Nothing Special   »   [go: up one dir, main page]

CN110146083A - A Crowded Indoor Image Recognition Cloud Navigation System - Google Patents

A Crowded Indoor Image Recognition Cloud Navigation System Download PDF

Info

Publication number
CN110146083A
CN110146083A CN201910396992.7A CN201910396992A CN110146083A CN 110146083 A CN110146083 A CN 110146083A CN 201910396992 A CN201910396992 A CN 201910396992A CN 110146083 A CN110146083 A CN 110146083A
Authority
CN
China
Prior art keywords
map
image recognition
mobile terminal
photographer
navigation system
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910396992.7A
Other languages
Chinese (zh)
Inventor
花罡辰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Institute of Information Technology
Original Assignee
Shenzhen Institute of Information Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Institute of Information Technology filed Critical Shenzhen Institute of Information Technology
Priority to CN201910396992.7A priority Critical patent/CN110146083A/en
Publication of CN110146083A publication Critical patent/CN110146083A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20Instruments for performing navigational calculations
    • G01C21/206Instruments for performing navigational calculations specially adapted for indoor navigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Navigation (AREA)

Abstract

本发明属于云导航技术领域且公开了一种拥挤室内图像识别云导航系统,能够实现不需维护的自我增殖的地图构建,在构建地图后,整个地图系统收集到视觉特征点8971个,随着导航的进行,系统会根据阈值自动记录新出现的视觉特征,在进行顺时针一周的导航后,系统记录的特征点数量达到了12482个,因此系统能够自动更新地图相关知识,此时导航识别率达到87.6%,识别精度为96.1%。The invention belongs to the technical field of cloud navigation and discloses a cloud navigation system for image recognition in crowded indoor spaces, which can realize the construction of self-proliferation maps without maintenance. After the map is constructed, the entire map system collects 8971 visual feature points. During the navigation, the system will automatically record the new visual features according to the threshold. After one week of clockwise navigation, the number of feature points recorded by the system has reached 12,482, so the system can automatically update the map-related knowledge. At this time, the navigation recognition rate It reaches 87.6%, and the recognition accuracy is 96.1%.

Description

一种拥挤室内图像识别云导航系统A Crowded Indoor Image Recognition Cloud Navigation System

技术领域technical field

本发明涉及云导航技术领域,尤其涉及一种拥挤室内图像识别云导航系统。The invention relates to the technical field of cloud navigation, in particular to a cloud navigation system for image recognition in crowded indoors.

背景技术Background technique

相较于非常成熟的室外定位技术(GPS、北斗系统、A-GPS等),室内定位目前尚处于萌芽期,还没有得到广泛应用。但作为室外导航技术的重要补充,室内导航技术承担着导航定位“最后一米”的重要任务。目前室内导航的主流技术有:Compared with very mature outdoor positioning technologies (GPS, Beidou system, A-GPS, etc.), indoor positioning is still in its infancy and has not been widely used. However, as an important supplement to outdoor navigation technology, indoor navigation technology undertakes the important task of navigation and positioning "the last meter". The current mainstream technologies for indoor navigation include:

1、基于专用设备的室内定位:通过在室内指定区域部署大量的专用硬件设施(超声波,红外传感),同时使用户佩戴设备,实现定位。1. Indoor positioning based on special equipment: By deploying a large number of special hardware facilities (ultrasonic, infrared sensing) in designated indoor areas, and at the same time enabling users to wear equipment to achieve positioning.

2、基于Wi-Fi信号测距:Wi-Fi的信号强度随着距离变大而衰减。通过衰减模型和三边定位可以确定用户位置。2. Ranging based on Wi-Fi signal: The signal strength of Wi-Fi decreases as the distance increases. The user location can be determined by the attenuation model and trilateration.

3、基于Wi-Fi指纹的定位:这种方法是目前最常见的解决方案。在位置A测到路由器1、2、3、4、5的信号强度是a、b、c、d、e,那你在另一位置B测出来的强度一般是不一样的。每隔几米测一次各路由器的信号强度,作为一条数据(我们称之为Wi-Fi指纹)存到一个数据库里,对于下一次某位置上的输入值,需寻找余弦相似度最高的。毕竟室内一般离不开Wi-Fi。3. Wi-Fi fingerprint-based positioning: This method is the most common solution at present. The signal strengths of routers 1, 2, 3, 4, and 5 measured at location A are a, b, c, d, and e, then the strengths you measured at another location B are generally different. Measure the signal strength of each router every few meters, and store it in a database as a piece of data (we call it a Wi-Fi fingerprint). For the next input value at a certain location, you need to find the one with the highest cosine similarity. After all, indoors are generally inseparable from Wi-Fi.

4、基于视觉信息的定位:基于视觉信息的定位是一种室内定位的新方法,这类方法基于图像特征的匹配进行闭环检出(loop-closure detecting),特点是成本较低。4. Localization based on visual information: Localization based on visual information is a new indoor localization method. This kind of method performs loop-closure detection based on the matching of image features, which is characterized by low cost.

发明内容SUMMARY OF THE INVENTION

本发明要解决的技术问题是克服现有的缺陷,提供一种拥挤室内图像识别云导航系统,能够实现不需维护的自我增殖的地图构建,在构建地图后,整个地图系统收集到视觉特征点8971个,随着导航的进行,系统会根据阈值自动记录新出现的视觉特征,在进行顺时针一周的导航后,系统记录的特征点数量达到了12482个,因此系统能够自动更新地图相关知识,此时导航识别率达到87.6%,识别精度为96.1%,可以有效解决背景技术中的问题。The technical problem to be solved by the present invention is to overcome the existing defects and provide a cloud navigation system for image recognition in crowded indoor spaces, which can realize self-proliferation map construction without maintenance. After the map is constructed, the entire map system collects visual feature points. 8,971. As the navigation progresses, the system will automatically record the new visual features according to the threshold. After clockwise navigation for one week, the number of feature points recorded by the system has reached 12,482, so the system can automatically update the map-related knowledge. At this time, the navigation recognition rate reaches 87.6%, and the recognition accuracy is 96.1%, which can effectively solve the problems in the background technology.

为了解决上述技术问题,本发明提供了如下的技术方案:In order to solve the above-mentioned technical problems, the present invention provides the following technical solutions:

本发明提供一种拥挤室内图像识别云导航系统,包括:The present invention provides a cloud navigation system for crowded indoor image recognition, including:

移动端和服务器端;mobile and server;

移动端连接服务器端,若连接成功,服务器发送抽象地图信息至移动端,抽象地图信息包括map.csv,为:The mobile terminal connects to the server terminal. If the connection is successful, the server sends the abstract map information to the mobile terminal. The abstract map information includes map.csv, which is:

0,0,00, 0, 0

1,0,01, 0, 0

100,5,0100, 5, 0

500,5,1500,5,1

从移动端拍摄当前行人视角的一个图片序列中提取出稳定的特征点发送至服务端;Extract stable feature points from a sequence of pictures taken by the mobile terminal from the current pedestrian perspective and send them to the server;

服务器端根据移动端发送的稳定特征点信息匹配存储于地图中的特征点,得出匹配度最高并且匹配度高于根据经验设定的阈值的一张图片编号;The server matches the feature points stored in the map according to the stable feature point information sent by the mobile terminal, and obtains a picture number with the highest matching degree and a matching degree higher than the threshold set according to experience;

由于已知与拍摄者最接近的图片编号,故能根据map.csv信息得知目前拍摄者的位置与视角;Since the picture number closest to the photographer is known, the current position and perspective of the photographer can be known according to the map.csv information;

根据目前拍摄者的位置与视角的信息,综合考虑离拍摄者最近店铺的编号、目的地的编号、地图中店铺总数、拍摄者当前朝向,在移动端利用算法一计算出最短路径,给出导航建议。According to the current location and angle of view of the photographer, comprehensively consider the number of the nearest store to the photographer, the number of the destination, the total number of stores in the map, and the current orientation of the photographer, and use algorithm 1 to calculate the shortest path on the mobile terminal, and give the navigation Suggest.

作为本发明的一种优选技术方案,所述map.csv第一列为地图中的图片编号,第二列为店铺代码,第三列为该图片的朝向,As a preferred technical solution of the present invention, the first column of the map.csv is the picture number in the map, the second column is the store code, and the third column is the orientation of the picture,

其中,0为顺时针,1为逆时针,Among them, 0 is clockwise, 1 is counterclockwise,

另外一个map.csv记录了每个店铺编号代表的店铺名称。Another map.csv records the store name represented by each store number.

作为本发明的一种优选技术方案,所述移动端拍摄为间隔500ms拍摄一张,共拍摄4张,耗时2秒。As a preferred technical solution of the present invention, the mobile terminal takes one picture at an interval of 500ms, and a total of 4 pictures are taken, which takes 2 seconds.

作为本发明的一种优选技术方案,所述稳定的特征点是指排除了行人移动物体干扰的环境特征点。As a preferred technical solution of the present invention, the stable feature points refer to environmental feature points that exclude the interference of pedestrian moving objects.

作为本发明的一种优选技术方案,所述建议包括两类:As a preferred technical solution of the present invention, the suggestion includes two categories:

①沿当前路径经过n1,n2..店铺即可到达目的地;①The destination can be reached by passing through n1, n2.. shops along the current path;

②转身途经经过n1,n2..店铺即可到达目的地。②Turn around and pass through n1, n2.. shops to reach the destination.

本发明中提供的一个或多个技术方案,至少具有如下技术效果或者优点:One or more technical solutions provided in the present invention have at least the following technical effects or advantages:

1、配置简便、人力成本低,本项目在一个地点使用之前,只需操作移动简易机器人对环境学习一次即可,由于地图能够自我增殖地更新,所以一次学习后不需要后续的维护,因此本项目能大大节省室内导航系统系统维护的人力成本。1. The configuration is simple and the labor cost is low. Before this project is used in one place, it is only necessary to operate the mobile simple robot to learn the environment once. Since the map can be updated with self-proliferation, subsequent maintenance is not required after one learning. The project can greatly save the labor cost of indoor navigation system system maintenance.

2、潜在用户量巨大,在移动互联网时代,智能手机的普及率很高(全国已达到58%,我市更高),本项目不需要额外硬件支持,只需要所有智能手机均具备的摄像功能和网络功能即可,所有智能手机用户都能够享受本项目带来的便利生活,所以本项目的潜在用户量是巨大的。2. The number of potential users is huge. In the era of mobile Internet, the penetration rate of smart phones is very high (the country has reached 58%, and the city is even higher). This project does not require additional hardware support, only the camera function that all smart phones have. All smart phone users can enjoy the convenient life brought by this project, so the potential users of this project are huge.

3、能在拥挤的环境下的学习和导航:利用创新专利技术能够有效地提取动态环境中稳定的视觉特征值,进而实现在拥挤环境的学习和导航。所以本项目可以在任何时间段进行学习或导航,特别适合人员众多的动态环境。3. Learning and navigation in crowded environments: Using innovative patented technology can effectively extract stable visual feature values in dynamic environments, thereby realizing learning and navigation in crowded environments. Therefore, this project can be learned or navigated at any time, especially suitable for dynamic environments with many people.

4、云端处理,本项目创新地使用云计算技术最大限度地把繁重的计算任务转移到配置强劲的云服务器端。这样可以不需消耗智能手机过多的电力,并能实现更快的响应速度,提升用户体验。4. Cloud processing, this project innovatively uses cloud computing technology to transfer heavy computing tasks to the cloud server with powerful configuration. In this way, the smartphone does not need to consume too much power, and can achieve a faster response speed and improve the user experience.

具体实施方式Detailed ways

本申请实施例通过一种拥挤室内图像识别云导航系统解决了现有技术中存在的问题。The embodiments of the present application solve the problems existing in the prior art through a crowded indoor image recognition cloud navigation system.

为了更好地理解上述技术方案,下面具体实施方式对上述技术方案进行详细的说明。In order to better understand the above technical solutions, the following specific embodiments describe the above technical solutions in detail.

实施例一:Example 1:

本发明一种拥挤室内图像识别云导航系统,包括:The present invention is a crowded indoor image recognition cloud navigation system, comprising:

移动端和服务器端;mobile and server;

移动端连接服务器端,若连接成功,服务器发送抽象地图信息至移动端,抽象地图信息包括map.csv,为:The mobile terminal connects to the server terminal. If the connection is successful, the server sends the abstract map information to the mobile terminal. The abstract map information includes map.csv, which is:

0,0,00, 0, 0

1,0,01, 0, 0

100,5,0100, 5, 0

500,5,1500,5,1

其中map.csv第一列为地图中的图片编号,第二列为店铺代码,第三列为该图片的朝向,其中,0为顺时针,1为逆时针;The first column of map.csv is the picture number in the map, the second column is the store code, and the third column is the orientation of the picture, where 0 is clockwise and 1 is counterclockwise;

例如500,5,1 意义为云端地图里第500张图片位置在第5个店铺附近,拍照朝向为逆时针;For example, 500,5,1 means that the 500th picture in the cloud map is located near the 5th store, and the photo orientation is counterclockwise;

另外一个map.csv记录了每个店铺编号代表的店铺名称;Another map.csv records the store name represented by each store number;

利用专利技术(Osamu Hasegawa, Gangchen Hua,FEATURE VALUE EXTRACTION DEVICEAND LOCATION INFERENCE DEVICE.(PCT专利,国际知识产权组织发明专利号:WO2014073204 A1)),从移动端拍摄当前行人视角的一个图片序列中提取出稳定的特征点发送至服务端;Using patented technology (Osamu Hasegawa, Gangchen Hua, FEATURE VALUE EXTRACTION DEVICEAND LOCATION INFERENCE DEVICE. (PCT patent, International Intellectual Property Organization Invention Patent No.: WO2014073204 A1)), a stable image is extracted from a sequence of pictures taken by the mobile terminal from the current pedestrian perspective. The feature points are sent to the server;

服务器端根据移动端发送的稳定特征点信息匹配存储于地图中的特征点(匹配方法仍然基于WO2014073204 A1),得出匹配度最高并且匹配度高于根据经验设定的阈值的一张图片编号(例如500),如果得不到这个编号则返回(3)重试,移动端拍摄为间隔500ms拍摄一张,共拍摄4张,耗时2秒,稳定的特征点是指排除了行人等移动物体干扰的环境特征点;The server matches the feature points stored in the map according to the stable feature point information sent by the mobile terminal (the matching method is still based on WO2014073204 A1), and obtains a picture number with the highest matching degree and a matching degree higher than the threshold set according to experience ( For example, 500), if you can't get this number, return to (3) and try again. The mobile terminal shoots one shot at an interval of 500ms, and a total of 4 shots are shot, which takes 2 seconds. The stable feature points refer to the exclusion of moving objects such as pedestrians. Interfering environmental feature points;

由于已知与拍摄者最接近的图片编号,故能根据map.csv信息得知目前拍摄者的位置与视角;例如知道与拍摄者最接近的图片编号为500就能知道拍摄者当前位置在第5个店铺附近,朝向为逆时针;Since the picture number closest to the photographer is known, the current photographer's position and angle of view can be known according to the map.csv information; for example, if the picture number closest to the photographer is 500, it can be known that the photographer's current position is in the first Near 5 stores, the orientation is counterclockwise;

根据目前拍摄者的位置与视角的信息,综合考虑离拍摄者最近店铺的编号、目的地的编号、地图中店铺总数、拍摄者当前朝向,在移动端利用算法一计算出最短路径,给出导航建议;According to the current location and angle of view of the photographer, comprehensively consider the number of the nearest store to the photographer, the number of the destination, the total number of stores in the map, and the current orientation of the photographer, and use algorithm 1 to calculate the shortest path on the mobile terminal, and give the navigation Suggest;

建议包括两类:Recommendations include two categories:

①沿当前路径经过n1,n2..店铺即可到达目的地;①The destination can be reached by passing through n1, n2.. shops along the current path;

②转身途经经过n1,n2..店铺即可到达目的地;②Turn around and pass through n1, n2.. shops to reach the destination;

其中,算法一为:Among them, the first algorithm is:

current_index 拍摄者当前最近店铺的编号,通过根据map.csv信息得知目前拍摄者的位置与视角得到:current_index The number of the photographer's current nearest store, obtained by knowing the current photographer's position and perspective according to the map.csv information:

objective_index 目的地的编号,用户指定;objective_index the number of the destination, specified by the user;

distance objective_index-current_index 的绝对值;The absolute value of distance objective_index-current_index;

shop_amount 地图中店铺总数;shop_amount The total number of shops in the map;

orentation 用户当前朝向,通过根据map.csv信息得知目前拍摄者的位置与视角得到,0为顺时针,1为逆时针。orentation The current orientation of the user, obtained by knowing the current photographer's position and angle of view according to the map.csv information, 0 is clockwise, 1 is counterclockwise.

具体操作为:The specific operations are:

if (current_index < objective_index) {if (current_index < objective_index) {

if (distance <= shop_amount / 2) {if (distance <= shop_amount / 2) {

if(orentation == 0){提示用户沿当前路径经过n1,n2..店铺即可到达目的地} if(orentation == 0){Prompt the user to reach the destination after passing n1, n2.. stores along the current path}

else{提示用户转身途经经过n1,n2..店铺即可到达目的地}}else{Prompt the user to turn around and pass through n1, n2.. shops to reach the destination}}

else{if(orentation == 1){提示用户沿当前路径经过n1,n2..店铺即可到达目的地}else{if(orentation == 1){Prompt the user to reach the destination after passing n1, n2.. shops along the current path}

else{提示用户转身途经经过n1,n2..店铺即可到达目的地}}}else{Prompt the user to turn around and go through the n1,n2.. stores to reach the destination}}}

else{if (distance <= shop_amount / 2) {if(orentation == 1){else{if (distance <= shop_amount / 2) {if(orentation == 1){

提示用户沿当前路径经过n1,n2..店铺即可到达目的地}Prompt the user to reach the destination after passing n1, n2.. stores along the current path}

else{提示用户转身途经经过n1,n2..店铺即可到达目的地}}else{Prompt the user to turn around and pass through n1, n2.. shops to reach the destination}}

else{if(orentation == 0){提示用户沿当前路径经过n1,n2..店铺即可到达目的地}else{if(orentation == 0){Prompt the user to reach the destination after passing n1, n2.. shops along the current path}

else{提示用户转身途经经过n1,n2..店铺即可到达目的地}}}。else{Prompt the user to turn around and go through the n1,n2.. stores to reach the destination}}}.

最后应说明的是:以上所述仅为本发明的优选实施例而已,并不用于限制本发明,尽管参照前述实施例对本发明进行了详细的说明,对于本领域的技术人员来说,其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换。凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。Finally, it should be noted that the above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, the The technical solutions described in the foregoing embodiments may be modified, or some technical features thereof may be equivalently replaced. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (5)

1.一种拥挤室内图像识别云导航系统,其特征在于:包括:1. a crowded indoor image recognition cloud navigation system, is characterized in that: comprise: 移动端和服务器端;mobile and server; 移动端连接服务器端,若连接成功,服务器发送抽象地图信息至移动端,抽象地图信息包括map.csv,为:The mobile terminal connects to the server terminal. If the connection is successful, the server sends the abstract map information to the mobile terminal. The abstract map information includes map.csv, which is: 0,0,00, 0, 0 1,0,01, 0, 0 100,5,0100, 5, 0 500,5,1500,5,1 从移动端拍摄当前行人视角的一个图片序列中提取出稳定的特征点发送至服务端;Extract stable feature points from a sequence of pictures taken by the mobile terminal from the current pedestrian perspective and send them to the server; 服务器端根据移动端发送的稳定特征点信息匹配存储于地图中的特征点,得出匹配度最高并且匹配度高于根据经验设定的阈值的一张图片编号;The server matches the feature points stored in the map according to the stable feature point information sent by the mobile terminal, and obtains a picture number with the highest matching degree and a matching degree higher than the threshold set according to experience; 由于已知与拍摄者最接近的图片编号,故能根据map.csv信息得知目前拍摄者的位置与视角;Since the picture number closest to the photographer is known, the current position and perspective of the photographer can be known according to the map.csv information; 根据目前拍摄者的位置与视角的信息,综合考虑离拍摄者最近店铺的编号、目的地的编号、地图中店铺总数、拍摄者当前朝向,在移动端利用算法一计算出最短路径,给出导航建议。According to the current location and angle of view of the photographer, comprehensively consider the number of the nearest store to the photographer, the number of the destination, the total number of stores in the map, and the current orientation of the photographer, and use algorithm 1 to calculate the shortest path on the mobile terminal, and give the navigation Suggest. 2.根据权利要求1所述的一种拥挤室内图像识别云导航系统,其特征在于:所述map.csv第一列为地图中的图片编号,第二列为店铺代码,第三列为该图片的朝向,2. A kind of crowded indoor image recognition cloud navigation system according to claim 1, it is characterized in that: described map.csv first column is the picture number in the map, the second column is the store code, and the third column is the the orientation of the picture, 其中,0为顺时针,1为逆时针,Among them, 0 is clockwise, 1 is counterclockwise, 另外一个map.csv记录了每个店铺编号代表的店铺名称。Another map.csv records the store name represented by each store number. 3.根据权利要求1所述的一种拥挤室内图像识别云导航系统,其特征在于:所述移动端拍摄为间隔500ms拍摄一张,共拍摄4张,耗时2秒。3 . The cloud navigation system for image recognition in crowded indoor rooms according to claim 1 , wherein the mobile terminal takes a picture at an interval of 500ms, and takes 4 pictures in total, which takes 2 seconds. 4 . 4.根据权利要求1所述的一种拥挤室内图像识别云导航系统,其特征在于:所述稳定的特征点是指排除了行人移动物体干扰的环境特征点。4 . The cloud navigation system for image recognition in a crowded indoor environment according to claim 1 , wherein the stable feature points refer to environmental feature points that exclude the interference of pedestrians and moving objects. 5 . 5.根据权利要求1所述的一种拥挤室内图像识别云导航系统,其特征在于:所述建议包括两类:5. The cloud-based navigation system for image recognition in crowded indoor rooms according to claim 1, wherein the suggestions include two categories: ①沿当前路径经过n1,n2..店铺即可到达目的地;①The destination can be reached by passing through n1, n2.. shops along the current path; ②转身途经经过n1,n2..店铺即可到达目的地。②Turn around and pass through n1, n2.. shops to reach the destination.
CN201910396992.7A 2019-05-14 2019-05-14 A Crowded Indoor Image Recognition Cloud Navigation System Pending CN110146083A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910396992.7A CN110146083A (en) 2019-05-14 2019-05-14 A Crowded Indoor Image Recognition Cloud Navigation System

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910396992.7A CN110146083A (en) 2019-05-14 2019-05-14 A Crowded Indoor Image Recognition Cloud Navigation System

Publications (1)

Publication Number Publication Date
CN110146083A true CN110146083A (en) 2019-08-20

Family

ID=67595319

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910396992.7A Pending CN110146083A (en) 2019-05-14 2019-05-14 A Crowded Indoor Image Recognition Cloud Navigation System

Country Status (1)

Country Link
CN (1) CN110146083A (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103398717A (en) * 2013-08-22 2013-11-20 成都理想境界科技有限公司 Panoramic map database acquisition system and vision-based positioning and navigating method
CN104897165A (en) * 2014-03-06 2015-09-09 苏州工业园区新国大研究院 Shot scenery-based navigation method and system thereof
CN104899603A (en) * 2015-06-03 2015-09-09 孙思宇 Optimization algorithm based on image matching indoor positioning
CN106940186A (en) * 2017-02-16 2017-07-11 华中科技大学 A kind of robot autonomous localization and air navigation aid and system
CN107806872A (en) * 2017-09-18 2018-03-16 东莞新吉凯氏测量技术有限公司 Augmented reality navigation method based on machine vision
CN108020225A (en) * 2016-10-28 2018-05-11 大辅科技(北京)有限公司 Map system and air navigation aid based on image recognition
CN109029444A (en) * 2018-06-12 2018-12-18 深圳职业技术学院 One kind is based on images match and sterically defined indoor navigation system and air navigation aid
CN109520510A (en) * 2018-12-26 2019-03-26 安徽智恒信科技有限公司 A kind of indoor navigation method and system based on virtual reality technology

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103398717A (en) * 2013-08-22 2013-11-20 成都理想境界科技有限公司 Panoramic map database acquisition system and vision-based positioning and navigating method
CN104897165A (en) * 2014-03-06 2015-09-09 苏州工业园区新国大研究院 Shot scenery-based navigation method and system thereof
CN104899603A (en) * 2015-06-03 2015-09-09 孙思宇 Optimization algorithm based on image matching indoor positioning
CN108020225A (en) * 2016-10-28 2018-05-11 大辅科技(北京)有限公司 Map system and air navigation aid based on image recognition
CN106940186A (en) * 2017-02-16 2017-07-11 华中科技大学 A kind of robot autonomous localization and air navigation aid and system
CN107806872A (en) * 2017-09-18 2018-03-16 东莞新吉凯氏测量技术有限公司 Augmented reality navigation method based on machine vision
CN109029444A (en) * 2018-06-12 2018-12-18 深圳职业技术学院 One kind is based on images match and sterically defined indoor navigation system and air navigation aid
CN109520510A (en) * 2018-12-26 2019-03-26 安徽智恒信科技有限公司 A kind of indoor navigation method and system based on virtual reality technology

Similar Documents

Publication Publication Date Title
CN112101339B (en) Map interest point information acquisition method and device, electronic equipment and storage medium
WO2019184542A1 (en) Indoor positioning method, indoor positioning system, indoor positioning device, and computer readable medium
CN104378735B (en) Indoor orientation method, client and server
CN106574975B (en) Trace matching using peripheral signals
CN104897161B (en) Indoor plane map making method based on laser ranging
US10949999B2 (en) Location determination using street view images
CN105516929B (en) Establish method, indoor orientation method and the corresponding intrument of indoor map data
CN107131883B (en) Full-automatic mobile terminal indoor positioning system based on vision
CN109583505A (en) A kind of object correlating method, device, equipment and the medium of multisensor
CN103747524A (en) Android terminal indoor positioning method based on cloud platform
US10660062B1 (en) Indoor positioning
CN109141444A (en) Localization method, device, storage medium and mobile device
CN113330278B (en) A navigation device, method and related equipment
CN111768430A (en) A tracking method for expressway outfield vehicles based on multi-feature cascade matching
TW202001892A (en) Indoor positioning system and method based on geomagnetic signals in combination with computer vision
CN109902681A (en) User group&#39;s relationship determines method, apparatus, equipment and storage medium
US11275939B2 (en) Movement intelligence using satellite imagery
TWM560099U (en) Indoor precise navigation system using augmented reality technology
CN113063421A (en) Navigation method and related device, mobile terminal and computer readable storage medium
CN109753901A (en) Indoor pedestrian tracing method, device, computer equipment and storage medium based on pedestrian recognition
CN116612493B (en) Pedestrian geographic track extraction method and device
Elhamshary et al. JustWalk: A crowdsourcing approach for the automatic construction of indoor floorplans
Xue et al. A fast visual map building method using video stream for visual-based indoor localization
CN107105404B (en) Pedestrian indoor positioning method based on step length matching
CN110146083A (en) A Crowded Indoor Image Recognition Cloud Navigation System

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20190820

WD01 Invention patent application deemed withdrawn after publication