CN113848878B - Indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data - Google Patents
Indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data Download PDFInfo
- Publication number
- CN113848878B CN113848878B CN202110976002.4A CN202110976002A CN113848878B CN 113848878 B CN113848878 B CN 113848878B CN 202110976002 A CN202110976002 A CN 202110976002A CN 113848878 B CN113848878 B CN 113848878B
- Authority
- CN
- China
- Prior art keywords
- data
- indoor
- outdoor
- dimensional
- crowd source
- 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.)
- Active
Links
- 238000010276 construction Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000001514 detection method Methods 0.000 claims description 36
- 238000012545 processing Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000008030 elimination Effects 0.000 claims description 9
- 238000003379 elimination reaction Methods 0.000 claims description 9
- 238000013135 deep learning Methods 0.000 claims description 6
- 238000011084 recovery Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000006399 behavior Effects 0.000 description 17
- 230000006870 function Effects 0.000 description 10
- 238000013527 convolutional neural network Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 230000001133 acceleration Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000001174 ascending effect Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 4
- 230000005284 excitation Effects 0.000 description 4
- 238000011176 pooling Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
- G05D1/0278—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using satellite positioning signals, e.g. GPS
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
- G05D1/0285—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using signals transmitted via a public communication network, e.g. GSM network
Abstract
The invention discloses an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data, which comprises the following steps: acquiring crowd source data, and identifying indoor data and outdoor data according to the crowd source data; performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data; layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data; and constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data. The method and the system can construct indoor and outdoor three-dimensional pedestrian road networks based on the crowd source data, so that more accurate route planning service can be conveniently provided for pedestrians when the pedestrians pass indoors and outdoors.
Description
Technical Field
The invention relates to the field of intelligent traffic, in particular to an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data.
Background
Due to the rapid development of urban road planning, the application requirements for navigation have also increased. The road data provided in the mainstream navigation application is mainly based on motor vehicle road network, but special walking road network data still have the defects, such as walking paths among high-rise buildings, underground passages, pedestrian overpasses, indoor passing paths and the like. In urban environments, a dense and complicated high-rise building group often needs guidance of navigation application, and when a navigation system provides route planning service, a real optimal path cannot be searched out due to the lack of indoor traffic path and three-dimensional path data.
Accordingly, the prior art is still in need of improvement and development.
Disclosure of Invention
The invention aims to solve the technical problems by providing an indoor and outdoor three-dimensional pedestrian road network construction method based on the crowd source data, aiming at solving the defects of the prior art, and providing more accurate route planning service for pedestrians when passing indoors and outdoors by constructing the three-dimensional pedestrian road network.
The technical scheme adopted for solving the technical problems is as follows:
in a first aspect, the present invention provides a method for constructing an indoor and outdoor three-dimensional pedestrian road network based on crowd source data, wherein the method comprises:
acquiring crowd source data, and identifying indoor data and outdoor data according to the crowd source data;
performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data;
layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data;
and constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data.
In one implementation, identifying indoor data and outdoor data from the crowd source data includes:
performing indoor and outdoor detection on the crowd source data to obtain a first detection result;
and dividing the crowd source data into indoor data and outdoor data according to the first detection result.
In one implementation manner, performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data includes:
performing pedestrian dead reckoning on the indoor data to obtain reckoning data;
performing error elimination processing on the calculated data to obtain complete indoor data;
and carrying out integrated processing on the complete indoor data and the outdoor data to obtain indoor and outdoor integrated track data.
In one implementation manner, the layering processing the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data includes:
performing three-dimensional road network node detection on the indoor and outdoor integrated track data through deep learning to obtain a second detection result;
layering the indoor and outdoor integrated track data according to the second detection result to obtain indoor and outdoor three-dimensional layered track data.
In one implementation manner, the constructing the indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data includes:
performing layered density calculation on the indoor and outdoor three-dimensional layered track data to obtain a layered density map;
extracting ridge lines in the density map of each layer to be an indoor and outdoor integrated pedestrian road network of each layer;
and jointing the positions of the indoor and outdoor integrated pedestrian road networks and the three-dimensional road network nodes to obtain the indoor and outdoor three-dimensional pedestrian road network.
In a second aspect, an embodiment of the present invention further provides an indoor and outdoor three-dimensional pedestrian road network device based on crowd source data, where the device includes:
the acquisition module is used for acquiring the crowd source data and identifying indoor data and outdoor data according to the crowd source data;
the estimating module is used for carrying out pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data;
the layering module is used for layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data;
and the road network construction module is used for constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data.
In a third aspect, an embodiment of the present invention further provides a terminal device, where the terminal device includes: a processor, a storage medium communicatively coupled to the processor, the storage medium adapted to store a plurality of instructions; the processor is adapted to invoke the instructions in the storage medium to perform an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data as described in any one of the above.
In a fourth aspect, an embodiment of the present invention further provides a computer readable storage medium, where the computer readable storage medium stores one or more programs, where the one or more programs are executable by one or more processors to implement a method for constructing an indoor and outdoor three-dimensional pedestrian road network based on crowd source data as set forth in any one of the above.
The invention has the beneficial effects that: compared with the prior art, the invention provides an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data, which comprises the steps of obtaining the crowd source data, obtaining indoor data and outdoor data according to the crowd source data, carrying out pedestrian dead reckoning on the indoor data to obtain complete indoor data because the indoor data have shortage or inaccuracy, combining the complete indoor data with the outdoor data to obtain indoor and outdoor integrated track data, layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data, and finally constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data, wherein the indoor and outdoor three-dimensional pedestrian road network constructed by the method can provide more accurate route planning service for pedestrians during indoor and outdoor traffic.
Drawings
Fig. 1 is a flowchart of a specific implementation of an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data according to an embodiment of the present invention.
Fig. 2 is a flowchart for determining indoor and outdoor integrated track data in the indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data according to the embodiment of the invention.
Fig. 3 is a schematic diagram of pedestrian dead reckoning in the indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data provided by the embodiment of the invention.
Fig. 4 is a flowchart of determining indoor and outdoor three-dimensional layered track data in the indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data according to the embodiment of the invention.
Fig. 5 is a three-axis acceleration under different environmental behaviors in the method for constructing the indoor and outdoor three-dimensional pedestrian road network based on the crowd source data according to the embodiment of the invention.
Fig. 6 is a flowchart of constructing an indoor and outdoor three-dimensional pedestrian road network in the indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data according to the embodiment of the invention.
Fig. 7 is a schematic block diagram of an indoor and outdoor three-dimensional pedestrian road network device based on crowd source data according to an embodiment of the invention.
Fig. 8 is a schematic block diagram of an internal structure of a terminal device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more clear and clear, the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
With the rapid development of urban construction, more and more roads are arranged in a staggered manner, so that the traveling problems of pedestrians are slowly revealed in the complicated roads, especially in dense and complicated high-rise building groups in urban environments, pedestrians often need guidance of navigation application, while the road data provided in the main stream navigation application at present are mainly based on motor vehicle road networks, and special walking road network data still have defects, such as walking paths among high-rise buildings, underground passages, pedestrian overpasses, indoor passing paths and the like. The pedestrian road network is a geographic data network suitable for specific requirements of pedestrians, is also a key basic geographic information data in a walking navigation system, and the defects of indoor traffic paths and three-dimensional path data can cause that the navigation system can not search out a real optimal path in reality when providing a route planning service.
The research shows that the main reason for the loss of the pedestrian road network is that the acquisition difficulty of the pedestrian road network data is high, and the professional acquisition and measurement vehicle for the road network is limited by the road and cannot enter the special road for the pedestrian.
In order to solve the problems in the prior art, the embodiment provides an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data, and by the indoor and outdoor three-dimensional pedestrian road network construction method based on the crowd source data, a more accurate three-dimensional pedestrian road network can be built, so that more accurate route planning service can be conveniently provided for pedestrians when the pedestrians pass indoors and outdoors. In specific implementation, the embodiment firstly acquires the crowd source data, obtains the indoor data and the outdoor data according to the crowd source data, carries out pedestrian dead reckoning on the indoor data to obtain complete indoor data because the indoor data have shortage or inaccuracy, combines the outdoor data to obtain indoor and outdoor integrated track data, carries out layering on the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data, and finally constructs an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data.
Exemplary method
The indoor and outdoor three-dimensional pedestrian road network construction method based on the crowd source data in the embodiment can be applied to terminal equipment, such as a smart phone, and the terminal equipment can acquire the crowd source data. In specific implementation, as shown in fig. 1, the indoor and outdoor three-dimensional pedestrian road network construction method based on the crowd source data in the embodiment includes the following steps:
and step S100, acquiring crowd source data, and identifying indoor data and outdoor data according to the crowd source data.
In specific implementation, the embodiment is an indoor and outdoor three-dimensional pedestrian road network constructed based on the crowd source data, so that the crowd source data needs to be acquired first. Along with the popularization of various sensors on intelligent terminal equipment, people holding the intelligent terminal equipment can become map data providers, and new opportunities are brought to acquisition of pedestrian road network data. The common sensors GPS, accelerometer, gyroscope, barometer and the like of the intelligent terminal equipment can acquire corresponding position information and gesture information, so that track data of user travel is obtained. Single user track data often has contingency, crowd source geographic data is open geographic space data which is acquired by a large number of non-professional persons volunteer and provided to masses or related institutions through the Internet, and a large number of track data can truly reflect topological structure information of roads. Thus, constructing a road network using crowd source data is a common and efficient road network construction method.
Because the acquired crowd source data comprises data information acquired indoors and outdoors, indoor data and outdoor data need to be acquired according to the crowd source data in order to construct a pedestrian road network more accurately. Specifically, indoor and outdoor detection is performed on the crowd source data to obtain a first detection result, and the crowd source data is divided into indoor data and outdoor data according to the first detection result.
Preferably, the crowd source data is collected and uploaded through an intelligent terminal device carried by a user, and the collected trail of pedestrians is as follows: t= { P 1 ，P 2 ,P 3 ,...P N }, wherein P i Representative of the trace sample point, denoted P i ＝(t i ,l i ,n i ,h i ,a i ,r i )，t i For the timestamp of the trace point, l i ＝(x i ,y i ) For the geographic location of the point, x i Representing the longitude, y of the point i Represents the latitude of the point, n i Is the number of valid satellites received by the locus point, h i Is the GPS horizontal precision of the track point, a i ＝(a xi ,a yi ,a zi ) The information of the trace point accelerometer is expressed as acceleration values of trace points in X-axis, Y-axis and Z-axis directions, and the gyroscope information r of the trace points i ＝(r xi ,r yi ,r zi ) The rotation rates of the gyroscope about the X, Y and Z axes, respectively, are shown. Under outdoor open environment, GPS can provide accurate positional information for the user, however under the condition that there is the building to shelter from in the room, GPS inaccuracy or lack completely, consequently need to carry out indoor outer detection to crowded source data, distinguish indoor data and outdoor data. The effective satellite number n and the horizontal precision h can directly reflect the signal intensity and the data accuracy of the GPS. The effective satellite number n refers to the number of satellites with a satellite signal-to-noise ratio lower than 30 in the visible satellites at the position point, and when the effective satellite number is smaller than a certain threshold value, the indoor track can be judged, namely n i <σ n . The GPS horizontal accuracy h means that the GPS equipment has 68% probability, in the horizontal plane taking the judging position as the origin and taking h as the radius, the smaller h indicates that the GPS position information is more accurate, and when the horizontal accuracy is greater than a certain threshold value, the GPS accuracy is insufficient, and the indoor track can be judged, namely h i >σ h 。
And step 200, performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data.
In this embodiment, after the indoor data is acquired, under the condition that the building is shielded in the room, the GPS is inaccurate or completely missing, so that the acquired indoor track is in a missing or inaccurate condition, and therefore, the indoor data needs to be calculated to obtain the complete indoor data.
In one implementation, as shown in fig. 2, the step S200 includes the following steps:
s201, performing pedestrian dead reckoning on the indoor data to obtain reckoning data;
s202, performing error elimination processing on the calculated data to obtain complete indoor data;
s203, carrying out integrated processing on the complete indoor data and the outdoor data to obtain indoor and outdoor integrated track data.
In the process of estimating the missing data, the GPS data of the missing data is lost or drifted, and therefore, it is necessary to obtain the trajectory data using other sensor data. Preferably, the Pedestrian Dead Reckoning (PDR) calculates the step frequency and the step length through the change of acceleration when a person walks, and estimates the course by combining with a gyroscope, so that the dead reckoning is realized, and the method can be used for acquiring indoor tracks. The principle of pedestrian dead reckoning is shown in fig. 3, wherein E is the direction of positive east, N is the direction of positive north, θ is the course angle, and L is the step size, and the formula is as follows:calculating according to a formula to obtain calculation data.
Because accumulated errors exist in the dead reckoning of the pedestrian, when error elimination processing is carried out on the reckoning data, a GPS lost point is taken as an origin, a cutterhead coordinate system is converted into a WGS-84 coordinate system, after conversion from a plane coordinate to a longitude and latitude coordinate is completed, a GPS recovery point is taken as a standard, the difference value of the angle and the position between the dead reckoning end point and the GPS recovery point is calculated, and the accumulated errors of the dead reckoning of the pedestrian are eliminated through coordinate rotation and position adjustment, so that complete indoor data are obtained.
In the embodiment, after the complete indoor data is obtained, the outdoor data is combined, and the complete indoor data and the outdoor data are subjected to integrated processing to obtain indoor and outdoor integrated track data. Preferably, the complete indoor data and the outdoor data are combined according to the time stamp, so that the indoor and outdoor integrated track data are obtained.
And step S300, layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data.
In one implementation, as shown in fig. 4, the step S300 includes the following steps:
and S301, performing three-dimensional road network node detection on the indoor and outdoor integrated track data through deep learning to obtain a second detection result.
S302, layering the indoor and outdoor integrated track data according to the second detection result to obtain indoor and outdoor three-dimensional layered track data.
In specific implementation, three-dimensional road network node detection is required to be carried out on the integrated track data to obtain a second detection result, and layering is carried out on the integrated track data according to the second detection result to obtain a layered pedestrian track. Specifically, when layering integrated track data to obtain an indoor and outdoor integrated pedestrian road network, performing cross-level behavior recognition on the integrated track data through deep learning, namely recognizing behaviors of going up and down stairs, going up and down elevators, going up and down slopes and the like, determining three-dimensional road network nodes, layering tracks, and obtaining indoor and outdoor three-dimensional layered track data.
The accelerometer, gyroscope, magnetometer and barometer data in the intelligent terminal equipment can be used for identifying pedestrians in different fieldsDifferent behaviors in the scene. For example, the bump behavior of pedestrians during ascending and descending stairs, the ascending and descending behavior of walking on a sloping road, the overweight and weightlessness behavior occurring during elevator riding, and the like, and the corresponding acceleration data have different characteristics, and it is necessary to obtain a layered track by layering integrated track data, as shown in fig. 5. Preferably, in this embodiment, the three-dimensional road network node detection is performed by identifying the behavior of the user carrying the intelligent terminal device, and in this embodiment, the purpose of behavior identification is achieved by adopting a convolutional neural network (Convolutional Neural Networks, CNN). The convolutional neural network (Convolutional Neural Networks, CNN) is a deep neural network with multi-layer supervised learning, and the core modules of the CNN network architecture are a convolutional layer and a pooling layer (sampling layer). The input layer data enter a convolution layer, an excitation layer and a pooling layer, extracted features serve as input of a first full-connection layer, and finally output tuples are finally classified through a classifier. Specifically, in this embodiment, sample data collection and processing are performed first, for example: and acquiring a plurality of sample data of different behaviors by using an APP preset on the intelligent terminal equipment, wherein each piece of data is provided with a corresponding behavior label. The sensor data acquired by the intelligent terminal equipment are time sequence data, the acquired data are required to be processed before behavior classification is carried out, a time sequence sight distance is divided by a sliding window with fixed length, samples of each behavior are obtained, the time window is set to be 2 seconds, R epsilon R is set assuming that the samples are divided into R samples. Data input at the input layer is then performed, such as: the input data is sensor data acquired by intelligent terminal equipment, the embodiment adopts three-axis acceleration and air pressure change data, the final data dimension after processing is a one-dimensional vector, n is the total number of data in a sliding time window, and the input vector is expressed asInput layer data is input into the convolutional layer, such as: the convolutional layer output is z= [ z 1 ,z 2 ...,z h ]Wherein h represents the size of the output vector, and the calculation formula of the layer is +.>Where m represents the number of convolution kernels (filters), ω= [ ω ] 1 ,ω 2 ,...,ω m-1 ,ω m ]. Wherein the relation between h, m, s (s is the sampling interval in convolution) and n is +.>Preferably, an activation function is arranged in the excitation layer, and the activation function has the function of adding some nonlinear factors into the neural network, so that the neural network can better solve the complex problem, and the excitation function adopted in the embodiment is a ReLu excitation function with the formula of->Where z is the output data of the convolutional layer. And then the features are further extracted and integrated through the pooling layer to obtain new features, such as: dividing a series of data after exciting the layer into a plurality of window data, namely segmenting the series of data, and independently carrying out summation or averaging operation on the data of each window to replace all the data in the window, wherein the data amount is reduced after the operation, and filling by using a filling function is needed at the moment so as to ensure that the number of the data is unchanged. And outputting final predicted values through a full connection layer, wherein the full connection layer is positioned behind a plurality of convolution layers and pooling layers. In this embodiment, two full connection layers are used, the number of nodes in the first full connection layer is 1000, after the first full connection layer is finished, the result is input into the second full connection layer, the number of nodes in the second full connection layer is identical to the number of categories to be output, in this embodiment, 4, and when the second full connection layer calculates, some of the connections between the nodes in the second full connection layer and the nodes in the first layer are randomly abandoned, so as to prevent overfitting. The calculation formula of the full connection layer is +.>Y rk Output of kth fully-connected layer node for the kth sliding time window sampleAs a result, the full link layer output may be represented as Y r ＝[Y r1 ,Y r2 ,Y r3 ,Y r4 ]. In this embodiment, a softmax layer is further provided, and the softmax layer is located after the full-connection layer, so as to obtain the probability that each time window belongs to each behavior, wherein the input data is the output data of the full-connection layer, and the calculation formula is->Where k=1, 2, 3, 4, respectively represent 4 behavior types. For each time window sample, there is y' r ＝[y' r1 ,y' r2 ,y' r3 ,y' r4 ]And taking the maximum value in the four probability values as a final category prediction result. Preferably, in the above process, the weight ω and the deviation b need to be continuously trained, and in this embodiment, a gradient descent method (Adam method) is used to perform minimization solution of a loss function, where the loss function is defined as->The weights in the network are continuously updated to achieve the optimal prediction effect. In the using process of the CNN model, firstly, a sliding window is used for dividing the acquired crowdsourcing data to obtain a data sample. And classifying the acquired crowdsourcing data by using the trained CNN model to obtain the behavior type of each data sample. According to longitude and latitude data of head and tail points of different behavior tracks, the positions of corresponding semantic nodes, namely an up-and-down stair position point, an up-and-down slope position point and an elevator position point, are recorded, and because different predicted points possibly appear in a plurality of track data for the same semantic position, the embodiment uses a K-means clustering algorithm to obtain the accurate position of the semantic nodes. Sample points at each position are input, a sample set is divided into K clusters according to the distance difference between samples, and K different semantic nodes C are represented, wherein C= (C) 1 、C 2 、...、C k ) Tightly linking the spots of the same cluster together, minimizing the square error +.>Wherein->And obtaining the optimal mass center of each cluster, namely the accurate position of each semantic node, by adopting a heuristic iteration method for the mean value vector, namely the mass center, of the cluster.
And step 400, constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data.
In one implementation, as shown in fig. 6, the step S400 includes the following steps:
s401, performing layering density calculation on the indoor and outdoor three-dimensional layering track data to obtain a layering density map.
S402, extracting ridge lines in the density map of each layer to be an indoor and outdoor integrated pedestrian road network of each layer.
And S403, jointing the positions of the indoor and outdoor integrated pedestrian road networks and the three-dimensional road network nodes of each layer to obtain the indoor and outdoor three-dimensional pedestrian road network.
Specifically, in the process of extracting a layered density map based on indoor and outdoor three-dimensional layered track data, the embodiment carries out kernel density analysis on the indoor and outdoor three-dimensional layered track data based on a Morse theory to obtain a density map, derives raster data, takes a density value of each raster as a height, analogizes the density map into a topographic map, then adopts a neighboring point height comparison method to extract key points, namely saddle points, peak points and valley points, calculates a path with the highest rising speed from each saddle point to the peak point, namely a ridge line, and finally extracts the ridge line in each layer of the density map as an indoor and outdoor integrated pedestrian road network of each layer. Specifically, the kernel density analysis mainly calculates the unit density of the track segment in the neighborhood, thereby generating a density map. E= { E 1 ,e 2 ,e 3 ,...e m The } represents a set of all track segments with longitude and latitude as nodes, L is a rectangular bounding box of all tracks, and the vertex set V= { V 1 ,v 2 ,v 3 ,...v n }. The kernel density analysis requires the determination of the search radius, i.e., bandwidth R. The smaller the bandwidth value, the more detailed the display information of the generated grid.
Wherein D is s Is the standard distance, i.e. the standard deviation of the distances between all trajectories and their centroid, D m For the median distance, n is the sum of the number of track segments.
The density map is subjected to nuclear density estimation, and the density function is as follows:
and (3) deriving a result of the kernel density estimation to obtain a density value of each grid, wherein each track of the density of the output grid pixels is the sum of all density values superposed in the center of the grid pixels.
And finally, reconstructing the road network of the indoor and outdoor integrated pedestrian road network to obtain the indoor and outdoor three-dimensional pedestrian road network. Specifically, the derived raster data is triangulated and split, and the derived raster data format is:
where m represents the number of rows of raster data and n represents the number of columns of raster data. Sequencing adjacent points in a clockwise direction to obtain a grid adjacent point set U, and comparing a grid point h with h in the set U i Density value (height). Let delta i ＝h-h i ，N C Is delta 1 ,Δ 2 ,...,Δ n The number of sign changes, n, is the number of neighbors in the neighbor set U. If N c ＝0,Δ i >0, h is the peak point; if N C ＝0,Δ i <0, h is the valley point; if N C And if not less than 4, h is a saddle point. Extracting corresponding key points, and taking an integral line connecting the key points as a key line to be obtained by integratingAll points on the line that flow into the key point are called stable manifolds, the stable manifold of peak points is only itself, and the stable manifold that flows from the integral line into the saddle point is the ridge line. As the walking track of most people is the real road track, in the density topography map, the grid density of the real road direction is higher than the grid density of two sides of the real road, and the grid density of the road intersection is higher than the grid density of surrounding track points, the topography ridge line is extracted through a density function, the stable manifold of saddle points is calculated as an indoor and outdoor integrated pedestrian road network, the extracted indoor and outdoor integrated pedestrian road network is jointed by taking the position of a three-dimensional road network node as a reference, and finally the indoor and outdoor three-dimensional pedestrian road network is obtained.
In summary, the embodiment firstly obtains crowd source data, obtains indoor data and outdoor data according to the crowd source data, and carries out pedestrian dead reckoning on the indoor data to obtain complete indoor data because the indoor data are deficient or inaccurate, combines the complete indoor data with the outdoor data to obtain indoor and outdoor integrated track data, carries out layering on the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data, and finally constructs an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data, so that more accurate route planning service can be provided through the three-dimensional pedestrian road network constructed by the method. For example, after the crowd source data are obtained through APP installed on a plurality of user mobile phones, the crowd source data are divided into indoor data and outdoor data, in the process of detecting the indoor data, it is found that GPS (global positioning system) can not provide accurate position information due to shielding of a building, so that the indoor data are lost, after the indoor data are calculated through dead reckoning, complete indoor data are obtained, the complete indoor data and the outdoor data are integrated to obtain indoor and outdoor integrated track data, the indoor and outdoor integrated track data are subjected to deep learning to detect the actions of ascending and descending stairs, elevators and slopes, the integrated tracks are layered to obtain indoor and outdoor integrated pedestrian road networks, finally the indoor and outdoor three-dimensional pedestrian road networks are constructed according to the fact that the indoor and outdoor integrated pedestrian road networks are connected with the three nodes of the ascending and descending stairs, the elevators and the slopes, and more accurate route planning services are conveniently provided for pedestrians during indoor and outdoor passing.
Exemplary apparatus
As shown in fig. 7, the present embodiment further provides an indoor and outdoor three-dimensional pedestrian road network device based on crowd source data, the device comprising: the system comprises an acquisition module 10, an estimation module 20, a layering module 30 and a road network construction module 40. Specifically, the acquiring module 10 is configured to acquire crowd source data, and identify indoor data and outdoor data according to the crowd source data. The estimating module 20 is configured to perform dead reckoning on the indoor data to obtain complete indoor data, and combine the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data. The road network construction module 30 is configured to perform layering processing on the indoor and outdoor integrated track data, so as to obtain indoor and outdoor three-dimensional layered track data. The road network construction module 40 is configured to construct an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data.
In one implementation, the acquisition module 10 includes:
the first detection unit is used for detecting the crowd source data indoors and outdoors to obtain a first detection result;
and the dividing unit is used for dividing the crowd source data into indoor data and outdoor data according to the first detection result.
In one implementation, the calculation module 20 includes:
the estimating unit is used for carrying out pedestrian dead reckoning on the indoor data to obtain estimated data;
the error elimination unit is used for carrying out error elimination processing on the calculated data to obtain complete indoor data;
and the integrated unit is used for carrying out integrated processing on the complete indoor data and the outdoor data to obtain indoor and outdoor integrated track data.
In one implementation, the layering module 30 includes:
the second detection unit is used for detecting the three-dimensional road network nodes of the indoor and outdoor integrated track data through deep learning to obtain a second detection result;
and the layering unit is used for layering the indoor and outdoor integrated track data according to the second detection result to obtain indoor and outdoor three-dimensional layered track data.
In one implementation, the road network construction module 40 includes:
the layered density map acquisition unit is used for carrying out layered density calculation on the indoor and outdoor three-dimensional layered track data to obtain a layered density map;
the extraction unit is used for extracting ridge lines in the density maps of all layers to be indoor and outdoor integrated pedestrian road networks of all layers;
and the road network construction unit is used for jointing the indoor and outdoor integrated pedestrian road networks of all layers with the positions of the three-dimensional road network nodes to obtain the indoor and outdoor three-dimensional pedestrian road networks.
Based on the above embodiment, the present invention further provides a terminal device, and a functional block diagram thereof may be shown in fig. 8. The terminal equipment comprises a processor, a memory, a network interface, a display screen and a temperature sensor which are connected through a system bus. Wherein the processor of the terminal device is adapted to provide computing and control capabilities. The memory of the terminal device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the terminal device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to realize an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data. The display screen of the terminal equipment can be a liquid crystal display screen or an electronic ink display screen, and the temperature sensor of the terminal equipment is preset in the terminal equipment and is used for detecting the running temperature of the internal equipment.
It will be appreciated by persons skilled in the art that the functional block diagram shown in fig. 8 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the terminal device to which the present inventive arrangements are applied, and that a particular terminal device may include more or fewer components than shown, or may combine some of the components, or may have a different arrangement of components.
In one embodiment, a terminal device is provided, where the terminal device includes a memory, a processor, and an indoor and outdoor three-dimensional pedestrian network construction program based on crowd source data stored in the memory and capable of running on the processor, and when the processor executes the indoor and outdoor three-dimensional pedestrian network construction program based on the crowd source data, the processor implements the following operation instructions:
acquiring crowd source data, and identifying indoor data and outdoor data according to the crowd source data;
performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data;
layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data;
and constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
In summary, the invention provides an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data, which comprises the steps of obtaining the crowd source data, obtaining indoor data and outdoor data according to the crowd source data, carrying out pedestrian dead reckoning on the indoor data to obtain complete indoor data because the indoor data have shortage or inaccuracy, combining the complete indoor data with the outdoor data to obtain indoor and outdoor integrated track data, layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data, and finally constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data, so that more accurate route planning service can be provided for pedestrians during indoor and outdoor traffic through the indoor and outdoor three-dimensional pedestrian road network constructed by the method.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (5)
1. An indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data is characterized by comprising the following steps:
acquiring crowd source data, and identifying indoor data and outdoor data according to the crowd source data;
the obtaining the crowd source data, and identifying indoor data and outdoor data according to the crowd source data comprises:
performing indoor and outdoor detection on the crowd source data to obtain a first detection result;
dividing the crowd source data into indoor data and outdoor data according to the first detection result;
the dividing the crowd source data into indoor data and outdoor data according to the first detection result includes:
dividing the crowd source data into indoor data and outdoor data according to the effective satellite number or horizontal precision in the first detection result;
when the effective satellite number is smaller than a first preset threshold value, the crowd source data are indoor data;
when the horizontal precision is greater than a second preset threshold, the crowd source data are indoor data;
performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data;
the step of performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data comprises the following steps:
performing pedestrian dead reckoning on the indoor data to obtain reckoning data;
performing error elimination processing on the calculated data to obtain complete indoor data;
carrying out integrated processing on the complete indoor data and the outdoor data to obtain indoor and outdoor integrated track data;
the performing error elimination processing on the calculated data to obtain complete indoor data includes:
converting a vehicle coordinate system into a WGS-84 coordinate system by taking a GPS lost point as an origin, completing the conversion from a plane coordinate to a longitude and latitude coordinate, taking a GPS recovery point as a standard, calculating the difference value of the angle and the position between a pedestrian dead reckoning end point and the GPS recovery point, and eliminating the accumulated error of the pedestrian dead reckoning by coordinate rotation and position adjustment to obtain complete indoor data;
layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data;
constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data;
the constructing the indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data comprises the following steps:
performing layered density calculation on the indoor and outdoor three-dimensional layered track data to obtain a layered density map;
extracting ridge lines in the density map of each layer to be an indoor and outdoor integrated pedestrian road network of each layer;
and jointing the positions of the indoor and outdoor integrated pedestrian road networks and the three-dimensional road network nodes to obtain the indoor and outdoor three-dimensional pedestrian road network.
2. The method for constructing an indoor and outdoor three-dimensional pedestrian road network based on crowd source data according to claim 1, wherein the layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data comprises:
performing three-dimensional road network node detection on the indoor and outdoor integrated track data through deep learning to obtain a second detection result;
layering the indoor and outdoor integrated track data according to the second detection result to obtain indoor and outdoor three-dimensional layered track data.
3. An indoor and outdoor three-dimensional pedestrian road network device based on crowd source data, the device comprising:
the acquisition module is used for acquiring the crowd source data and identifying indoor data and outdoor data according to the crowd source data;
the obtaining the crowd source data, and identifying indoor data and outdoor data according to the crowd source data comprises:
performing indoor and outdoor detection on the crowd source data to obtain a first detection result;
dividing the crowd source data into indoor data and outdoor data according to the first detection result;
the dividing the crowd source data into indoor data and outdoor data according to the first detection result includes:
dividing the crowd source data into indoor data and outdoor data according to the effective satellite number or horizontal precision in the first detection result;
when the effective satellite number is smaller than a first preset threshold value, the crowd source data are indoor data;
when the horizontal precision is greater than a second preset threshold, the crowd source data are indoor data;
the estimating module is used for carrying out pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data;
the step of performing pedestrian dead reckoning on the indoor data to obtain complete indoor data, and combining the complete indoor data with outdoor data to obtain indoor and outdoor integrated track data comprises the following steps:
performing pedestrian dead reckoning on the indoor data to obtain reckoning data;
performing error elimination processing on the calculated data to obtain complete indoor data;
carrying out integrated processing on the complete indoor data and the outdoor data to obtain indoor and outdoor integrated track data;
the performing error elimination processing on the calculated data to obtain complete indoor data includes:
converting a vehicle coordinate system into a WGS-84 coordinate system by taking a GPS lost point as an origin, completing the conversion from a plane coordinate to a longitude and latitude coordinate, taking a GPS recovery point as a standard, calculating the difference value of the angle and the position between a pedestrian dead reckoning end point and the GPS recovery point, and eliminating the accumulated error of the pedestrian dead reckoning by coordinate rotation and position adjustment to obtain complete indoor data;
the layering module is used for layering the indoor and outdoor integrated track data to obtain indoor and outdoor three-dimensional layered track data;
the road network construction module is used for constructing an indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data;
the constructing the indoor and outdoor three-dimensional pedestrian road network according to the indoor and outdoor three-dimensional layered track data comprises the following steps:
performing layered density calculation on the indoor and outdoor three-dimensional layered track data to obtain a layered density map;
extracting ridge lines in the density map of each layer to be an indoor and outdoor integrated pedestrian road network of each layer;
and jointing the positions of the indoor and outdoor integrated pedestrian road networks and the three-dimensional road network nodes to obtain the indoor and outdoor three-dimensional pedestrian road network.
4. A terminal device, characterized in that the terminal device comprises: a processor, a storage medium communicatively coupled to the processor, the storage medium adapted to store a plurality of instructions; the processor is adapted to invoke instructions in the storage medium to perform an indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data implementing any of the above claims 1-2.
5. A computer-readable storage medium storing one or more programs executable by one or more processors to implement a method of three-dimensional pedestrian road network construction based on crowd source data as claimed in any one of claims 1-2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110976002.4A CN113848878B (en) | 2021-08-24 | 2021-08-24 | Indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110976002.4A CN113848878B (en) | 2021-08-24 | 2021-08-24 | Indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113848878A CN113848878A (en) | 2021-12-28 |
| CN113848878B true CN113848878B (en) | 2023-08-08 |
Family
ID=78976134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110976002.4A Active CN113848878B (en) | 2021-08-24 | 2021-08-24 | Indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113848878B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115952364B (en) * | 2023-03-07 | 2023-05-23 | 之江实验室 | Route recommendation method and device, storage medium and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105263113A (en) * | 2015-10-26 | 2016-01-20 | 深圳大学 | Wi-Fi location fingerprint map building method and system based on crowd-sourcing |
| CN106017476A (en) * | 2016-07-12 | 2016-10-12 | 中国地质大学（武汉） | Method for generating indoor positioning and navigating map model |
| CN106448431A (en) * | 2016-08-24 | 2017-02-22 | 电子科技大学 | Mobile phone sensor-based indoor planar graph construction method adopting crowdsourcing mode |
| CN107688665A (en) * | 2017-09-27 | 2018-02-13 | 深圳大学 | A kind of indoor map method for auto constructing, device and storage medium |
| CN111288999A (en) * | 2020-02-19 | 2020-06-16 | 深圳大学 | Pedestrian road network attribute detection method, device and equipment based on mobile terminal |
| CN112013862A (en) * | 2020-07-31 | 2020-12-01 | 深圳大学 | Pedestrian network extraction and updating method based on crowdsourcing trajectory |
| CN113124857A (en) * | 2021-06-17 | 2021-07-16 | 腾讯科技（深圳）有限公司 | Method for determining entrance and exit positions, related device, equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9733091B2 (en) * | 2007-05-31 | 2017-08-15 | Trx Systems, Inc. | Collaborative creation of indoor maps |
| US9464903B2 (en) * | 2011-07-14 | 2016-10-11 | Microsoft Technology Licensing, Llc | Crowd sourcing based on dead reckoning |
| US9335175B2 (en) * | 2012-08-15 | 2016-05-10 | Google Inc. | Crowd-sourcing indoor locations |
-
2021
- 2021-08-24 CN CN202110976002.4A patent/CN113848878B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105263113A (en) * | 2015-10-26 | 2016-01-20 | 深圳大学 | Wi-Fi location fingerprint map building method and system based on crowd-sourcing |
| CN106017476A (en) * | 2016-07-12 | 2016-10-12 | 中国地质大学（武汉） | Method for generating indoor positioning and navigating map model |
| CN106448431A (en) * | 2016-08-24 | 2017-02-22 | 电子科技大学 | Mobile phone sensor-based indoor planar graph construction method adopting crowdsourcing mode |
| CN107688665A (en) * | 2017-09-27 | 2018-02-13 | 深圳大学 | A kind of indoor map method for auto constructing, device and storage medium |
| CN111288999A (en) * | 2020-02-19 | 2020-06-16 | 深圳大学 | Pedestrian road network attribute detection method, device and equipment based on mobile terminal |
| CN112013862A (en) * | 2020-07-31 | 2020-12-01 | 深圳大学 | Pedestrian network extraction and updating method based on crowdsourcing trajectory |
| CN113124857A (en) * | 2021-06-17 | 2021-07-16 | 腾讯科技（深圳）有限公司 | Method for determining entrance and exit positions, related device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113848878A (en) | 2021-12-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tang et al. | CLRIC: Collecting lane-based road information via crowdsourcing | |
| US10970542B2 (en) | Scalable three dimensional object segmentation | |
| EP4318397A2 (en) | Method of computer vision based localisation and navigation and system for performing the same | |
| CN105241445B (en) | A kind of indoor navigation data capture method and system based on intelligent mobile terminal | |
| JP2019527832A (en) | System and method for accurate localization and mapping | |
| WO2018057331A1 (en) | Discovering points of entry to a location | |
| EP3762856A1 (en) | Automatic identification of roadside objects for localization | |
| CN112380312B (en) | Laser map updating method based on grid detection, terminal and computer equipment | |
| CN102208013A (en) | Scene matching reference data generation system and position measurement system | |
| US11232582B2 (en) | Visual localization using a three-dimensional model and image segmentation | |
| WO2021082464A1 (en) | Method and device for predicting destination of vehicle | |
| CN111275969B (en) | Vehicle track filling method based on intelligent identification of road environment | |
| CN114964212B (en) | Multi-machine collaborative fusion positioning and mapping method oriented to unknown space exploration | |
| CN110727714A (en) | Resident travel feature extraction method integrating space-time clustering and support vector machine | |
| Zhou et al. | A pedestrian network construction system based on crowdsourced walking trajectories | |
| Gervasoni et al. | Convolutional neural networks for disaggregated population mapping using open data | |
| Bierlaire et al. | Modeling route choice behavior from smartphone GPS data | |
| CN113848878B (en) | Indoor and outdoor three-dimensional pedestrian road network construction method based on crowd source data | |
| CN110009054B (en) | Hierarchical airborne LiDAR point cloud classification method utilizing geometric and intensity features | |
| CN114577191B (en) | Surveying and mapping data acquisition method and system based on geospatial information data | |
| CN109612458A (en) | The method for realizing user location tracking in mountain area environment using mobile phone air pressure and acceleration transducer | |
| CN115344655A (en) | Method and device for finding change of feature element, and storage medium | |
| WO2022021209A9 (en) | Electronic map generation method and apparatus, computer device, and storage medium | |
| Wang et al. | Intersection and stop bar position extraction from vehicle positioning data | |
| CN114674317A (en) | Self-correcting dead reckoning system and method based on activity recognition and fusion filtering |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |