the abscissa of the second target position ₁ Is the ordinate, y, of the third target position ₂ Is the ordinate, t, of the second target position ₁ Is the first distance, t ₂ Is the second distance; y when the absolute value of the abscissa of the third target position is larger than the absolute value of the abscissa of the second target position ₁ Is the ordinate of the second target position, y2 is the ordinate of the third target position, t ₁ Is the second distance, t ₂ Is the first distance.

An embodiment of the present invention further provides an electronic device, as shown in fig. 5, including a processor 501, a communication interface 502, a memory 503 and a communication bus 504, where the processor 501, the communication interface 502 and the memory 503 complete mutual communication through the communication bus 504,

a memory 503 for storing a computer program;

the processor 501 is configured to implement the steps of the method for implementing indoor positioning by using the homography matrix according to the embodiment of the present invention when executing the program stored in the memory 503.

The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the electronic equipment and other equipment.

The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.

In yet another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any one of the above-mentioned methods for implementing indoor positioning using homography matrices.

In yet another embodiment of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform any one of the above-mentioned methods for indoor positioning using homography matrices.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to be performed in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, 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, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

It should be noted that, in this document, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for apparatuses, devices, storage media, etc., since they are substantially similar to the method embodiments, the description is relatively simple, and reference may be made to some descriptions of the method embodiments for relevant points.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for realizing indoor positioning by using a homography matrix is characterized by comprising the following steps:

acquiring a first query image and a second query image; the first query image and the second query image are respectively images of indoor areas obtained by acquiring images towards the extending directions of an x axis and a y axis at the position of an object to be positioned; the x-axis and the y-axis are coordinate axes in a plane coordinate system constructed for the indoor area;

aiming at each query image in the first query image and the second query image, determining an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image by utilizing a pre-constructed database corresponding to a specified direction; the database comprises image data of the fingerprint images, each fingerprint image is an image obtained by acquiring an image towards the specified direction at a target position of the indoor area in advance, and the target position is a coordinate point in the plane coordinate system; the specified direction is the direction toward which the query image is acquired;

respectively mapping the target fingerprint image and the matching image to the query image by utilizing a homography matrix between the target fingerprint image and the query image and a homography matrix between the matching image and the query image to obtain a first mapping image and a second mapping image corresponding to the query image, determining a first distance between a first pixel point of the first mapping image and a second pixel point of the query image, and determining a second distance between the first pixel point of the second mapping image and the second pixel point of the query image, wherein the first pixel point is a point at a specified position in each mapping image, and the second pixel point is a point at the specified position in the query image;

2. The method according to claim 1, wherein before determining the target fingerprint image having the highest degree of matching with the query image from the two reference fingerprint images corresponding to the matching image, the method further comprises:

3. The method according to claim 1 or 2, wherein when the specified direction is an extending direction toward a y-axis, the determining a coordinate value of the position where the object to be positioned is located with respect to a second coordinate axis by using the first distance, the second distance, the coordinate of the second target position, and the coordinate of a third target position comprises:

based on the first distance, the second distance, the coordinates of the second target position and the coordinates of the third target position, a preset first coordinate calculation formula is adopted to obtain coordinate values of the position of the object to be positioned, wherein the coordinate values are related to the x axis;

the first coordinate calculation formula includes:

wherein x is _l Coordinate values about the x-axis for the location of the object to be positioned; x when the absolute value of the abscissa of the third target position is smaller than the absolute value of the abscissa of the second target position ₁ Is the abscissa, x, of the third target position ₂ Is the abscissa, t, of the second target position ₁ Is the first distance, t ₂ Is the second distance; x when the absolute value of the abscissa of the third target position is larger than the absolute value of the abscissa of the second target position ₁ Is the abscissa, x, of the second target position ₂ Is the abscissa, t, of the third target position ₁ As a result of the second distance being said,t ₂ is the first distance;

the second coordinate calculation formula includes:

wherein, y _l Coordinate values about the y-axis for the position of the object to be positioned; y when the absolute value of the abscissa of the third target position is smaller than the absolute value of the abscissa of the second target position ₁ Is the ordinate, y, of the third target position ₂ Is the ordinate, t, of the second target position ₁ Is the first distance, t ₂ Is the second distance; when the absolute value of the abscissa of the third target position is larger than the absolute value of the abscissa of the second target position, ₁ is the ordinate, y, of the second target position ₂ Is the ordinate, t, of the third target position ₁ Is the second distance, t ₂ Is the first distance.

4. The method according to claim 1 or 2, wherein the plurality of fingerprint images belong to a plurality of image groups, each fingerprint image having a unique number; for each image group, the coordinates of the target positions corresponding to all the fingerprint images in the image group on the first coordinate axis are the same, and the serial number of each fingerprint image in the image group is increased or decreased according to the coordinates of the second coordinate axis;

5. An apparatus for implementing indoor positioning by using homography matrix, comprising:

the image matching module is used for determining an image with the highest matching degree with the query image from a plurality of fingerprint images as a matching image by utilizing a database which is constructed in advance and corresponds to a specified direction aiming at each query image in the first query image and the second query image; the database comprises image data of the fingerprint images, each fingerprint image is an image obtained by acquiring an image towards the specified direction at a target position of the indoor area in advance, and the target position is a coordinate point in the plane coordinate system; the specified direction is the direction toward which the query image is acquired;

the image determining module is used for determining a target fingerprint image which has the highest matching degree with the query image from the two reference fingerprint images corresponding to the matching image; each reference fingerprint image is an image collected at a first target position, the first target position is a target position which is adjacent to a second target position and has the same coordinate with a first coordinate axis of the second target position, and the second target position is a target position where the matching image is collected; the first coordinate axis is a coordinate axis parallel to the specified direction;

the position determining module is used for determining coordinate values of the position where the object to be positioned is located relative to a second coordinate axis, wherein the coordinate values are the first distance, the second distance, the coordinate of the second target position and the coordinate of a third target position; the third target position is a target position where the target fingerprint image is located when being collected, and the second coordinate axis is a coordinate axis other than the first coordinate axis.

6. The apparatus of claim 5, further comprising:

judging whether the third distance is greater than a preset threshold value, if so, determining a target fingerprint image with the highest matching degree with the query image from two reference fingerprint images corresponding to the matching image;

7. The apparatus of claim 5 or 6, wherein the location determination module is specifically configured to:

the first coordinate calculation formula includes:

the second coordinate calculation formula includes:

8. The apparatus according to claim 5 or 6, wherein the plurality of fingerprint images belong to a plurality of image groups, each fingerprint image having a unique number; for each image group, the coordinates of the target positions corresponding to all the fingerprint images in the image group on the first coordinate axis are the same, and the serial number of each fingerprint image in the image group is increased or decreased according to the coordinates of a second coordinate axis, wherein the second coordinate axis is a coordinate axis except the first coordinate axis;

9. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;

a memory for storing a computer program;

a processor for implementing the method steps of any of claims 1 to 4 when executing a program stored in the memory.

10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of claims 1 to 4.