阅读说明：本技术 一种基于ar智能眼镜的图像识别系统 (Image recognition system based on AR intelligent glasses ) 是由 张晓峰 王青峰 邱小锋 夏志龙 于 2021-09-17 设计创作，主要内容包括：本发明公开了一种基于AR智能眼镜的图像识别系统,属于AR智能领域,包括红外捕捉模块、图像获取模块、图像预处理模块、第一特征获取模块、第二特征获取模块、存储模块、数据库、对比模块、搜索导入模块、GPS定位模块、线路规划模块、尺寸获取模块、判断模块、画面生成模块和显示模块,其中,所述GPS定位模块与线路规划模块通讯连接；所述红外捕捉模块与图像获取模块通讯连接,本发明能够快速识别图像,降低了AR智能眼镜的卡顿,提高了用户的使用体感,且能够精确用户位置,并清晰直观的指引用户。(The invention discloses an image recognition system based on AR intelligent glasses, which belongs to the field of AR intelligence and comprises an infrared capture module, an image acquisition module, an image preprocessing module, a first characteristic acquisition module, a second characteristic acquisition module, a storage module, a database, a comparison module, a search import module, a GPS positioning module, a line planning module, a size acquisition module, a judgment module, a picture generation module and a display module, wherein the GPS positioning module is in communication connection with the line planning module; the infrared capturing module is in communication connection with the image acquisition module, so that the image can be quickly identified, the blocking of AR intelligent glasses is reduced, the use feeling of a user is improved, the position of the user can be accurately obtained, and the user can be clearly and intuitively guided.)

1. An image recognition system based on AR intelligent glasses is characterized by comprising an infrared capturing module, an image acquiring module, an image preprocessing module, a first feature acquiring module, a second feature acquiring module, a storage module, a database, a comparison module, a search importing module, a GPS positioning module, a line planning module, a size acquiring module, a judging module, a picture generating module and a display module;

the GPS positioning module is in communication connection with the circuit planning module; the infrared capturing module is in communication connection with the image acquisition module.

2. The image recognition system based on AR smart glasses according to claim 1, wherein the infrared capturing module is configured to capture a degree of pupillary zoom of the user, generate a real gaze point, and send the gaze point to the image obtaining module; the image acquisition module is used for acquiring a depth-of-field image and sending the depth-of-field image to the image preprocessing module and the size acquisition module; the image preprocessing module is used for processing the depth-of-field image; the specific operation of processing the depth-of-field image is as follows:

s1, converting the depth-of-field image into image frame data through an analog-to-digital converter;

s2, labeling the image frame data to form labeled frame data;

s3, in the process of the step S2, the labeled frame data after labeling immediately judges whether the label is an odd number, if so, the labeled frame data is sent to the first feature acquisition module, and if not, the labeled frame data is sent to the second feature acquisition module;

the first characteristic acquisition module is used for acquiring and sending the acquired information to the storage module; and the second characteristic acquisition module is used for acquiring special matters in the label frame data and sending the special matters to the storage module.

3. The AR smart glasses-based image recognition system of claim 2, wherein the special things include landmark buildings and road names; the real fixation point is generated by adopting an eyeball tracking technology; the obtained depth of field image is obtained by shooting with a depth of field shooting device.

4. The AR smart glasses-based image recognition system of claim 1, wherein the storage module is configured to store a special thing and send the special thing to the comparison module; the database is used for storing map data and sending the map data to the comparison module; the comparison module is used for comparing special things with map data and generating an actual position; the specific operation of generating the actual position is as follows:

SS1 for calculating special things and map dataCorrelation value, denoted as a₁、a₂、a₃……a_n；

SS2, judging whether the correlation value is qualified, specifically:

A. if the correlation value is less than 90%, the product is unqualified and is marked as E;

B. if the relevance value is more than or equal to 90 percent, the result is qualified and is marked as F;

and the SS3 reserves the map data corresponding to the F, records the map data as an actual position, and sends the actual position to the route planning module.

5. The AR smart glasses-based image recognition system of claim 4, wherein a is₁+a₂+a₃+……+a_nThe calculation method of the relevance value adopts a similarity measurement algorithm, namely a ROST SEAT algorithm.

6. The AR smart glasses-based image recognition system of claim 1, wherein the search import module is configured to import a name of a search place and send the name to the GPS location module; the GPS positioning module is used for positioning a search place, generating a target position and sending the target position to the route planning module; the route planning module is used for planning a traveling route and generating a virtual beacon; the specific process of generating the virtual beacon is as follows:

SSS1, marking actual and target locations on a map;

SSS2, avoiding a building connection target position and an actual position, and generating a preparation travel route;

SSS3, sorting the preliminary travel routes from the beginning to the end according to the lengths of the preliminary travel routes, and selecting the minimum value, wherein the preliminary travel route corresponding to the minimum value is the travel route;

SSS4, generating a beacon direction according to the traveling route, further generating a virtual beacon, and sending the virtual beacon to the size acquisition module.

7. The image recognition system based on the AR smart glasses according to claim 1, wherein the size obtaining module is configured to obtain a virtual pointing target size, obtain an actual object size in the depth image, and send the virtual pointing target size and the actual object size to the judging module; the judging module is used for judging the similarity between the size of the virtual pointing object and the size of the actual object and marking the actual object according to the similarity; the picture generation module generates a playing picture according to the actual object marked by the virtual guide mark, and sends the playing picture to the display module; the display module is used for displaying a playing picture.

Technical Field

The invention relates to the field of AR intelligence, in particular to an image recognition system based on AR intelligent glasses.

Background

Image recognition, which refers to a technology for processing, analyzing and understanding an image by using a computer to recognize targets and objects in various modes, is generally divided into two modules, namely face recognition and object recognition, and AR smart glasses are smart devices for applying virtual objects to real scenes by using an augmented reality technology.

Through retrieval, chinese patent No. CN202110004483.2 discloses an image recognition system based on AR smart glasses, which solves some problems of image recognition, but the rate of image recognition is still slow, thereby resulting in the stuttering of AR smart glasses, reducing the use feeling of AR smart glasses, and general AR smart glasses do not have a navigation function, while general navigation software is not accurate enough in positioning the user position, and also cannot clearly and intuitively guide the user, so the navigation function of AR smart glasses is very urgent, and a system defect is caused.

Disclosure of Invention

The invention aims to solve the defects in the prior art, and provides an image recognition system based on AR intelligent glasses.

In order to achieve the purpose, the invention adopts the following technical scheme:

an image recognition system based on AR intelligent glasses comprises an infrared capture module, an image acquisition module, an image preprocessing module, a first characteristic acquisition module, a second characteristic acquisition module, a storage module, a database, a comparison module, a search import module, a GPS positioning module, a line planning module, a size acquisition module, a judgment module, a picture generation module and a display module;

the GPS positioning module is in communication connection with the circuit planning module; the infrared capturing module is in communication connection with the image acquisition module.

Furthermore, the infrared capturing module is used for capturing the pupil zooming degree of the user, generating a real fixation point and sending the real fixation point to the image acquisition module; the image acquisition module is used for acquiring a depth-of-field image and sending the depth-of-field image to the image preprocessing module and the size acquisition module; the image preprocessing module is used for processing the depth-of-field image; the specific operation of processing the depth-of-field image is as follows:

s1, converting the depth-of-field image into image frame data through an analog-to-digital converter;

s2, labeling the image frame data to form labeled frame data;

s3, in the process of the step S2, the labeled frame data after labeling immediately judges whether the label is an odd number, if so, the labeled frame data is sent to the first feature acquisition module, and if not, the labeled frame data is sent to the second feature acquisition module;

the first characteristic acquisition module is used for acquiring special matters in the label frame data and sending the special matters to the storage module; and the second characteristic acquisition module is used for acquiring special matters in the label frame data and sending the special matters to the storage module.

Further, the special things include landmark buildings and road names; the real fixation point is generated by adopting an eyeball tracking technology; the obtained depth of field image is obtained by shooting with a depth of field shooting device.

Further, the storage module is used for storing special things and sending the special things to the comparison module; the database is used for storing map data and sending the map data to the comparison module; the comparison module is used for comparing special things with map data and generating an actual position; the specific operation of generating the actual position is as follows:

SS1, calculating the relevance value between special things and map data, and marking as a₁、a₂、a₃……a_n；

SS2, judging whether the correlation value is qualified, specifically:

A. if the correlation value is less than 90%, the product is unqualified and is marked as E;

B. if the relevance value is more than or equal to 90 percent, the result is qualified and is marked as F;

and the SS3 reserves the map data corresponding to the F, records the map data as an actual position, and sends the actual position to the route planning module.

Further, said a₁+a₂+a₃+……+a_nThe method of calculating the relevance value employs a similarity measure algorithm, i.e. ROSThe T SEAT algorithm.

Furthermore, the search import module is used for importing a search place name and sending the search place name to the GPS positioning module; the GPS positioning module is used for positioning a search place, generating a target position and sending the target position to the route planning module; the route planning module is used for planning a traveling route and generating a virtual beacon; the specific process of generating the virtual beacon is as follows:

SSS1, marking actual and target locations on a map;

SSS2, avoiding a building connection target position and an actual position, and generating a preparation travel route;

SSS3, sorting the preliminary travel routes from the beginning to the end according to the lengths of the preliminary travel routes, and selecting the minimum value, wherein the preliminary travel route corresponding to the minimum value is the travel route;

SSS4, generating a beacon direction according to the traveling route, further generating a virtual beacon, and sending the virtual beacon to the size acquisition module.

Further, the size acquisition module is used for acquiring the size of the virtual pointing object, acquiring the size of an actual object in the depth-of-field image, and sending the size of the virtual pointing object and the size of the actual object to the judgment module; the judging module is used for judging the similarity between the size of the virtual pointing object and the size of the actual object and marking the actual object according to the similarity; the picture generation module generates a playing picture according to the actual object marked by the virtual guide mark, and sends the playing picture to the display module; the display module is used for displaying a playing picture.

Compared with the prior art, the invention has the beneficial effects that:

1. according to the method, the real fixation point of the user glasses is obtained through the infrared capture module, the field depth image is shot according to the real fixation point, the image preprocessing module processes the field depth image, the field depth image is decomposed into image frame data firstly, then the image frame data is labeled to form labeled frame data, whether a label in the labeled frame data is an odd number or not is judged finally, if the label is the odd number, the labeled frame data is sent to the first feature obtaining module, and if the label is not the odd number, the labeled frame data is sent to the second feature obtaining module.

2. The invention obtains the actual position through the comparison module to achieve the purpose of accurate user position, then introduces the name of the search place through the search introduction module, determines the target position through the GPS positioning module, determines the advancing route of the user through the route planning module after the actual position and the target position are all determined, generates the virtual guide mark, determines the actual object to be covered in the depth of field image through the size acquisition module, covers the actual object with the virtual guide mark, sends the covered depth of field image to the image generation module to generate the playing image, and finally displays the playing image through the display module to achieve the purpose of guiding the user, so that the guiding is clear and intuitive.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.

Fig. 1 is a schematic diagram of the operation of the modules of an image recognition system based on AR smart glasses according to 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.

In the description of the present invention, it is to be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention.

Example 1:

referring to fig. 1, the present invention provides a technical solution: an image recognition system based on AR intelligent glasses comprises an infrared capture module, an image acquisition module, an image preprocessing module, a first characteristic acquisition module, a second characteristic acquisition module, a storage module, a database, a comparison module, a search import module, a GPS positioning module, a line planning module, a size acquisition module, a judgment module, a picture generation module and a display module;

the GPS positioning module is in communication connection with the circuit planning module; the infrared capturing module is in communication connection with the image acquisition module and is used for capturing the pupil contraction and contraction degree of a user, generating a real fixation point and sending the real fixation point to the image acquisition module; the image acquisition module is used for acquiring a depth-of-field image and sending the depth-of-field image to the image preprocessing module and the size acquisition module; the image preprocessing module is used for processing the depth-of-field image; the specific operations for processing the depth-of-field image are as follows:

firstly, converting a depth-of-field image into image frame data through an analog-to-digital converter; secondly, labeling the image frame data to form labeled frame data; finally, the labeled frame data after labeling immediately judges whether the label is an odd number, if the label is the odd number, the labeled frame data is sent to the first characteristic acquisition module, and if the label is not the odd number, the labeled frame data is sent to the second characteristic acquisition module;

the first characteristic acquisition module is used for acquiring special matters in the label frame data and sending the special matters to the storage module; the second characteristic acquisition module is used for acquiring special things in the label frame data and sending the special things to the storage module, wherein the special things comprise landmark buildings and road names; generating the real fixation point adopts an eyeball tracking technology; the obtained depth of field image is obtained by shooting with a depth of field shooting device.

Specifically, in the process of identifying the image, the infrared capture module acquires the real fixation point of the user glasses through an eyeball tracking technology, the depth-of-field camera device in the image acquisition module shoots a depth-of-field image according to the real fixation point, the depth-of-field image is processed through the image preprocessing module, the depth-of-field image is decomposed into image frame data, then the image frame data are labeled to form labeled frame data, and finally whether the label in the labeled frame data is an odd number or not is judged.

Example 2:

referring to fig. 1, the storage module is used for storing a special thing and sending the special thing to the comparison module; the database is used for storing map data and sending the map data to the comparison module; the comparison module is used for comparing special things with map data and generating an actual position; the specific operation of generating the actual position is as follows:

SS1, calculating the relevance value between special things and map data, and marking as a₁、a₂、a₃……a_n；

SS2, judging whether the correlation value is qualified, specifically:

A. if the correlation value is less than 90%, the product is unqualified and is marked as E;

B. if the relevance value is more than or equal to 90 percent, the result is qualified and is marked as F;

SS3, reserving the map data corresponding to F, recording as the actual position, and sending to the route planning module, a₁+a₂+a₃+……+a_nThe calculation method of the relevance value adopts a similarity measurement algorithm, namely a ROST SEAT algorithm, and the search import module is used for importing a search place name and sending the search place name to the GPS positioning module; the GPS positioning module is used to locate a search site,

generating a target position and sending the target position to a line planning module; the route planning module is used for planning a traveling route and generating a virtual beacon; the specific process of generating the virtual beacon comprises the following steps: marking an actual position and a target position on a map; avoiding the position of a building connection target and the actual position, and generating a prepared traveling route; according to the length of the preparation advancing route, sorting the preparation advancing route from the small to the small, and selecting the minimum value, wherein the preparation advancing route corresponding to the minimum value is the advancing route; generating a direction of a pointer according to the advancing route, further generating a virtual pointer, and sending the virtual pointer to a size acquisition module, wherein the size acquisition module is used for acquiring the size of the virtual pointer, acquiring the size of an actual object in the depth-of-field image, and sending the size of the virtual pointer and the size of the actual object to a judgment module; the judging module is used for judging the similarity between the size of the virtual pointing standard and the size of the actual object and marking the actual object according to the similarity; the picture generation module generates a playing picture according to the actual object marked by the virtual guide mark, and sends the playing picture to the display module; the display module is used for displaying a playing picture.

Specifically, in the process of guiding the user, the first characteristic acquisition module and the second characteristic acquisition module respectively acquire special objects in label frame data, store the special objects in the storage module, then the comparison module compares the special objects according to map data in the database to obtain the actual position, so as to achieve the purpose of accurately positioning the user, the user introduces the name of a search place through the search introduction module, and determines the target position through the GPS positioning module, after the actual position and the target position are all determined, the route planning module determines the traveling route of the user and generates a virtual pointing target, then the size acquisition module determines the actual object to be covered in the depth image, and the actual object is covered by the virtual pointing target, the covered depth image is sent to the image generation module to generate a playing image, and finally the display module displays the playing image to achieve the purpose of guiding the user, so that the guidance is clear and intuitive.

The working principle and the using process of the invention are as follows: when a user uses the AR intelligent glasses, the infrared capturing module acquires the real fixation point of the user glasses through an eyeball tracking technology, the depth-of-field camera in the image acquisition module shoots a depth-of-field image according to the real fixation point, the depth-of-field image is processed through the image preprocessing module, the depth-of-field image is firstly decomposed into image frame data, then the image frame data is labeled to form labeled frame data, finally whether the label in the labeled frame data is an odd number or not is judged, if the label is the odd number, the labeled frame data is sent to the first feature acquisition module, if the label is not the odd number, the labeled frame data is sent to the second feature acquisition module, in the process, the processes of labeling the image frame data and judging the labeled frame data are synchronously carried out, the aim of quickly identifying the image is achieved, the stutter of the AR intelligent glasses is reduced, the user feeling is improved, and the first feature acquisition module and the second feature acquisition module respectively acquire special matters in the labeled frame data, the special objects are stored in a storage module, the special objects are compared by a comparison module according to map data in a database, so that the actual position is obtained, the purpose of accurate user position is achieved, a user introduces a search place name through a search introduction module and determines the target position through a GPS positioning module, after the actual position and the target position are completely determined, a route planning module determines the traveling route of the user and generates a virtual pointer, a size acquisition module determines the actual object to be covered in a depth image and covers the actual object with the virtual pointer, the covered depth image is sent to a picture generation module to generate a playing picture, and finally the playing picture is displayed by a display module, so that the purpose of guiding the user is achieved, the guiding is clear and visual, and the operation is completed.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.