CN107481287A - It is a kind of based on the object positioning and orientation method and system identified more - Google Patents

Abstract

The present invention relates to a method and system for positioning and attitude determination of an object based on multiple marks. The method comprises the following steps: setting a plurality of identifiable marks different from each other in the motion space of the object to be positioned; The camera device shoots an image containing at least one logo; identifies, processes and screens all the logos in the image to obtain a unique target logo, and determines the position information of the target logo in the image coordinate system; decodes the target logo to obtain the target logo The identification number; determine the identification coordinates of the target identification in the world coordinate system according to the number of the target identification; determine the pose of the object to be positioned according to the position information and identification coordinates. The multi-marker-based object positioning and attitude determination method and system provided by the present invention realize the large-scale positioning and attitude determination of the object to be positioned, and have stronger practicability and lower cost.

Description

A method and system for object positioning and attitude determination based on multiple markers

technical field

The invention relates to the field of computer vision/robot navigation and positioning, in particular to a multi-marker-based object positioning and attitude determination method and system.

Background technique

With the rapid development of positioning technology and LBS (Location-Based Service, location-based service), the demand for positioning has reached a new height, and location-based services are widely used in navigation, tracking and tour guides. The key technology of LBS is navigation positioning, including outdoor positioning, indoor positioning and visual positioning.

Among them, visual positioning is a method of pose measurement using computer vision to recognize artificially set marks and natural feature marks, and is widely used in robot systems, moving body control systems and precision detection systems. At present, the visual positioning technology has the following disadvantages: the range of positioning and attitude determination is small, and the accuracy is not high enough. During the movement of the robot or aircraft, accurate positioning and attitude determination cannot be achieved. For the identification of artificially set signs and natural feature signs, the recognition is accurate. low rate.

Contents of the invention

The technical problem to be solved by the present invention is to provide a method and system for positioning and attitude determination of an object based on multiple markers in view of the deficiencies in the prior art.

The technical scheme that the present invention solves the problems of the technologies described above is as follows:

A method for positioning and attitude determination of an object based on multiple identifications, comprising the following steps:

Step 1, setting multiple identifiable signs different from each other in the motion space of the object to be positioned;

Step 2, taking an image containing at least one of the logos by a camera device arranged on the object to be positioned;

Step 3, identifying, processing and screening all the marks in the image to obtain a unique target mark, and determining the position information of the target mark in the image coordinate system;

Step 4, decoding the target identifier to obtain the number of the target identifier;

Step 5, determining the coordinates of the target mark in the world coordinate system according to the number of the target mark;

Step 6: Determine the pose of the object to be positioned according to the location information and the identification coordinates.

The beneficial effects of the present invention are: the present invention provides a method for positioning and attitude determination of an object based on multiple marks, by setting a plurality of identifiable marks different from each other in the motion space of the object to be positioned, and identifying these marks by the object to be positioned , process these marks, determine the pose of the object to be positioned, and realize the large-scale positioning and pose determination of the object to be positioned, which has stronger practicability and lower cost.

On the basis of the above technical solutions, the present invention can also be improved as follows.

Further, in step 3, specifically include:

Step 3.1, identify the image according to the prestored opencv identification algorithm, obtain the binary contour map of the image, and determine the vertices of each contour in the image;

Step 3.2, processing the image according to the transformation matrix obtained from the vertex, to obtain the identification to be determined;

Step 3.3, judging the number of the identifiers to be determined, when the number of identifiers to be determined is equal to 1, using the identifiers to be determined as target identifiers; when the number of identifiers to be determined is greater than 1, judging the The contour area to be identified;

Step 3.4, when the contour areas of the to-be-determined marks are different, use the to-be-determined mark with the largest contour area as the target mark; when the contour areas of the to-be-determined marks are the same, judge the to-be-determined marks The distance between the midpoint of the image and the principal point of the image, select the identification to be determined closest to the principal point of the image as the target identification;

Step 3.5, respectively determining the vertex coordinates of each vertex of the target mark in the image coordinate system to obtain the position information of the target mark in the image coordinate system.

Further, in step 3.1, specifically include:

Step 3.1.1, performing grayscale processing and binarization processing on the image in turn to obtain a binary image, and denoising the binary image;

Step 3.1.2, extracting the contour of the binary image according to a preset contour extraction algorithm;

Step 3.1.3, removing contours whose area is smaller than the contour threshold according to the preset contour threshold;

Step 3.1.4, performing polygonal approximation on the retained outline to obtain a polygonal outline;

Step 3.1.5, judging whether the polygonal outline is a convex polygon, and removing polygonal outlines that are not convex polygons;

Step 3.1.6, extracting the retained vertices of the polygonal outline.

Further, in step 6, specifically include:

Step 6.1: Determine the pose of the target marker in the world coordinate system relative to the image coordinate system according to the position information, the marker coordinates, and a preset pose calculation algorithm;

Step 6.2: Determine the pose of the image coordinate system relative to the world coordinate system according to a preset inversion operation, and obtain the pose of the object to be positioned.

Further, the object positioning and attitude determination method also includes:

Step 7, when the object to be positioned translates or rotates, re-execute steps 2 to 6 to determine the pose of the object to be positioned.

The beneficial effect of the above further solution is: when the object to be positioned is moving, by identifying a plurality of identifiable marks that are different from each other, it is realized that even if the object to be positioned is moving in translation or rotating, the object to be positioned can be positioned and fixed in a timely and accurate manner , not only achieves three-dimensional position positioning, but also obtains deflection angle information in three-dimensional directions to realize six-degree-of-freedom pose measurement, which improves practicability and expands the range of positioning and attitude determination.

Another kind of technical scheme that the present invention solves the problems of the technologies described above is as follows:

A multi-marker-based object positioning and attitude determination system, including: multiple identifiable signs that are different from each other, and an object to be positioned, wherein:

The mark is set in the movement space of the object to be positioned;

The objects to be positioned include:

a camera device configured to capture an image containing at least one of said markers;

a processor, configured to identify, process and screen all the markers in the image to obtain a unique target marker, determine the position information of the target marker in the image coordinate system, and decode the target marker, Obtain the number of the target mark, and determine the mark coordinates of the target mark in the world coordinate system according to the number of the target mark, and determine the pose of the object to be positioned according to the position information and the mark coordinates .

Further, the processor specifically includes:

The image recognition unit is used to recognize the image according to the prestored opencv recognition algorithm, obtain the binary contour map of the image, and determine the vertices of each contour in the image;

An image processing unit, configured to process the image according to the transformation matrix obtained from the vertex, to obtain the identifier to be determined;

A judging unit, configured to judge the number of the identifiers to be determined, when the number of the identifiers to be determined is equal to 1, use the identifiers to be determined as target identifiers; when the number of identifiers to be determined is greater than 1, judge each The contour area of the logo to be determined;

When the contour areas of the marks to be determined are different, the mark to be determined with the largest contour area is used as the target mark; when the contour areas of the marks to be determined are the same, determine the midpoint of the marks to be determined The distance between the principal point of the image and the principal point of the image, select the identification to be determined closest to the principal point of the image as the target identification;

The coordinate determining unit is configured to respectively determine the vertex coordinates of each vertex of the target mark in the image coordinate system, and obtain the position information of the target mark in the image coordinate system.

Further, the image recognition unit is specifically configured to sequentially perform grayscale processing and binarization processing on the image to obtain a binary image, and denoise the binary image, and perform contour extraction according to a preset algorithm. extracting the contour of the binary image, and removing contours whose area is smaller than the contour threshold according to a preset contour threshold, and performing polygonal approximation on the retained contour to obtain a polygonal contour, and judging whether the polygonal contour is Convex polygons, remove polygon outlines that are not convex polygons, and extract vertices of the polygon outlines that remain.

Further, the processor also includes:

A calculation unit, configured to determine the pose of the target marker in the world coordinate system relative to the image coordinate system according to the position information, the marker coordinates, and a preset pose calculation algorithm, and determine the pose according to the preset The inverse operation provided determines the pose of the image coordinate system relative to the world coordinate system to obtain the pose of the object to be positioned.

Further, the processor is further configured to re-determine the pose of the object to be positioned through the processor when the object to be positioned translates or rotates.

Advantages of additional aspects of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.

Description of drawings

FIG. 1 is a schematic flowchart of a multi-marker-based object positioning and attitude determination method provided by an embodiment of the present invention;

Fig. 2 is a flow chart of a multi-marker-based object positioning and attitude determination method provided by another embodiment of the present invention;

FIG. 3 is a schematic diagram of the spatial structure of a multi-marker-based object positioning and attitude determination system provided by another embodiment of the present invention;

Fig. 4 is a schematic structural diagram of an object to be positioned in a multi-marker-based object positioning and attitude determination system according to another embodiment of the present invention.

detailed description

The principles and features of the present invention are described below in conjunction with the accompanying drawings, and the examples given are only used to explain the present invention, and are not intended to limit the scope of the present invention.

As shown in FIG. 1 , it is a schematic flow chart of a multi-label-based object positioning and attitude determination method provided by an embodiment of the present invention. The method includes the following steps:

S1, set multiple identifiable marks different from each other in the motion space of the object to be positioned. The marks are for the identification of the object to be positioned. These marks have specific identifiable features, and the identifiable features can be texture pattern features Or pattern features composed of feature points. The pattern features of these marks are different from each other to distinguish each mark. These pattern features can be converted into specific digital numbers, and the objects to be positioned are distinguished by digital numbers.

The identifiable features of each sign may include or be transformed into a set of feature points with at least 4 numbers. This set of feature points should be convenient for determining the position information of the sign relative to the world coordinate system. For example, the sign may be spray-painted There is a black border around the outer edge of the print, and the inside is a combination of black and white squares. The combination of black and white squares inside each logo pattern can be decoded by Hamming code to obtain the corresponding number of the logo. For example, you can choose 5 inside The grid of *5 excludes the repeated texture after rotation, which can be combined to correspond to 1204 numbers.

Preferably, the mark can also be directly selected on the surface of the reference object around the motion space of the object to be positioned.

These identifications also need to meet the following conditions to ensure that at least one complete identification can be obtained at different positions of the object to be positioned in the motion space.

A method for arranging signs is given below. Select a spatial reference point, that is, the origin of the world coordinate system, and fix the sign in a space that is convenient for measuring the three-dimensional positions of its four rectangular vertices. You can also use a flat plate or a bracket to fix the sign , the density of these marks should ensure that at least one mark appears in the shooting field of view of the camera device moving with the object to be positioned.

S2, taking an image containing at least one logo by using a camera device arranged on the object to be positioned, where the camera device refers to a visual sensor, usually an optical camera can be used to collect image data in real time at a preset frequency.

It should be noted that before shooting, the camera device needs to be calibrated. For example, the camera device is calibrated using the checkerboard calibration method of "Zhang Zhengyou Calibration" to obtain the focal lengths f _x , f _y , offsets C _x , C _y The internal parameters and distortion parameters of the camera device are used to determine the imaging mathematical model of the camera device, and then the internal parameters and distortion parameters of the camera device are saved for pose calculation and call.

S3, identify, process and screen all the markers in the image to obtain a unique target marker, and determine the position information of the target marker in the image coordinate system.

It should be noted that the captured image may contain images of multiple logos or other disturbances. This requires image processing to select a logo that meets the requirements. For example, it can be identified from the image through the opencv recognition algorithm Find out all possible identifications, and then measure and calibrate the identifications, for example, select a unique target identification based on the Hamming distance.

S4, decoding the target identifier to obtain the number of the target identifier.

For example, when the logo is a pattern composed of 5*5 black and white squares, the number of the logo can be obtained through Hamming distance decoding.

S5, determine the mark coordinates of the target mark in the world coordinate system according to the number of the target mark, the coordinates of each mark in the world coordinate system can be set in advance, pre-stored in the object to be positioned, when the object to be positioned recognizes a certain mark , you can get the coordinates of the logo.

S6. Determine the pose of the object to be positioned according to the position information and the identification coordinates.

For example, one of the pose calculation algorithms provided by opencv can be used to solve the PNP problem (perspective-n-point, multi-point perspective) by iteratively finding the solution with the smallest reprojection error as the optimal solution of the problem. According to the world coordinates of the identified logo, the pose of the world coordinate system where the logo is located relative to the camera coordinate system is solved, and then the pose of the camera coordinate system in the world coordinate system is solved by inverse operation, and the pose data is combined with the camera device to The pose of the object to be positioned is calculated, that is, the pose of the object to be positioned is calculated.

This embodiment provides a multi-marker-based object positioning and attitude determination method, by setting multiple identifiable marks that are different from each other in the motion space of the object to be positioned, and identifying these marks by the object to be positioned, and processing these marks , to determine the pose of the object to be positioned, and realize the positioning and pose determination of the object to be positioned in a large range, with stronger practicability and lower cost.

As shown in FIG. 2 , it is a flow chart of a multi-label-based object positioning and attitude determination method provided by another embodiment of the present invention. The method will be described in detail below. The method includes the following steps:

S1, set multiple identifiable marks different from each other in the motion space of the object to be positioned. The marks are for the identification of the object to be positioned. These marks have specific identifiable features, and the identifiable features can be texture pattern features Or pattern features composed of feature points. The pattern features of these marks are different from each other to distinguish each mark. These pattern features can be converted into specific digital numbers, and the objects to be positioned are distinguished by digital numbers.

The identifiable features of each sign may include or be transformed into a set of feature points with at least 4 numbers. This set of feature points should be convenient for determining the position information of the sign relative to the world coordinate system. For example, the sign may be spray-painted There is a black border around the outer edge of the print, and the inside is a combination of black and white squares. The combination of black and white squares inside each logo pattern can be decoded by Hamming code to obtain the corresponding number of the logo. For example, you can choose 5 inside The grid of *5 excludes the repeated texture after rotation, which can be combined to correspond to 1204 numbers.

Preferably, the mark can also be directly selected on the surface of the reference object around the motion space of the object to be positioned.

These identifications also need to meet the following conditions to ensure that at least one complete identification can be obtained at different positions of the object to be positioned in the motion space.

A method for arranging signs is given below. Select a spatial reference point, that is, the origin of the world coordinate system, and fix the sign in a space that is convenient for measuring the three-dimensional positions of its four rectangular vertices. You can also use a flat plate or a bracket to fix the sign , the density of these marks should ensure that at least one mark appears in the shooting field of view of the camera device moving with the object to be positioned.

S2, taking an image containing at least one logo by using a camera device arranged on the object to be positioned, where the camera device refers to a visual sensor, usually an optical camera can be used to collect image data in real time at a preset frequency.

It should be noted that before shooting, the camera device needs to be calibrated. For example, the camera device is calibrated using the checkerboard calibration method of "Zhang Zhengyou Calibration" to obtain the focal lengths f _x , f _y , offsets C _x , C _y The internal parameters and distortion parameters of the camera device are used to determine the imaging mathematical model of the camera device, and then the internal parameters and distortion parameters of the camera device are saved for pose calculation and call.

S3, identify, process and screen all the markers in the image to obtain a unique target marker, and determine the position information of the target marker in the image coordinate system.

It should be noted that the captured image may contain images of multiple logos or other disturbances. This requires image processing to select a logo that meets the requirements. For example, it can be identified from the image through the opencv recognition algorithm Find out all possible identifications, and then measure and calibrate the identifications, for example, select a unique target identification based on the Hamming distance.

Specifically, step S3 may be subdivided into the following steps.

S31. Recognize the image according to the pre-stored opencv recognition algorithm, obtain a binary contour map of the image, and determine vertices of each contour in the image. This process will be described in detail below.

After the image is acquired, grayscale processing is performed on the image in sequence. It can be understood that if the camera device is a high-speed camera that directly collects grayscale images, it is not necessary to perform grayscale processing on the image.

Then convert the obtained grayscale image into a binary image, and the binarization process can adopt an adaptive threshold segmentation algorithm, which can adapt to certain illumination changes.

Then, the morphological opening operation is performed on the obtained binary image, the binary image is denoised, and the contour of the denoised binary image is extracted according to a preset contour extraction algorithm.

Then preset the contour threshold to remove the contour whose area is smaller than the contour threshold. For example, assuming that the minimum area of the logo is s, then the contour threshold can be set to s. It is understandable that the side length of the contour can also be judged here. Remove smaller outlines.

Perform polygonal approximation on the retained contour to obtain a polygonal contour, judge whether the polygonal contour is a convex polygon, and remove the polygonal contours that are not convex polygons.

Then judge the side length of the obtained convex polygonal contour according to the preset side length threshold, and remove the contours that do not meet the side length threshold condition. For example, assuming that the minimum side length of the logo is b, then the side length threshold can be set as b, Contours that do not meet the conditions can be further removed through this step.

Finally, the vertices of the retained convex polygonal contours are extracted.

S32. Process the image according to the transformation matrix obtained from the vertices to obtain the logo to be determined. It is assumed that the logo is a square logo containing 5*5 black and white squares inside. The details will be described below.

First, obtain the transformation matrix according to the vertices of the convex polygon outline to obtain the front view of the image, perform perspective transformation on the image by the transformation matrix, and transform it into a 70*70 rectangular logo image, then perform otsu algorithm threshold segmentation on the perspective image, and then segment The post-binary contour image uses 10*10 squares to traverse the edge area to determine whether there is a black border around it. If there is, the contour is kept, and if not, it is removed. Then decode the internal black and white squares of the logo, and obtain the logo number information through Hamming distance decoding. Obtain the identification to be determined.

S33, judge the number of to-be-determined marks, when the number of to-be-determined marks is equal to 1, use the to-be-determined marks as target marks;

S34. When the contour areas of the marks to be determined are different, use the mark to be determined with the largest contour area as the target mark; the closer to the object to be located, the larger the area of the mark, so that the nearest mark can be determined as the target mark. When the contour area of each to-be-determined mark is the same, judge the distance between the midpoint of each to-be-determined mark and the image principal point of the image, and select the to-be-determined mark with the closest distance to the image principal point as the target mark.

S35, respectively determine the vertex coordinates of each vertex of the target mark in the image coordinate system, and obtain the position information of the target mark in the image coordinate system.

S4, decoding the target identifier to obtain the number of the target identifier.

S5, determine the mark coordinates of the target mark in the world coordinate system according to the number of the target mark, the coordinates of each mark in the world coordinate system can be set in advance, pre-stored in the object to be positioned, when the object to be positioned recognizes a certain mark , you can get the coordinates of the logo.

S6. Determine the pose of the object to be positioned according to the position information and the identification coordinates.

Specifically, the pose of the target marker in the world coordinate system relative to the image coordinate system is determined according to the position information, marker coordinates, and a preset pose calculation algorithm.

Determine the pose of the image coordinate system relative to the world coordinate system according to the preset inversion operation, and obtain the pose of the object to be positioned.

For example, one of the pose calculation algorithms provided by opencv can be used to solve the PNP problem by iteratively finding the solution with the smallest reprojection error as the optimal solution of the problem, and determine the target logo in the world coordinate system relative to the image coordinates pose of the system.

When the object to be positioned translates or rotates, step S2 to step S6 are re-executed to determine the pose of the object to be positioned.

Only through a single mark, the pose of the object to be positioned can only be determined when the mark appears in the camera's field of view. When the object to be positioned moves or rotates in a large range, making a single mark deviate from the field of view, it can be identified by identifying multiple marks in the field of view. Another mark is used to determine the pose of the object, and then the range of positioning and pose determination can be expanded through multiple marks.

As shown in FIG. 3 , it is a schematic diagram of the spatial structure of a multi-marker-based object positioning and attitude determination system provided by another embodiment of the present invention, including: multiple identifiable marks 10 different from each other, and an object to be positioned 20, in:

The marker 10 is arranged in the movement space of the object 20 to be positioned.

The object 20 to be positioned may be a movable robot vehicle, robot or aircraft, etc., specifically, may include:

The camera device 21 is used to take images containing at least one sign 10, and can use an optical camera to collect image data in real time at a certain frequency.

The connection mode between the camera 21 and the object to be positioned 20 can be: the bottom end of the telescopic rod is fixed on the object to be positioned 20, the top is connected to a small spherical platform, and the cloud platform is connected to the camera 21, and the camera can be adjusted by the telescopic rod. The height of 21, the rotation angle of camera 21 can be adjusted through the small spherical head, and by adjusting the height and viewing angle of camera 21, it can be ensured that at least one logo appears in the field of view of camera 21.

The processor 22 is configured to identify, process and screen all the markers 10 in the image to obtain a unique target marker, determine the position information of the target marker in the image coordinate system, and decode the target marker to obtain the target marker number, and determine the identification coordinates of the target identification in the world coordinate system according to the number of the target identification, and determine the pose of the object 20 to be positioned according to the position information and identification coordinates.

It can be seen from this embodiment that the processor 22 in the object to be positioned 20 performs complex processing on the image, and the object to be positioned 20 will be described in detail below through another embodiment.

As shown in FIG. 4 , it is a schematic structural diagram of an object to be positioned 20 of a multi-marker-based object positioning and attitude determination system provided by another embodiment of the present invention. The object to be positioned 20 includes: a camera 21 and a processor 22, Processor 22 specifically includes:

The image recognition unit 221 is used to recognize the image according to the pre-stored opencv recognition algorithm, obtain the binary contour map of the image, and determine the vertices of each contour in the image.

The image recognition unit 221 is specifically configured to sequentially perform grayscale processing and binarization processing on the image to obtain a binary image, and denoise the binary image, and extract the contour of the binary image according to a preset contour extraction algorithm, and According to the preset contour threshold, remove the contours whose area is smaller than the contour threshold, and perform polygon approximation on the retained contours to obtain polygonal contours, and judge whether the polygonal contours are convex polygons, remove non-convex polygonal polygonal contours, and extract the retained ones The vertices of the polygonal outline.

The image processing unit 222 is configured to process the image according to the transformation matrix obtained from the vertices to obtain the identifier to be determined.

The judging unit 223 is configured to judge the number of identifiers to be determined, and when the number of identifiers to be determined is equal to 1, use the identifier to be determined as the target identifier. When the number of to-be-determined markers is greater than 1, determine the contour area of each to-be-determined marker.

When the contour areas of the to-be-determined marks are different, the to-be-determined mark with the largest contour area is taken as the target mark. When the contour area of each to-be-determined mark is the same, judge the distance between the midpoint of each to-be-determined mark and the image principal point of the image, and select the to-be-determined mark with the closest distance to the image principal point as the target mark.

The coordinate determining unit 224 is configured to respectively determine the vertex coordinates of each vertex of the target mark in the image coordinate system, and obtain the position information of the target mark in the image coordinate system.

The decoding unit 225 decodes the target mark to obtain the number of the target mark, and determines the mark coordinates of the target mark in the world coordinate system according to the number of the target mark.

The calculation unit 226 is used to determine the pose of the target marker in the world coordinate system relative to the image coordinate system according to the position information, the marker coordinates and the preset pose calculation algorithm, and determine the image coordinate system according to the preset inversion operation Relative to the pose of the world coordinate system, the pose of the object to be positioned 20 is obtained.

Preferably, the processor 22 is also used to re-determine the pose of the object to be positioned 20 through the processor 22 when the object to be positioned 20 translates or rotates.

Readers should understand that in the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" mean that the embodiments or examples are combined A particular feature, structure, material, or characteristic is described as included in at least one embodiment or example of the invention. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the described specific features, structures, materials or characteristics may be combined in any suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine different embodiments or examples and features of different embodiments or examples described in this specification without conflicting with each other.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described devices and units can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed devices and methods may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated. to another system, or some features may be ignored, or not implemented.

A unit described as a separate component may or may not be physically separated, and a component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment of the present invention.

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

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of software products, and the computer software products are stored in a storage medium In, several instructions are included to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-OnlyMemory), random access memory (RAM, RandomAccessMemory), magnetic disk or optical disk and other media that can store program codes.

The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or modifications within the technical scope disclosed in the present invention. Replacement, these modifications or replacements shall all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (10)

1. a method for positioning and attitude determination of an object based on multiple marks, is characterized in that, comprising the following steps: Step 1, setting multiple identifiable signs different from each other in the motion space of the object to be positioned; Step 2, taking an image containing at least one of the logos by a camera device arranged on the object to be positioned; Step 3, identifying, processing and screening all the marks in the image to obtain a unique target mark, and determining the position information of the target mark in the image coordinate system; Step 4, decoding the target identifier to obtain the number of the target identifier; Step 5, determining the coordinates of the target mark in the world coordinate system according to the number of the target mark; Step 6: Determine the pose of the object to be positioned according to the location information and the identification coordinates. 2. The object positioning and attitude determination method according to claim 1, characterized in that, in step 3, specifically comprising: Step 3.1, identify the image according to the prestored opencv identification algorithm, obtain the binary contour map of the image, and determine the vertices of each contour in the image; Step 3.2, processing the image according to the transformation matrix obtained from the vertex, to obtain the identification to be determined; Step 3.3, judging the number of the identifiers to be determined, when the number of identifiers to be determined is equal to 1, using the identifiers to be determined as target identifiers; when the number of identifiers to be determined is greater than 1, judging the The contour area to be identified; Step 3.4, when the contour areas of the to-be-determined marks are different, use the to-be-determined mark with the largest contour area as the target mark; when the contour areas of the to-be-determined marks are the same, judge the to-be-determined marks The distance between the midpoint of the image and the principal point of the image, select the identification to be determined closest to the principal point of the image as the target identification; Step 3.5, respectively determining the vertex coordinates of each vertex of the target mark in the image coordinate system to obtain the position information of the target mark in the image coordinate system. 3. The object positioning and attitude determination method according to claim 2, characterized in that, in step 3.1, specifically comprising: Step 3.1.1, performing grayscale processing and binarization processing on the image in turn to obtain a binary image, and denoising the binary image; Step 3.1.2, extracting the contour of the binary image according to a preset contour extraction algorithm; Step 3.1.3, removing contours whose area is smaller than the contour threshold according to the preset contour threshold; Step 3.1.4, performing polygonal approximation on the retained outline to obtain a polygonal outline; Step 3.1.5, judging whether the polygonal outline is a convex polygon, and removing polygonal outlines that are not convex polygons; Step 3.1.6, extracting the retained vertices of the polygonal outline. 4. The object positioning and attitude determination method according to any one of claims 1 to 3, characterized in that, in step 6, specifically comprising: Step 6.1: Determine the pose of the target marker in the world coordinate system relative to the image coordinate system according to the position information, the marker coordinates, and a preset pose calculation algorithm; Step 6.2: Determine the pose of the image coordinate system relative to the world coordinate system according to a preset inversion operation, and obtain the pose of the object to be positioned. 5. The object positioning and attitude determination method according to claim 4, further comprising: Step 7, when the object to be positioned translates or rotates, re-execute steps 2 to 6 to determine the pose of the object to be positioned. 6. An object positioning and attitude determination system based on multiple identifications, characterized in that it includes: a plurality of mutually different identifiable identifications, and an object to be located, wherein: The mark is set in the movement space of the object to be positioned; The objects to be positioned include: a camera device configured to capture an image containing at least one of said markers; a processor, configured to identify, process and screen all the markers in the image to obtain a unique target marker, determine the position information of the target marker in the image coordinate system, and decode the target marker, Obtain the number of the target mark, and determine the mark coordinates of the target mark in the world coordinate system according to the number of the target mark, and determine the pose of the object to be positioned according to the position information and the mark coordinates . 7. The object positioning and attitude determination system according to claim 6, wherein the processor specifically includes: The image recognition unit is used to recognize the image according to the prestored opencv recognition algorithm, obtain the binary contour map of the image, and determine the vertices of each contour in the image; An image processing unit, configured to process the image according to the transformation matrix obtained from the vertex, to obtain the identifier to be determined; A judging unit, configured to judge the number of the identifiers to be determined, when the number of the identifiers to be determined is equal to 1, use the identifiers to be determined as target identifiers; when the number of identifiers to be determined is greater than 1, judge each The contour area of the logo to be determined; When the contour areas of the marks to be determined are different, the mark to be determined with the largest contour area is used as the target mark; when the contour areas of the marks to be determined are the same, determine the midpoint of the marks to be determined The distance between the principal point of the image and the principal point of the image, select the identification to be determined closest to the principal point of the image as the target identification; The coordinate determining unit is configured to respectively determine the vertex coordinates of each vertex of the target mark in the image coordinate system, and obtain the position information of the target mark in the image coordinate system. 8. The object positioning and attitude determination system according to claim 7, wherein the image recognition unit is specifically configured to sequentially perform grayscale processing and binarization processing on the image to obtain a binary image, and The binary image is denoised, and the contour of the binary image is extracted according to a preset contour extraction algorithm, and according to a preset contour threshold, contours whose area is smaller than the contour threshold are removed, and the retained contours are Performing polygonal approximation to obtain a polygonal outline, judging whether the polygonal outline is a convex polygon, removing polygonal outlines that are not convex polygons, and extracting the retained vertices of the polygonal outline. 9. The object positioning and attitude determination system according to any one of claims 6 to 8, wherein the processor further comprises: A calculation unit, configured to determine the pose of the target marker in the world coordinate system relative to the image coordinate system according to the position information, the marker coordinates, and a preset pose calculation algorithm, and determine the pose according to the preset The inverse operation provided determines the pose of the image coordinate system relative to the world coordinate system to obtain the pose of the object to be positioned. 10. The object positioning and attitude determination system according to claim 9, wherein the processor is further used to re-determine the object to be positioned by the processor when the object to be positioned translates or rotates pose.