WO2021017675A1 - 轨迹数据处理方法、设备以及存储介质 - Google Patents
轨迹数据处理方法、设备以及存储介质 Download PDFInfo
- Publication number
- WO2021017675A1 WO2021017675A1 PCT/CN2020/096747 CN2020096747W WO2021017675A1 WO 2021017675 A1 WO2021017675 A1 WO 2021017675A1 CN 2020096747 W CN2020096747 W CN 2020096747W WO 2021017675 A1 WO2021017675 A1 WO 2021017675A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- trajectory
- layer
- pixel position
- movement
- flow
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
- G01C21/3804—Creation or updating of map data
- G01C21/3807—Creation or updating of map data characterised by the type of data
- G01C21/3815—Road data
- G01C21/3819—Road shape data, e.g. outline of a route
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
- G01C21/3804—Creation or updating of map data
- G01C21/3833—Creation or updating of map data characterised by the source of data
- G01C21/3841—Data obtained from two or more sources, e.g. probe vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
- G01C21/3863—Structures of map data
- G01C21/387—Organisation of map data, e.g. version management or database structures
- G01C21/3878—Hierarchical structures, e.g. layering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Definitions
- This application relates to the field of network technology, and in particular to a method, device and storage medium for processing track data.
- Trajectory data refers to data information formed by sampling the movement process of one or more moving objects.
- the above-mentioned moving objects usually refer to vehicles.
- collecting trajectory data and processing the collected trajectory data that is, analyzing trajectory data through trajectory modeling methods, has become an in-depth study of trajectory data and mining moving objects in the real world.
- the related technology After the related technology collects the trajectory data, it simply processes the trajectory data. For example, consider the trajectory as a 3D (three-dimensional) polyline formed by the movement of a moving object in time-space, where the polyline includes a series of position points arranged in time; then, the polyline is placed in a fixed space Projection is performed on the scale to obtain the sampling position sequence, that is, the trajectory is defined as a simple sampling position sequence. In the subsequent process, further in-depth analysis is completed based on the sampling position sequence, that is, trajectory mining is performed.
- 3D three-dimensional
- the above-mentioned trajectory data processing method simply converts the trajectory into a sequence of sampling positions, and the method is single, and the trajectory data usually contains rich feature information, such as rich in-depth knowledge of the road network, so the above-mentioned trajectory data processing method will As a result, the characteristic information of the trajectory data is severely lost, which not only fails to accurately mine the movement patterns of moving objects in the real world, but also seriously affects the subsequent trajectory mining results. For this reason, how to process trajectory data has become an urgent problem to be solved by those skilled in the art.
- the embodiments of the present application provide a trajectory data processing method, device, and storage medium, which have a good trajectory modeling effect, can accurately mine the motion pattern of a moving object in the real world, and ensure the subsequent trajectory mining effect.
- the technical solution is as follows:
- a trajectory data processing method is provided, the method is applied to a trajectory data processing device, and the method includes:
- trajectory data within a target geographic range, where the trajectory data includes movement trajectories of multiple moving objects
- the feature layer including at least a trajectory flow layer, a moving speed layer, and a traffic direction layer;
- a trajectory data processing method is provided, the method is applied to a trajectory data processing device, and the method includes:
- the feature layers including at least a trajectory flow layer, a moving speed layer, and a traffic direction layer;
- different colors in the feature layer represent the trajectory flow, moving speed, and traffic direction respectively.
- the displaying the trajectory traffic at the pixel position, the moving speed and the traffic direction of the moving object includes:
- the trajectory flow volume at the pixel position, the moving speed and the traffic direction of the moving object are displayed on the feature display layer.
- a trajectory data processing device in another aspect, includes:
- the first display module is configured to display characteristic layers, the characteristic layers including at least a trajectory flow layer, a moving speed layer, and a traffic direction layer;
- the second display module is configured to display the trajectory flow rate, the moving speed and the traffic direction of the moving object at the pixel position in response to the acquisition of the selection instruction for any pixel position on the feature layer.
- different colors in the feature layer represent the trajectory flow, moving speed, and traffic direction respectively.
- the second display module is further configured to superimpose and display a feature display layer on the feature layer; display the track at the pixel position on the feature display layer Traffic, moving speed and direction of movement of moving objects.
- a trajectory data processing device in another aspect, includes:
- An acquisition module configured to acquire trajectory data within a target geographic range, the trajectory data including the movement trajectories of multiple moving objects;
- a generating module for generating a characteristic layer of the trajectory data including at least a trajectory flow layer, a moving speed layer, and a traffic direction layer;
- a processing module configured to perform fusion processing on each layer included in the feature layer to obtain a fused feature layer
- the display module is used to display the fusion feature layer; in response to obtaining a selection instruction for any pixel location on the fusion feature layer, display the trajectory flow at the pixel location, the moving speed and the traffic direction of the moving object.
- the display module is configured to superimpose and display a feature display layer on the fused feature layer; display the trajectory flow at the pixel position on the feature display layer, The speed and direction of movement of the moving object.
- the device further includes:
- the zoom module is used to zoom each grid area into a track image of uniform size
- the conversion module is configured to convert the coordinate system of each movement track included in the grid area for any grid area to obtain the pixel position of each sampling point on each movement track in the corresponding track image;
- the grid area is obtained by dividing the target geographic range according to space, and each movement track included in the trajectory data is divided according to the grid area to which it belongs.
- the conversion module is used to obtain the latitude and longitude coordinates of the region boundary of the grid area; obtain the height value and width value of the corresponding trajectory image; for any sampling point, according to the grid
- the latitude and longitude coordinates of the left and right boundaries of the grid area, the width value, and the longitude coordinates of the sampling point to obtain the horizontal axis coordinates of the sampling point in the corresponding trajectory image; according to the latitude and longitude coordinates of the upper and lower boundaries of the grid area,
- the height value and the latitude coordinate of the sampling point are used to obtain the vertical coordinate of the sampling point in the corresponding trajectory image.
- the generation module in response to the feature layer including the trajectory flow layer, is configured to, for each pixel position in any trajectory image, based on the corresponding grid area Each sampling point on each movement track is at the corresponding pixel position in the track image, and it is detected whether each movement track intersects the pixel position; whenever a movement track intersects the pixel position, the Add one to the track flow at the pixel position;
- the trajectory flow layer includes the trajectory flow of each pixel position in the trajectory image.
- the generation module in response to the feature layer including the movement speed layer, is configured to, for each pixel position in any trajectory image, based on the corresponding grid area Each sampling point on each movement track is at the corresponding pixel position in the track image, and it is detected whether each movement track intersects the pixel position; whenever a movement track intersects the pixel position, the The movement speed of the sampling points with the pixel position on the movement track is accumulated; the average movement speed is obtained based on the total number of accumulated sampling points and the sum of the accumulated movement speed; in response to the average movement speed being less than the speed threshold, the The average moving speed is taken as the moving speed of the moving object at the pixel position;
- the moving speed layer includes the moving speed of the moving object at each pixel position in the trajectory image.
- the generating module in response to the feature layer including the traffic direction layer, is configured to, for each pixel position in any trajectory image, based on the corresponding grid area Each sampling point on each movement track is at the corresponding pixel position in the track image, and it is detected whether each movement track intersects the pixel position; whenever a movement track intersects the pixel position, the Accumulating the vector directions of the sampling points with the pixel positions on the movement track;
- the traffic direction layer includes the vector direction of each pixel position in the trajectory image, and the vector direction represents the traffic direction of the moving object.
- the display module is configured to display the historical trajectory traffic, the historical average moving speed of the moving object, and the historical traffic direction at the geographic location corresponding to the pixel position.
- the processing module is configured to perform flow characteristic processing on the trajectory flow layer to obtain a first characteristic layer; based on the traffic direction layer, the movement speed layer Projecting the moving speed of ⁇ in the direction space to obtain a second feature layer; splicing and merging the first feature layer and the second feature layer to obtain the fused feature layer;
- the first feature layer and the second feature layer have the same value space.
- the processing module is configured to obtain the upper limit value of the flow, the first smoothing factor and the second smoothing factor; to obtain the maximum trajectory flow in the trajectory flow layer; Any pixel position in the trajectory flow layer, based on the trajectory flow at the pixel position, the first smoothing factor, the second smoothing factor, the maximum trajectory flow, and the upper limit of the flow Obtain the trajectory flow of the pixel position.
- the processing module is configured to obtain the moving speed of the moving object at the pixel position in the moving speed layer for any pixel position; in the traffic direction layer Obtain the vector direction of the moving object at the pixel position in, use the product of the moving speed and the sine of the vector direction as the moving speed of the pixel position in the vertical axis direction; combine the moving speed with the The product of the cosine value in the vector direction is used as the moving speed of the pixel position in the horizontal axis direction.
- the device further includes:
- the preprocessing module is used to preprocess the trajectory data after acquiring the trajectory data
- the preprocessing includes at least one of the following:
- a storage medium is provided, and at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement the aforementioned trajectory data processing method.
- a trajectory data processing device in another aspect, includes a processor and a memory, and at least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement the following steps:
- trajectory data within a target geographic range, where the trajectory data includes movement trajectories of multiple moving objects
- the characteristic layer including at least a trajectory flow layer, a moving speed layer, and a traffic direction layer;
- the feature layer includes at least a trajectory flow layer, a moving speed layer, and a traffic direction layer, which is the embodiment of the present application It is to model the trajectory data in a new three-dimensional space, that is, to model the trajectory data as new three-dimensional spatial data in the three dimensions of flow, speed and direction. After that, the various layers are fused to form a fusion Feature layer and visual analysis of the fusion feature layer.
- the embodiment of the present application comprehensively considers the attribute information of multiple dimensions such as trajectory flow, moving speed, and traffic direction in the trajectory modeling stage through the multi-channel feature fusion method, effectively retaining the important features implicit in the trajectory data Therefore, the embodiments of the present application can not only accurately mine the movement pattern of the moving object in the real world, but also improve the subsequent trajectory mining effect.
- the embodiment of the present application can intuitively analyze the trajectory flow at each pixel position, the moving speed and the traffic direction of the moving object, and the effect is better.
- FIG. 1 is a schematic diagram of an implementation environment of a trajectory data processing method provided by an embodiment of the present application
- FIG. 2 is an overall flow chart of processing trajectory data provided by an embodiment of the present application
- FIG. 3 is a schematic diagram of the construction of a multi-feature layer provided by an embodiment of the present application.
- FIG. 4 is a flowchart of a trajectory data processing method provided by an embodiment of the present application.
- FIG. 5 is an execution flowchart of a trajectory modeling process provided by an embodiment of the present application.
- FIG. 6 is a flow chart of a method for trajectory modeling process provided by an embodiment of the present application.
- Fig. 7 is a flowchart of a feature layer construction provided by an embodiment of the present application.
- FIG. 8 is a schematic diagram of a trajectory traffic layer construction provided by an embodiment of the present application.
- FIG. 9 is a schematic diagram of a traffic speed layer construction provided by an embodiment of the present application.
- FIG. 10 is a schematic diagram of a traffic direction layer construction provided by an embodiment of the present application.
- FIG. 11 is a schematic diagram of performing flow characteristic processing according to an embodiment of the present application.
- FIG. 12 is an effect diagram after multi-channel feature fusion provided by an embodiment of the present application.
- FIG. 13 is a schematic diagram of performing visual analysis on a fusion feature layer provided by an embodiment of the present application.
- FIG. 14 is a flowchart of a process of combining trajectory modeling and trajectory mining provided by an embodiment of the present application.
- 15 is a flowchart of a method for processing trajectory data provided by an embodiment of the present application.
- FIG. 16 is a flowchart of a method for processing trajectory data according to an embodiment of the present application.
- FIG. 17 is a schematic structural diagram of a trajectory data processing device provided by an embodiment of the present application.
- FIG. 18 is a schematic structural diagram of a trajectory data processing device provided by an embodiment of the present application.
- FIG. 19 is a schematic structural diagram of a trajectory data processing device provided by an embodiment of the present application.
- At least one refers to one or more than one.
- the at least one moving object may be one moving object, two moving objects, three moving objects, and any integer number of moving objects greater than or equal to one.
- multiple refers to two or more than two.
- the multiple moving objects may be any integer number greater than or equal to two, such as two moving objects and three moving objects.
- Trajectory data It is data information formed by sampling the movement process of one or more moving objects. Among them, the trajectory data usually contains thousands of trajectories.
- the trajectory data usually includes position information of sampling points, sampling time information, and speed of moving objects.
- the location information of the sampling point is usually expressed in latitude and longitude, which is not specifically limited in the embodiment of the present application.
- trajectory data are usually diverse and complex.
- GPS Global Positioning System
- mobile terminal services such as mobile phone services
- communication base stations and other methods can be used.
- the trajectory data also has the characteristics of massive, real-time and diverse.
- Road network It refers to a road system composed of various roads in a certain area, which is interconnected and woven into a network.
- a traffic road composed of many roads, such as main roads, auxiliary roads, branch roads, fork roads, etc.
- a traffic hub or a traffic network can be called a road network.
- the embodiments of the present application provide a new method for modeling massive trajectories based on multi-channel feature expression, which can solve the following problems in related technologies:
- the in-depth knowledge of the road network includes but is not limited to: historical trajectory flow represents the importance of the road, speed represents the traffic capacity of the road, etc.
- the embodiment of the application does not specifically limit this.
- trajectory data are different from traditional spatial data.
- the inherent spatiotemporal coupling, high dimensionality, complex spatial scale information, and large data volume are different from traditional spatial data.
- the massive trajectory modeling method based on multi-channel feature expression realizes the modeling of historical massive trajectory data in a new three-dimensional space.
- the trajectory flow layer, The movement speed layer and the traffic direction layer realize the depiction of massive historical trajectories, effectively retaining the important characteristic information of the trajectory data.
- the trajectory modeling method proposed in the embodiments of the present application also has good versatility, is convenient to expand and enrich subsequent mining methods, and can be used in the road network update system to achieve simple, efficient, active, and intelligent Trajectory information mining provides a technical basis.
- the embodiment of the present application also introduces a visualization module to facilitate intuitive analysis of the movement mode of the moving object in the real world.
- trajectory data processing method also referred to as a trajectory modeling method
- the implementation environment includes a trajectory data processing device 101, which is used to collect massive trajectory data through various channels.
- the trajectory data processing device 101 can access multiple trajectory data sources to collect trajectory data;
- the trajectory data processing device 101 is also configured to execute the massive trajectory modeling method based on multi-channel feature expression provided by the embodiment of the present application based on the collected massive trajectory data.
- the trajectory data processing device 101 can obtain trajectory data collected through various methods such as GPS locator, mobile phone service, communication base station, etc., which is not specifically limited in the embodiment of the present application.
- the trajectory data processing device 101 is a computer device with computing capability.
- the trajectory data processing device 101 may be a computer device such as a desktop computer, a notebook computer, or a server, which is also not specifically limited in the embodiment of the present application.
- the massive trajectory modeling method based on multi-channel feature expression provided by the embodiments of the present application can be applied to trajectory mining.
- trajectory mining can be directly performed based on the trajectory modeling result.
- the embodiments of this application can provide a technical basis for realizing simple, efficient, active, and intelligent trajectory information mining.
- the key to trajectory information mining in the road network update system is how to use historical massive trajectories to discover the similarities and differences between road network data and the real world, so as to quickly update the spatial location, topological relationship and some attribute characteristics of road network data.
- a more complete and accurate road network database is formed, which is of great significance for improving the timeliness of road network data and reducing data collection costs. That is, massive trajectory modeling method is a very basic and key technology in trajectory information mining.
- the trajectory data processing device 101 may include a data collection module 1011, a trajectory preprocessing module 1012, a trajectory modeling module 1013, and a trajectory mining module 1014.
- the trajectory data processing device 101 first performs unified management and storage of multi-source trajectory data; then, it eliminates abnormal trajectories and noises in the collected trajectory data; then, performs multi-layer features on the trajectory data
- the channel is constructed and visualized and analyzed through the visualization tool.
- trajectory mining can be performed based on the trajectory modeling results to complete in-depth research on trajectory data.
- the data collection module 1011 is responsible for accessing multiple trajectory data sources for trajectory data collection. To put it another way, the data collection module 1011 is used for unified management and storage of trajectory data collected from the real world. Wherein, the trajectory data provided by each trajectory data source may be different, and the data collection module 1011 can uniformly manage and store trajectory data in all data formats for subsequent use.
- the trajectory preprocessing module 1012 is responsible for quickly and effectively removing abnormal trajectories and noises in the collected trajectory data to Improve data quality.
- the trajectory modeling module 1013 is the core part of the embodiment of the present application.
- the module is used to model the collected trajectory data as shown in FIG. 3 by constructing a trajectory flow layer, a traffic direction layer, and a moving speed layer.
- the new three-dimensional spatial data, this kind of trajectory modeling method can effectively retain the important characteristic information of the trajectory data, and at the same time introduce the visualization module to facilitate the intuitive analysis of the movement mode of the moving object in the real world.
- the trajectory can be regarded as a 3D polyline formed by the movement of a moving object in time-space, and the polyline is usually represented by a series of position points arranged in time.
- the coordinates (x, y) of the spatial position, where x and y are usually expressed in latitude and longitude, which is not specifically limited in the embodiment of the present application.
- the top layer in the second column refers to the trajectory flow layer
- the second column in the middle refers to the movement speed layer
- the bottom layer refers to the traffic direction layer
- the trajectory data is modeled as new three-dimensional spatial data in the three dimensions of flow, speed, and direction.
- red is used to represent traffic
- green is used to represent speed
- blue is used to represent direction.
- trajectory mining module 1014 it is responsible for combining different mining modes to realize in-depth research on the trajectory data after the trajectory modeling is completed, so as to mine the movement mode of the moving object in the real world.
- the embodiments of the present application propose a new method for modeling massive trajectories based on multi-feature expression.
- multi-channel feature fusion multiple dimensions of attribute information such as trajectory flow, speed, and direction are comprehensively considered.
- the spatiotemporal information and related attributes of the trajectory data are effectively retained, which is convenient for spatiotemporal analysis, exploration and visualization.
- the embodiment of this application relates to a feature modeling method in trajectory mining technology, which can accurately express the movement pattern of a moving object in the real world on the basis of massive trajectory data, thereby updating the system on the road network Realize simple, efficient, active and intelligent trajectory information mining to provide a technical basis.
- Fig. 4 is a flowchart of a method for processing trajectory data provided by an embodiment of the present application. Referring to Fig. 4, the method flow provided by the embodiment of the present application includes:
- the trajectory data processing device acquires trajectory data within the target geographic range.
- This step is used to collect trajectory data.
- the collected trajectory data may be trajectory data for the target geographic range.
- the target geographic range may be the whole country, the whole city, the whole district, etc., which is not specifically limited in the embodiment of the application.
- multiple trajectory data sources can be accessed for trajectory data collection.
- the embodiment of the present application can perform unified management and storage of trajectory data collected from the real world.
- the trajectory data provided by each trajectory data source may be different, and the embodiment of the present application can uniformly manage and store trajectory data in all data formats for subsequent use.
- the collected trajectory data is usually massive, that is, the collected trajectory data usually includes thousands of moving trajectories of thousands of moving objects.
- the embodiment of the present application can model massive historical trajectories.
- the moving object usually refers to a vehicle, which is not specifically limited in the embodiment of the present application.
- the trajectory data processing device preprocesses the acquired trajectory data.
- preprocessing includes at least one of the following:
- This step is used to divide the collected trajectory data according to the grid area. That is, the target geographic area is divided into multiple grid areas according to space. The grid area is shown in the left figure in Figure 3. After that, each movement trajectory included in the collected trajectory data is divided according to the grid. The area is divided.
- the collected trajectory data for the whole country can be divided into multiple rectangles according to the space for grid management.
- a rectangle with a width of 12 kilometers and a height of 8 kilometers is used as the unit for dividing the grid. Accordingly, the whole country is divided into more than 300,000 grid areas; after that, the collected All trajectory data are divided according to their grid area.
- trajectory across multiple grid regions can be truncated according to the grid region boundary to facilitate data decoupling and subsequent parallel processing and analysis, which is not specifically limited in the embodiment of the present application.
- This step is to filter each movement track included in the collected track data according to the set filter condition.
- this embodiment of the present application can filter out trajectories that meet at least one of the following conditions: travel time is lower than a certain time threshold, travel distance is lower than a certain distance threshold, and the number of sampling points included in the trajectory Less than a certain number threshold.
- the number of sampling points is usually the number of GPS points.
- This step is to perform abnormal elimination processing on each movement track included in the collected track data. To put it another way, this step is used to truncate the trajectory twice.
- This step is to perform data interpolation processing on each movement trajectory included in the collected trajectory data.
- a linear interpolation algorithm can be used to estimate the value of the moving object at the unsampled position, where the value here can refer to position information, time information, or speed, etc. This is the case in this application. There is no specific limitation.
- the trajectory data processing device performs feature modeling on the preprocessed trajectory data.
- This step is used to model the collected trajectory data in a new three-dimensional space as shown in Figure 3.
- a trajectory flow layer By constructing a trajectory flow layer, a traffic direction layer and a moving speed layer, the historical massive trajectory is modeled , This kind of trajectory modeling method can effectively retain the important characteristic information of trajectory data.
- trajectory modeling includes the following processing flow:
- the region mapping is used to respectively scale the divided grid regions to a certain fixed size image.
- the grid areas obtained by spatially segmenting geographical regions across the country are respectively scaled to a certain fixed-size image.
- the size of the trajectory image can be 8192 pixels high and 9420 pixels wide, corresponding to the actual geographic area, and the resolution of each pixel is about 1.5 meters accuracy, which is not specifically limited in the embodiment of the application. .
- This step is used to convert geographic coordinates into image coordinates.
- geographic coordinates usually refer to GPS coordinates, that is, latitude and longitude coordinates.
- the pixel position of each sampling point on each movement trajectory falling in the corresponding trajectory image will be calculated, and the pixel position can be ( loc_x, loc_y) refer to.
- the coordinate system conversion is performed on each movement track included in the grid area to obtain the pixel position of each sampling point on each movement track in the corresponding track image, It can be achieved in the following ways:
- the regional boundary includes the left boundary, the right boundary, the upper boundary and the lower boundary.
- the latitude and longitude coordinates of the above regional boundary include: the latitude and longitude coordinates of the left boundary, the latitude and longitude coordinates of the right boundary, the latitude and longitude coordinates of the upper boundary, and the latitude and longitude coordinates of the lower boundary. coordinate.
- the height value and the width value respectively represent the pixel size of the height and width of the track image.
- the horizontal axis refers to the x axis
- the vertical axis refers to the y axis. This step is to calculate loc_x and loc_y.
- each parameter represents the longitude and latitude coordinates of the area boundary of the grid area
- the parameter left refers to the longitude and latitude coordinates of the left boundary
- the parameter right refers to the latitude and longitude coordinates of the right boundary
- the parameter top refers to the latitude and longitude coordinates of the upper boundary
- the parameter bottom refers to the latitude and longitude coordinates of the lower boundary
- the trajectory image zoomed to by the grid area is img(height, width), where,
- the parameter height represents the height of the trajectory image, and the parameter width represents the width of the trajectory image; taking each sampling point as Pt(x,y) as an example, where the parameters x and y represent the longitude and latitude of the sampling point, then
- the calculation formulas for the above-mentioned horizontal axis and vertical axis are as follows:
- loc_x (Pt.x-Rect.left)/(Rect.right-Rect.left)*img.width
- loc_y (Rect.top-Pt.y)/(Rect.top-Rect.bottom)*img.height
- this embodiment of the application will construct a trajectory flow layer, a moving speed layer, and a traffic direction layer.
- the physical meaning of the trajectory flow layer is: represents the actual flow value in the corresponding geographic space area, where the flow can reflect the importance of the road, and the larger the flow, the more important the road;
- the physical meaning of the speed layer is: the corresponding geographic The actual traffic speed in the spatial area;
- the physical meaning of the direction layer is: represents the actual traffic direction in the corresponding geographic spatial area.
- the embodiment of the present application constructs the feature layer based on the grid area as a unit.
- a trajectory image zoomed to by a grid area corresponds to a trajectory flow layer, a movement speed layer, and a traffic direction layer.
- the feature layer construction includes:
- the process of constructing the trajectory flow layer includes:
- the trajectory flow layer includes the trajectory flow of each pixel position in the trajectory image.
- the construction method of the trajectory flow layer is to traverse each GPS point of each movement trajectory, calculate its pixel position in the corresponding trajectory image based on the aforementioned step 4032, and perform flow accumulation at the pixel position.
- the trajectory flow can be calculated according to the following formula:
- Flow ij represents the flow value at C ij in the trajectory flow layer, and its initial value is 0, Traj ⁇ C ij represents whether the current moving trajectory intersects C ij ; therefore, the value space of the last output trajectory flow layer Is [0,+ ⁇ ].
- the intuitive effect of the trajectory flow layer is shown in FIG. 8, where the left image in FIG. 8 is a real-world trajectory image, and the right image is the trajectory flow layer of the trajectory image.
- the brighter position in the trajectory flow layer represents the greater the traffic flow at that position and the stronger the road capacity.
- trajectory flow layer is only a possible implementation method for the construction of the trajectory flow layer.
- other methods can be used to construct the trajectory flow layer, which is not specifically limited in the embodiment of this application. .
- the construction process of the moving speed layer includes:
- the moving speed layer includes the moving speed of the moving object at each pixel position in the trajectory image.
- the first point that needs to be explained is that the moving speed mentioned in this article usually refers to the average moving speed.
- the construction method of the movement speed layer is to traverse each GPS point of each movement trajectory, calculate its pixel position in the corresponding trajectory image based on the aforementioned step 4032, and perform speed accumulation at the pixel position.
- the speed unit used in this application is meter/second, and a speed threshold is set; wherein, the value of the speed threshold can be 50 meters/second, which is not specifically limited in the embodiments of this application . Assuming that the maximum value of the default moving speed is 50 meters/second, the embodiment of the present application performs numerical truncation (set to 50) on the sampling points exceeding the maximum speed to reduce abnormal interference.
- the mobile traffic can be calculated according to the following formula:
- Speed ij represents the speed value at C ij in the moving speed flow layer, and its initial value is 0.
- Traj ⁇ C ij represents whether the current moving trajectory intersects with C ij ; in addition, because there are multiple accumulated at each pixel position The moving speed of the sampling point. Therefore, in order to calculate the average moving speed at the position C ij , it needs to be divided by the sum of the number of sampling points to obtain Traj.speed. In addition, if the obtained Traj.speed is greater than the speed threshold, the size of Traj.speed is set to 50, that is, the value space of the final output moving speed layer is [0,50].
- FIG. 9 the intuitive effect of the moving speed layer is shown in FIG. 9, where the left image in FIG. 9 is the real world trajectory image, and the right image is the moving speed layer of the trajectory image.
- the brighter position in the moving speed layer represents the greater the passing speed at that position.
- the construction process of the traffic direction layer includes:
- the traffic direction layer includes the vector direction of each pixel position in the trajectory image, and the vector direction represents the traffic direction of the moving object.
- the construction method of the traffic direction layer is to traverse each GPS point of each movement track, calculate its pixel position in the corresponding track image based on the aforementioned step 4032, and perform vector direction accumulation at the pixel position.
- the direction coordinate system adopted in the embodiment of the present application is based on the true north included angle, and the vector direction can be calculated according to the following formula:
- the intuitive effect of the traffic direction layer is shown in FIG. 10.
- the left picture in Fig. 10 is the real world trajectory image
- the middle picture is the traffic direction layer of the trajectory image
- the right picture is used to represent the quadrant color.
- the embodiment of this application divides the directions into four quadrants, and the vector directions falling in different quadrants sample different colors for display; for example, the first quadrant adopts red, the second quadrant adopts green, and the second quadrant adopts green.
- the three quadrants are white, and the fourth quadrant is blue.
- the traffic direction layer is displayed in four colors of red, green, white and blue.
- This step is used to splice and merge the trajectory flow layer, movement speed layer and traffic direction layer constructed in step 4033.
- Two problems are mainly solved here:
- the value space of the trajectory flow layer is [0,+ ⁇ ]
- the value space of the moving speed layer is [0,50]
- the value space of the traffic direction layer is [0,360].
- the directional feature is a categorical feature, and the value of its own value is meaningless when layer fusion is performed, and the effect of direct splicing and merging is not ideal.
- the embodiment of the present application will perform flow characteristic processing, speed and direction characteristic processing in the following ways, and the process is as follows:
- the flow characteristic processing on the trajectory flow layer includes:
- the upper limit value of the flow rate is set to 255, which is not specifically limited in the embodiment of the present application.
- the embodiment of the present application can further optimize the trajectory traffic layer by using the following formula:
- Flow ij represents the flow value at a certain pixel position of the trajectory flow layer
- max (Flow) refers to the maximum flow value in the trajectory flow layer
- ⁇ , ⁇ represent the smoothing factor
- the values are constant.
- ⁇ is also referred to as the first smoothing factor
- ⁇ is also referred to as the second smoothing factor.
- the embodiment of the application in order to effectively process the traffic direction layer, the embodiment of the application will comprehensively consider the speed characteristics and the direction characteristics, and project the speed in the direction space. On the premise of preserving the details of the speed characteristics, it can be more reasonable and effective
- the directional feature is integrated into the multi-layer channel.
- projecting the moving speed of the moving speed layer in the direction space includes:
- any pixel position obtain the moving speed of the moving object at that pixel position in the moving speed layer; obtain the vector direction of the moving object at the pixel position in the traffic direction layer; the moving speed and the vector direction
- the product of the sine value is used as the moving speed of the pixel position in the vertical axis direction; the product of the moving speed and the vector direction cosine value is used as the moving speed of the pixel position in the horizontal axis direction.
- the embodiment of the present application divides the direction angle sine
- the product of the value and the moving speed is taken as the average moving speed of the pixel position in the y-axis direction
- the product of the direction angle cosine value and the moving speed is taken as the average moving speed of the pixel position in the x-axis direction; so far, the x-axis and y
- the value space of the calculation result of the axis is [-50,50], where positive and negative represent opposite directions; finally, the value interval is converted from [-50,50] to
- Sy ij represents the average moving speed in the y-axis direction
- Sx ij represents the average moving speed in the x-axis direction
- the left picture in Fig. 12 is the effect picture after the multi-channel layer fusion is performed before the feature processing
- the right picture is the effect picture after the multi-channel layer fusion is performed after the feature processing.
- This step is mainly used to visually analyze the historical trajectory flow, historical average moving speed and historical traffic direction at each pixel location.
- the functions provided by the interface elements include but are not limited to position, size, color, feature display, etc.
- Position refers to the ability to visually display the flow characteristics, velocity characteristics, and direction characteristics at the pixel position in response to a selection instruction for any pixel position in the fusion feature layer; size refers to the corresponding fusion feature layer In the aforementioned grid area; color refers to the RGB components, as mentioned above, R represents flow, G represents speed, and B represents direction; feature display is to display the flow characteristics, speed characteristics and direction at the corresponding pixel position feature.
- the fusion feature layer is displayed through a visualization tool, in response to obtaining a selection instruction for any pixel location on the fusion feature layer, as shown in FIG. 13, the pixel location is displayed
- the trajectory flow at the location, the moving speed and the direction of the moving object.
- the historical track traffic, historical average moving speed, and historical traffic direction at the actual geographic location corresponding to the pixel position can be displayed.
- a feature display layer can be superimposed on the fusion feature layer; the trajectory flow, movement speed, and traffic direction can be displayed on the feature display layer.
- This application is implemented The example does not specifically limit this.
- the embodiment of this application first preprocesses the collected trajectory data; then, the preprocessed trajectory data is modeled in a new three-dimensional space to construct a trajectory flow layer and traffic direction Layers and moving speed layers, that is, the trajectory data is modeled as new three-dimensional spatial data in the three dimensions of flow, speed and direction. Finally, a fusion feature layer is formed through the multi-channel layer fusion technology, and the visualization tool is used Perform visual analysis on the fusion feature layer.
- the embodiment of the present application comprehensively considers the attribute information of multiple dimensions such as trajectory flow, moving speed, and traffic direction in the trajectory modeling stage through the multi-channel feature fusion method, effectively retaining the important features implicit in the trajectory data Information, such as spatio-temporal information and related attributes. Based on this, on the one hand, the embodiments of the present application can accurately mine the motion patterns of moving objects in the real world, that is, they can extract real motion attributes and form a set of time, geography, and trajectory.
- the trajectory modeling framework integrating three-dimensional visualization, etc. can be used to study the theory of moving objects and as a real data set; on the other hand, the embodiments of the present application are convenient to expand and enrich the subsequent trajectory mining, which is simple and easy to implement in the road network update system Efficient, active and intelligent trajectory information mining provides a technical basis.
- the embodiments of the present application can not only accurately mine the movement patterns of moving objects in the real world based on massive trajectory data, but also improve the subsequent trajectory mining effect.
- the embodiment of the present application can intuitively analyze the historical trajectory traffic, historical average moving speed, and historical traffic direction at each pixel position, and the effect is better.
- the trajectory data after trajectory modeling is used for trajectory mining. That is, referring to FIG. 14, the method flow provided by the embodiment of the present application further includes:
- this step is used to combine different mining modes to achieve in-depth research on trajectory data.
- the trajectory mining mode includes but is not limited to frequent mode, classification mode, regression mode, and clustering mode, which are not specifically limited in the embodiment of the present application.
- frequent itemsets there are a series of sets, these sets have some identical elements, and the elements that appear frequently at the same time in the set form a subset, which meets a certain threshold condition, that is, frequent items set.
- This mode can portray user portraits based on trajectory data.
- the classification mode is used to classify the trajectory data or classify users based on the trajectory data, which is not specifically limited in the embodiment of the present application.
- the regression model can be used to predict the congestion of the road based on the trajectory data.
- the clustering mode can predict places where users frequently appear based on trajectory data.
- the collected trajectory data is first preprocessed; then, the preprocessed trajectory data is modeled in a new three-dimensional space to construct a trajectory flow diagram Layer, traffic direction layer, and movement speed layer, that is, trajectory data is modeled as new three-dimensional spatial data in the three dimensions of flow, speed and direction, and a fusion feature layer is formed through multi-channel layer fusion technology. And visually analyze the fusion feature layer through visualization tools.
- the embodiment of the present application comprehensively considers the attribute information of multiple dimensions such as trajectory flow, moving speed, and traffic direction in the trajectory modeling stage through the multi-channel feature fusion method, effectively retaining the important features implicit in the trajectory data Information, such as spatio-temporal information and related attributes.
- the embodiments of the present application can accurately mine the motion patterns of moving objects in the real world, that is, they can extract real motion attributes and form a set of time, geography, and trajectory.
- the trajectory modeling framework integrating three-dimensional visualization, etc. can be used to study the theory of moving objects and as a real data set; on the other hand, the embodiments of the present application are convenient to expand and enrich the subsequent trajectory mining, which is simple and easy to implement in the road network update system. Efficient, active and intelligent trajectory information mining provides a technical basis.
- the embodiments of the present application can not only accurately mine the movement patterns of moving objects in the real world based on massive trajectory data, but also improve the subsequent trajectory mining effect.
- trajectory data processing method provided in the embodiment of the present application further includes the method flow shown in FIG. 15:
- the feature layer includes at least a trajectory flow layer, a moving speed layer, and a traffic direction layer.
- Display the fusion feature layer in response to obtaining a selection instruction for any pixel location on the fusion feature layer, display the trajectory flow at the pixel location, the moving speed and the traffic direction of the moving object.
- the embodiment of the application models the trajectory data in a new three-dimensional space to construct a trajectory flow layer, a traffic direction layer, and a moving speed layer, that is, the trajectory data is modeled as a flow,
- the new three-dimensional spatial data in the three dimensions of speed and direction are then used to form a fusion feature layer through multi-channel layer fusion technology, and visually analyze the fusion feature layer.
- the embodiment of the present application comprehensively considers the attribute information of multiple dimensions such as trajectory flow, moving speed, and traffic direction in the trajectory modeling stage through the multi-channel feature fusion method, effectively retaining the important features implicit in the trajectory data Information, such as spatio-temporal information and related attributes.
- the embodiments of the present application can accurately mine the motion patterns of moving objects in the real world, that is, they can extract real motion attributes and form a set of time, geography, and trajectory.
- the trajectory modeling framework integrating three-dimensional visualization, etc. can be used to study the theory of moving objects and as a real data set; on the other hand, the embodiments of the present application are convenient to expand and enrich the subsequent trajectory mining, which is simple and easy to implement in the road network update system. Efficient, active and intelligent trajectory information mining provides a technical basis.
- the embodiments of the present application can not only accurately mine the movement patterns of moving objects in the real world based on massive trajectory data, but also improve the subsequent trajectory mining effect.
- the embodiment of the present application can intuitively analyze the trajectory flow at each pixel position, the moving speed and the traffic direction of the moving object, and the effect is better.
- trajectory data processing method provided in the embodiment of the present application further includes the method flow shown in FIG. 16:
- a feature layer is displayed.
- the feature layer includes at least a trajectory flow layer, a movement speed layer, and a traffic direction layer.
- the feature layer in this step refers to the aforementioned fusion feature layer, that is, the feature layer in this step is the fusion result of each layer included in the feature layer of the trajectory data, and each layer includes at least the trajectory Traffic layer, moving speed layer and traffic direction layer.
- the fusion feature layer in this step is the feature layer obtained through the fusion process in step 4034. That is, after acquiring the trajectory data in the target geographic range, a feature layer of the trajectory data is generated, and the fusion feature layer is obtained by performing fusion processing on each layer included in the feature layer.
- the trajectory data includes movement trajectories of multiple moving objects.
- the functions provided by the interface elements include but are not limited to position, size, color, feature display, etc.
- Position refers to the ability to visually display the flow characteristics, velocity characteristics, and direction characteristics at the pixel position in response to a selection instruction for any pixel position in the fusion feature layer; size refers to the corresponding fusion feature layer In the aforementioned grid area; color refers to the RGB components, as mentioned above, R represents flow, G represents speed, and B represents direction; feature display is to display the flow characteristics, speed characteristics and direction at the corresponding pixel position feature.
- displaying the trajectory flow at the pixel position, the moving speed and the traffic direction of the moving object including but not limited to: superimposing and displaying a feature display layer on the fusion feature layer; in the feature display layer
- the trajectory volume at the pixel position, the moving speed of the moving object and the direction of travel are displayed on the top.
- the fusion feature layer can also be partially enlarged and displayed, which is not specifically limited in the embodiment of the present application.
- the method provided by the embodiments of the present application can perform visual analysis on trajectory data. For example, after displaying a feature layer, in response to obtaining a selection instruction for any pixel location on the feature layer, it can visually display the location at that pixel location.
- the trajectory flow, the moving speed and the traffic direction of the moving object that is, the embodiment of the application can intuitively analyze the flow characteristics, speed characteristics, and direction characteristics at each pixel position, and the effect is better.
- the embodiment of this application directly models the collected mass movement trajectories. In some trajectory mining application scenarios, it may be necessary to pay attention to a specific time period. Therefore, the embodiment of this application can also be extended to carry out mass movement trajectories. Model separately by time period.
- FIG. 17 is a schematic structural diagram of a trajectory data processing device provided by an embodiment of the present application. Referring to Figure 17, the device includes:
- the obtaining module 1701 is configured to obtain trajectory data within a target geographic area, where the trajectory data includes movement trajectories of multiple moving objects;
- a generating module 1702 configured to generate a characteristic layer of the trajectory data, the characteristic layer including at least a trajectory flow layer, a moving speed layer, and a traffic direction layer;
- the processing module 1703 is configured to perform fusion processing on each layer included in the feature layer to obtain a fused feature layer;
- the display module 1704 is configured to display the fusion feature layer; in response to obtaining a selection instruction for any pixel location on the fusion feature layer, display the trajectory flow at the pixel location, the moving speed and the traffic direction of the moving object.
- the device provided by the embodiment of the present application first generates a characteristic layer of the trajectory data after acquiring the trajectory data within the target geographic area, where the characteristic layer includes at least a trajectory flow layer, a movement speed layer, and a traffic direction
- the layer that is, the embodiment of the application is to model the trajectory data in a new three-dimensional space, that is, to model the trajectory data as new three-dimensional spatial data in the three dimensions of flow, speed and direction.
- the layers are fused to form a fusion feature layer, and the fusion feature layer is visually analyzed.
- the embodiment of the present application comprehensively considers the attribute information of multiple dimensions such as trajectory flow, moving speed, and traffic direction in the trajectory modeling stage through the multi-channel feature fusion method, effectively retaining the important features implicit in the trajectory data Therefore, the embodiments of the present application can not only accurately mine the movement pattern of the moving object in the real world, but also improve the subsequent trajectory mining effect.
- the embodiment of the present application can intuitively analyze the trajectory flow at each pixel position, the moving speed and the traffic direction of the moving object, and the effect is better.
- the display module is configured to superimpose and display a feature display layer on the fusion feature layer; display the trajectory flow and the moving object at the pixel position on the feature display layer The speed and direction of travel.
- the device further includes:
- the zoom module is used to zoom each grid area into a track image of uniform size
- the conversion module is configured to convert the coordinate system of each movement track included in the grid area for any grid area to obtain the pixel position of each sampling point on each movement track in the corresponding track image;
- the grid area is obtained by dividing the target geographic range according to space, and each movement track included in the trajectory data is divided according to the grid area to which it belongs.
- the conversion module is used to obtain the latitude and longitude coordinates of the regional boundary of the grid area; obtain the height value and width value of the corresponding track image; for any sampling point, according to the grid area
- the latitude and longitude coordinates of the left and right boundaries of the grid area, the width value and the longitude coordinates of the sampling point to obtain the horizontal axis coordinates of the sampling point in the corresponding trajectory image;
- the height value and the latitude coordinate of the sampling point are used to obtain the vertical coordinate of the sampling point in the corresponding trajectory image.
- a generating module in response to the feature layer including the trajectory flow layer, is configured to generate, for each pixel position in any trajectory image, based on each pixel position included in the corresponding grid area Each sampling point on the movement track is at the corresponding pixel position in the track image, and it is detected whether each movement track intersects the pixel position; whenever a movement track intersects the pixel position, the pixel The trajectory flow at the position is processed by adding one;
- the trajectory flow layer includes the trajectory flow of each pixel position in the trajectory image.
- the generating module in response to the feature layer including the movement speed layer, is further configured to, for each pixel position in any trajectory image, based on each pixel position included in the corresponding grid area Each sampling point on each movement track is at the corresponding pixel position in the track image, and it is detected whether each movement track intersects the pixel position; whenever a movement track intersects the pixel position, the The movement speed of the sampling points with the pixel position on the movement track is accumulated; the average movement speed is obtained based on the total number of accumulated sampling points and the sum of the accumulated movement speed; in response to the average movement speed being less than the speed threshold, the average Moving speed as the moving speed of the moving object at the pixel position;
- the moving speed layer includes the moving speed of the moving object at each pixel position in the trajectory image.
- the generating module in response to the feature layer including the traffic direction layer, is further configured to, for each pixel position in any trajectory image, based on each pixel position included in the corresponding grid area Each sampling point on each movement track is at the corresponding pixel position in the track image, and it is detected whether each movement track intersects the pixel position; whenever a movement track intersects the pixel position, the Accumulate the vector directions of the sampling points with the pixel positions on the movement track;
- the traffic direction layer includes the vector direction of each pixel position in the trajectory image, and the vector direction represents the traffic direction of the moving object.
- the display module is configured to display the historical trajectory traffic, the historical average moving speed of the moving object, and the historical traffic direction at the geographic location corresponding to the pixel position.
- the processing module is configured to perform flow characteristic processing on the trajectory flow layer to obtain a first characteristic layer; based on the traffic direction layer, the movement of the movement speed layer Projecting the velocity in the direction space to obtain a second feature layer; splicing and merging the first feature layer and the second feature layer to obtain the fused feature layer;
- the first feature layer and the second feature layer have the same value space.
- the processing module is configured to obtain the upper limit value of the flow rate, the first smoothing factor and the second smoothing factor; obtain the maximum trajectory flow in the trajectory flow layer; for the trajectory flow At any pixel position in the layer, based on the trajectory flow at the pixel position, the first smoothing factor, the second smoothing factor, the maximum trajectory flow, and the upper limit of the flow The trajectory flow of the pixel position.
- the processing module is configured to obtain, in the movement speed layer, the movement speed of the moving object at the pixel position for any pixel position; and obtain it in the traffic direction layer.
- the vector direction of the moving object at the pixel position; the product of the moving speed and the sine of the vector direction is used as the moving speed of the pixel position in the vertical axis direction; the moving speed is compared with the vector direction
- the product of the cosine value is used as the moving speed of the pixel position in the horizontal axis direction.
- the device further includes:
- the preprocessing module is used to preprocess the trajectory data after acquiring the trajectory data
- the preprocessing includes at least one of the following:
- FIG. 18 is a schematic structural diagram of a trajectory data processing device provided by an embodiment of the present application. Referring to Figure 18, the device includes:
- the first display module 1801 is configured to display characteristic layers, the characteristic layers including at least a trajectory flow layer, a moving speed layer, and a traffic direction layer;
- the second display module 1802 is configured to display the trajectory flow at the pixel position, the moving speed and the traffic direction of the moving object in response to the acquisition of the selection instruction for any pixel position on the feature layer.
- the device provided in the embodiment of the present application can perform visual analysis on trajectory data. For example, after displaying a feature layer, in response to obtaining a selection instruction for any pixel location on the feature layer, it can visually display the pixel location at that pixel location.
- the trajectory flow, the moving speed and the traffic direction of the moving object that is, the embodiment of the application can intuitively analyze the flow characteristics, speed characteristics, and direction characteristics at each pixel position, and the effect is better.
- different colors in the feature layer represent the trajectory flow, moving speed, and traffic direction respectively.
- the second display module is also used to superimpose and display a feature display layer on the feature layer; display the trajectory flow at the pixel position on the feature display layer, The speed and direction of movement of the moving object.
- trajectory data processing device processes trajectory data
- only the division of the above functional modules is used as an example for illustration.
- the above functions can be allocated by different functional modules as needed.
- the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.
- the trajectory data processing device provided in the foregoing embodiment and the trajectory data processing method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, and will not be repeated here.
- FIG. 19 is a schematic structural diagram of a trajectory data processing device provided by an embodiment of the present application.
- the device 1900 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 1901 and one or more memories 1902, wherein the memory 1902 stores At least one instruction, and the at least one instruction is loaded and executed by the processor 1901 to implement the trajectory data processing method provided by the foregoing method embodiments.
- processors Central Processing Units, CPU
- memories 1902 wherein the memory 1902 stores At least one instruction, and the at least one instruction is loaded and executed by the processor 1901 to implement the trajectory data processing method provided by the foregoing method embodiments.
- the device may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface for input and output, and the device may also include other components for implementing device functions, which will not be described in detail here.
- a computer-readable storage medium such as a memory including instructions, which may be executed by a processor in a terminal to complete the trajectory data processing method in the foregoing embodiment.
- the computer-readable storage medium may be ROM (Read Only Memory), RAM (Random Access Memory, random access memory), CD-ROM (Compact Disc Read-Only Memory, optical disk read-only memory) , Magnetic tapes, floppy disks and optical data storage devices.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Automation & Control Theory (AREA)
- Computer Graphics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Architecture (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Instructional Devices (AREA)
- Traffic Control Systems (AREA)
- Processing Or Creating Images (AREA)
Abstract
本申请公开了一种轨迹数据处理方法、设备以及存储介质,属于网络技术领域。所述方法包括:获取目标地理范围内的轨迹数据,轨迹数据中包括多个移动对象的移动轨迹;生成轨迹数据的特征图层,特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;对特征图层中包括的各个图层进行融合处理,显示融合特征图层;响应于获取到对融合特征图层上任意一个像素位置的选中指令,显示像素位置处的轨迹流量、移动对象的移动速度和通行方向。本申请在轨迹建模阶段综合考虑了流量、速度、方向等多个维度的属性信息,不但能够精确挖掘移动对象在真实世界中的运动模式,而且还能够提升后续的轨迹挖掘效果,另外还可以对融合特征图层进行可视化分析。
Description
本申请要求于2019年07月29日提交的申请号为201910690467.6、发明名称为“轨迹数据处理方法、装置、存储介质以及设备”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请涉及网络技术领域,特别涉及一种轨迹数据处理方法、设备以及存储介质。
轨迹数据,是指通过对一个或多个移动对象的运动过程进行采样而形成的数据信息。以道路网为例,上述移动对象通常指代车辆。时下随着网络技术的快速发展,收集轨迹数据并对收集到的轨迹数据进行处理,即通过轨迹建模方法对轨迹数据进行分析,已经成为对轨迹数据进行深入研究,挖掘移动对象在真实世界中的运动模式的重要手段之一。
相关技术在收集到轨迹数据后,仅是对轨迹数据进行简单处理。比如,将轨迹看作是移动对象在时间-空间中运动而形成的3D(three dimensional,三维)折线,其中,该折线上包括一系列按照时间排列的位置点;之后,将该折线在固定空间尺度下进行投影,得到采样位置序列,即将轨迹定义为简单的采样位置序列,后续过程中基于该采样位置序列完成进一步地深入分析,即进行轨迹挖掘。
上述轨迹数据处理方式,仅是简单地将轨迹转化为采样位置序列,方式单一,而轨迹数据中通常会隐含丰富的特征信息,比如道路网丰富的深层次知识,所以上述轨迹数据处理方式会造成轨迹数据的特征信息严重丢失,不但无法精确挖掘移动对象在真实世界中的运动模式,而且还会严重影响后续的轨迹挖掘效果。为此,如何进行轨迹数据处理,成为了本领域技术人员亟待解决的一个问题。
发明内容
本申请实施例提供了一种轨迹数据处理方法、设备以及存储介质,其轨迹建模效果佳,能够精确挖掘移动对象在真实世界中的运动模式,确保了后续轨迹挖掘效果。所述技术方案如下:
一方面,提供了一种轨迹数据处理方法,所述方法应用于轨迹数据处理设备,所述方法包括:
获取目标地理范围内的轨迹数据,所述轨迹数据中包括多个移动对象的移动轨迹;
生成所述轨迹数据的特征图层,所述特征图层至少包括轨迹流量图层、移 动速度图层和通行方向图层;
对所述特征图层中包括的各个图层进行融合处理,得到融合特征图层;
显示所述融合特征图层;
响应于获取到对所述融合特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
另一方面,提供了一种轨迹数据处理方法,所述方法应用于轨迹数据处理设备,所述方法包括:
显示特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;
响应于获取到对所述特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
在一种可能的实现方式中,所述特征图层中以不同的颜色分别表征轨迹流量、移动速度和通行方向。
在一种可能的实现方式中,所述显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向,包括:
在所述特征图层上叠加显示一个特征展示图层;
在所述特征展示图层上显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
另一方面,提供了一种轨迹数据处理装置,所述装置包括:
第一显示模块,用于显示特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;
第二显示模块,用于响应于获取到对所述特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
在一种可能的实现方式中,所述特征图层中以不同的颜色分别表征轨迹流量、移动速度和通行方向。
在一种可能的实现方式中,所述第二显示模块,还用于在所述特征图层上叠加显示一个特征展示图层;在所述特征展示图层上显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
另一方面,提供了一种轨迹数据处理装置,所述装置包括:
获取模块,用于获取目标地理范围内的轨迹数据,所述轨迹数据中包括多个移动对象的移动轨迹;
生成模块,用于生成所述轨迹数据的特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;
处理模块,用于对所述特征图层中包括的各个图层进行融合处理,得到融合特征图层;
显示模块,用于显示融合特征图层;响应于获取到对所述融合特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的 移动速度和通行方向。
在一种可能的实现方式中,所述显示模块,用于在所述融合特征图层上叠加显示一个特征展示图层;在所述特征展示图层上显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
在一种可能的实现方式中,所述装置还包括:
缩放模块,用于将每一个网格区域分别缩放为一个统一尺寸大小的轨迹图像;
转换模块,用于对于任意一个网格区域,对所述网格区域包括的每条移动轨迹进行坐标系转换,得到每条移动轨迹上各个采样点在相应轨迹图像中的像素位置;
其中,所述网格区域是对所述目标地理范围按照空间进行切分得到的,所述轨迹数据中包括的各条移动轨迹是按照所属网格区域进行划分的。
在一种可能的实现方式中,所述转换模块,用于获取所述网格区域的区域边界的经纬度坐标;获取相应轨迹图像的高度数值和宽度数值;对于任意一个采样点,根据所述网格区域的左右边界的经纬度坐标、所述宽度数值和所述采样点的经度坐标,获取所述采样点在相应轨迹图像中的横轴坐标;根据所述网格区域的上下边界的经纬度坐标、所述高度数值和所述采样点的纬度坐标,获取所述采样点在相应轨迹图像中的纵轴坐标。
在一种可能的实现方式中,响应于所述特征图层包括所述轨迹流量图层,所述生成模块,用于对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述像素位置处的轨迹流量进行加一处理;
其中,所述轨迹流量图层包括所述轨迹图像中每一个像素位置的轨迹流量。
在一种可能的实现方式中,响应于所述特征图层包括所述移动速度图层,所述生成模块,用于对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的移动速度进行累积;基于累积的采样点总数和累积的移动速度总和,获取平均移动速度;响应于所述平均移动速度小于速度阈值,将所述平均移动速度作为移动对象在所述像素位置处的移动速度;
其中,所述移动速度图层包括移动对象在所述轨迹图像中每一个像素位置处的移动速度。
在一种可能的实现方式中,响应于所述特征图层包括所述通行方向图层,所述生成模块,用于对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的矢量方向进行累积;
其中,所述通行方向图层包括所述轨迹图像中每一个像素位置的矢量方向,所述矢量方向表征了移动对象的通行方向。
在一种可能的实现方式中,所述显示模块,用于显示所述像素位置对应的地理位置处的历史轨迹流量、移动对象的历史平均移动速度和历史通行方向。
在一种可能的实现方式中,所述处理模块,用于对所述轨迹流量图层进行流量特征处理,得到第一特征图层;基于所述通行方向图层,将所述移动速度图层的移动速度在方向空间进行投影,得到第二特征图层;将所述第一特征图层和所述第二特征图层进行拼接合并,得到所述融合特征图层;
其中,所述第一特征图层和所述第二特征图层具有相同的取值空间。
在一种可能的实现方式中,所述处理模块,用于获取流量取值上限值、第一平滑因子和第二平滑因子;获取所述轨迹流量图层中的最大轨迹流量;对于所述轨迹流量图层中的任意一个像素位置,基于所述像素位置的轨迹流量、所述第一平滑因子、所述第二平滑因子、所述最大轨迹流量和所述流量取值上限值,重新获取所述像素位置的轨迹流量。
在一种可能的实现方式中,所述处理模块,用于对于任意一个像素位置,在所述移动速度图层中获取移动对象在所述像素位置处的移动速度;在所述通行方向图层中获取移动对象在所述像素位置处的矢量方向;将所述移动速度与所述矢量方向正弦值的乘积,作为所述像素位置在纵轴方向的移动速度;将所述移动速度与所述矢量方向余弦值的乘积,作为所述像素位置在横轴方向的移动速度。
在一种可能的实现方式中,所述装置还包括:
预处理模块,用于在获取到所述轨迹数据后,对所述轨迹数据进行预处理;
其中,预处理包括以下至少一种:
将所述目标地理范围按照空间切分为多个网格区域,将所述轨迹数据中包括的各条移动轨迹按照所属网格区域进行划分;
按照设置的过滤条件,对所述轨迹数据中包括各条移动轨迹进行过滤处理;
对所述轨迹数据中包括的各条移动轨迹进行异常剔除处理;
对所述轨迹数据中包括的各条移动轨迹进行数据插值处理。
另一方面,提供了一种存储介质,所述存储介质中存储有至少一条指令,所述至少一条指令由处理器加载并执行以实现上述的轨迹数据处理方法。
另一方面,提供了一种轨迹数据处理设备,所述设备包括处理器和存储器,所述存储器中存储有至少一条指令,所述至少一条指令由所述处理器加载并执行以实现如下步骤:
获取目标地理范围内的轨迹数据,所述轨迹数据中包括多个移动对象的移动轨迹;
生成所述轨迹数据的特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;
对所述特征图层中包括的各个图层进行融合处理,得到融合特征图层;
显示所述融合特征图层;
响应于获取到对所述融合特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
本申请实施例提供的技术方案带来的有益效果是:
在获取到目标地理范围内的轨迹数据后,首先生成该轨迹数据的特征图层,其中,该特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层,即本申请实施例是对轨迹数据在新的三维空间内进行建模,也即将轨迹数据建模为在流量、速度和方向三个维度上的新的三维空间数据,之后,对各个图层进行融合处理形成一个融合特征图层,并对融合特征图层进行可视化分析。
即,本申请实施例通过多通道特征融合的方式,在轨迹建模阶段综合考虑了轨迹流量、移动速度、通行方向等多个维度的属性信息,有效地保留了轨迹数据中隐含的重要特征信息,因此,本申请实施例不但能够精确挖掘移动对象在真实世界中的运动模式,而且还能够提升后续的轨迹挖掘效果。
另外,在显示融合特征图层后,响应于获取到对该融合特征图层上任意一个像素位置的选中指令,还能够直观显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向,即本申请实施例能够直观分析每个像素位置处的轨迹流量、移动对象的移动速度以及通行方向,效果较佳。
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的一种轨迹数据处理方法的实施环境的示意图;
图2是本申请实施例提供的一种对轨迹数据进行处理的整体流程架构图;
图3是本申请实施例提供的一种多特征图层的构建示意图;
图4是本申请实施例提供的一种轨迹数据处理方法的流程图;
图5是本申请实施例提供的一种轨迹建模过程的执行流程图;
图6是本申请实施例提供的一种轨迹建模过程的方法流程图;
图7是本申请实施例提供的一种特征图层构建的流程图;
图8是本申请实施例提供的一种轨迹流量图层构建的示意图;
图9是本申请实施例提供的一种通行速度图层构建的示意图;
图10是本申请实施例提供的一种通行方向图层构建的示意图;
图11是本申请实施例提供的一种进行流量特征处理的示意图;
图12是本申请实施例提供的一种多通道特征融合后的效果图;
图13是本申请实施例提供的一种对融合特征图层进行可视化分析的示意图;
图14是本申请实施例提供结合轨迹建模和轨迹挖掘过程的流程图;
图15是本申请实施例提供的一种轨迹数据处理方法的流程图;
图16是本申请实施例提供的一种轨迹数据处理方法的流程图;
图17是本申请实施例提供的一种轨迹数据处理装置的结构示意图;
图18是本申请实施例提供的一种轨迹数据处理装置的结构示意图;
图19是本申请实施例提供的一种轨迹数据处理设备的结构示意图。
为使本申请的目的、技术方案和优点更加清楚,下面将结合附图对本申请实施方式作进一步地详细描述。
可以理解,本申请所使用的术语“第一”、“第二”等可在本文中用于描述各种概念,但除非特别说明,这些概念不受这些术语限制。这些术语仅用于将一个概念与另一个概念区分。
其中,至少一个是指一个或一个以上,例如,至少一个移动对象可以是一个移动对象、两个移动对象、三个移动对象等任意大于等于一的整数个移动对象。而多个是指两个或者两个以上,例如,多个移动对象可以是两个移动对象、三个移动对象等任意大于等于二的整数个移动对象。
在对本申请实施例进行详细地解释说明之前,先对本申请实施例涉及的一些缩略语和关键术语进行定义。
轨迹数据:其是通过对一个或多个移动对象的运动过程进行采样而形成的数据信息。其中,轨迹数据中通常包含成千上万条轨迹。
在一种可能的实现方式中,轨迹数据中通常包括采样点的位置信息、采样时间信息以及移动对象的速度等。作为一个示例,采样点的位置信息通常以经纬度进行表示,本申请实施例对此不进行具体限定。
另外,轨迹数据的来源通常多样且复杂,在一种可能的实现方式中,可以通过GPS(Global Positioning System,全球定位系统)定位器、移动终端服务(比如手机服务)、通信基站等多种方式进行获取,其中,轨迹数据还具有海量、实时以及多样的特点。
道路网:其是指在一定区域内由各种道路组成的相互联络、交织成网状分布的道路系统。作为一个示例,由主路、辅路、支路、岔路等很多条道路构成的交通道路,交通枢纽或交通网络,均可称之为道路网。
针对前文提及的相关技术,本申请实施例提供了一种新的基于多通道特征表达的海量轨迹建模方法,该方法能够解决相关技术存在的以下问题:
(1)、缺乏对历史海量轨迹数据进行刻画。
历史海量轨迹数据能够更真实地刻画真实世界中移动对象的运动情况,而相关技术由于缺乏基于历史海量轨迹数据挖掘隐含的道路网深层次知识,所以无法有效运用于路网更新中。其中,道路网深层次知识包括但不限于:历史的轨迹流量代表道路的重要性、速度快慢代表道路的通行能力等,本申请实施例对此不进行具体限定。
(2)、轨迹建模方法单一。
如前文所述,相关技术仅是进行了简单地轨迹建模,然而轨迹数据的表达 形式、产生机理与传统空间数据不同,内在的时空耦合、高维、空间尺度信息复杂、数据量大等特点,更需要通过合理的轨迹建模方法,来对轨迹数据进行分析处理,以便于扩展和丰富后续挖掘手段,提升挖掘效果。
在本申请实施例中,基于多通道特征表达的海量轨迹建模方法,实现了对历史海量轨迹数据在新的三维空间中进行建模,换一种表达方式,通过构建出轨迹流量图层、移动速度图层和通行方向图层,实现对历史海量轨迹进行刻画,有效保留了轨迹数据的重要特征信息。另外,本申请实施例提出的该种轨迹建模方式,还具有较好的通用性,便于扩展和丰富后续挖掘手段,在路网更新系统中能够为实现简单、高效、主动式、智能化的轨迹情报挖掘提供技术基础。另外,本申请实施例还引入了可视化模块,以便于直观分析移动对象在真实世界中的运动模式。
下面结合图1,对本申请实施例提供的一种轨迹数据处理方法(也称之为轨迹建模方法)的实施环境进行介绍说明。
参见图1,该实施环境中包括轨迹数据处理设备101,轨迹数据处理设备101用于通过各种渠道收集海量轨迹数据,比如轨迹数据处理设备101可以接入多个轨迹数据源来收集轨迹数据;另外,轨迹数据处理设备101还用于基于收集到的海量轨迹数据执行本申请实施例提供的基于多通道特征表达的海量轨迹建模方法。
在一种可能的实现方式中,轨迹数据处理设备101可以获取通过GPS定位器、手机服务、通信基站等多种方式收集到的轨迹数据,本申请实施例对此不进行具体限定。其中,轨迹数据处理设备101为具有计算能力的计算机设备。作为一个示例,轨迹数据处理设备101可为台式电脑、笔记本电脑或服务器等计算机设备,本申请实施例对此同样不进行具体限定。
其中,本申请实施例提供的基于多通道特征表达的海量轨迹建模方法,可运用于轨迹挖掘中。如图2所述,在基于本申请实施例提供的轨迹建模方法,完成对预处理后的轨迹数据进行分析处理之后,可直接基于轨迹建模结果进行轨迹挖掘。换一种表达方式,在路网更新系统中,本申请实施例可以为实现简单、高效、主动式、智能化的轨迹情报挖掘提供技术基础。其中,路网更新系统中的轨迹情报挖掘,关键在于如何利用历史海量轨迹发现道路网数据和真实世界之间的异同点,以便于快速更新道路网数据的空间位置、拓扑关系以及一些属性特征,最终形成一个更加完备、精准的道路网数据库,这对于提高道路网数据的时效性、降低数据采集成本来说,都具有十分重要的意义。也即,海量轨迹建模方法是轨迹情报挖掘中的一个非常基础且关键的技术。
在一种可能的实现方式中,参见图2,轨迹数据处理设备101可以包括数据收集模块1011、轨迹预处理模块1012、轨迹建模模块1013以及轨迹挖掘模块1014。
简言之,基于上述模块,轨迹数据处理设备101首先是对多源轨迹数据进行统一管理存储;之后,剔除收集到的轨迹数据中的异常轨迹与噪声;之后,对轨迹数据进行多图层特征通道构建,并通过可视化工具进行可视化分析。另 外,还可以基于轨迹建模结果进行轨迹挖掘,完成对轨迹数据的深入研究。
详细来讲,数据收集模块1011负责接入多个轨迹数据源进行轨迹数据收集。换一种表达方式,数据收集模块1011用于对由真实世界收集而来的轨迹数据进行统一管理存储。其中,各个轨迹数据源提供的轨迹数据可能并不相同,而数据收集模块1011可以对所有数据格式的轨迹数据进行统一管理存储,以便于后续使用。
其中,由于从真实世界收集的轨迹数据的数据质量与挖掘目标所需的质量要求存在一定差距,因此轨迹预处理模块1012负责快速且有效地剔除收集到的轨迹数据中的异常轨迹与噪声,以提高数据质量。
轨迹建模模块1013是本申请实施例的核心部分,该模块用于通过构建轨迹流量图层、通行方向图层以及移动速度图层,将收集到的轨迹数据建模为如图3所示的新的三维空间数据,该种轨迹建模方式能够有效保留轨迹数据的重要特征信息,同时引入可视化模块,以便于直观分析移动对象在真实世界中的运动模式。
参见图3,轨迹可以被看作是移动对象在时间-空间中运动而形成的3D折线,该折线通常是由一系列按照时间排列的位置点来表示。例如,图3中示出了轨迹s,其中,轨迹s可用S:p
1→p
2→...→p
n表示,p
i=(x,y,t),表示在时刻t采样的地理空间位置的坐标(x,y),其中,x和y通常用经纬度表示,本申请实施例对此不进行具体限定。
在图3中,在基于海量轨迹数据构建多特征图层时,第二列最上方的图层指代轨迹流量图层,第二列位于中间的图层指代移动速度图层,第二列最下方的图层指代通行方向图层,进而实现将轨迹数据建模为在流量、速度和方向三个维度上的新的三维空间数据。在一种可能的实现方式中,本申请实施例采用红色代表流量,采用绿色代表速度,采用蓝色代表方向。
针对轨迹挖掘模块1014,负责结合不同的挖掘模式实现对完成轨迹建模后的轨迹数据的深入研究,以挖掘移动对象在真实世界中的运动模式。
综上所述,本申请实施例提出了一种新的基于多特征表达的海量轨迹建模方法,通过多通道特征融合,实现综合考虑了轨迹流量、速度和方向等多个维度的属性信息,有效地保留了轨迹数据的时空信息和相关属性,便于时空分析、探索及可视化。换一种表达方式,本申请实施例涉及轨迹挖掘技术中的特征建模方法,能够在海量轨迹数据的基础上,精准地表达移动对象在真实世界中的运动模式,进而为在路网更新系统中实现简单、高效、主动式、智能化的轨迹情报挖掘提供技术基础。
下面对本申请实施例提供的一种轨迹数据处理方法进行详细地解释说明。
需要说明的是,后续出现的诸如第一、第二、第三、第四等描述,仅是为了用于区分不同的对象,而不构成任何其他的限定。
图4是本申请实施例提供的一种轨迹数据处理方法的流程图。参见图4,本申请实施例提供的方法流程包括:
401、轨迹数据处理设备获取目标地理范围内的轨迹数据。
本步骤用于收集轨迹数据。
在一种可能的实现方式中,收集到的轨迹数据可为针对目标地理范围内的轨迹数据。比如,该目标地理范围可为全国范围、全市范围、全区范围等,本申请实施例对此不进行具体限定。
在本申请实施例中,可以接入多个轨迹数据源进行轨迹数据收集。其中,本申请实施例可以对由真实世界收集而来的轨迹数据进行统一管理存储。另外,各个轨迹数据源提供的轨迹数据可能并不相同,而本申请实施例可以对所有数据格式的轨迹数据进行统一管理存储,以便于后续使用。
其中,收集到的轨迹数据通常海量,即收集到的轨迹数据中通常包括成千上万个移动对象的成千上万条移动轨迹。相应地,对本申请实施例可以对历史海量轨迹进行建模。
以道路网为例,则移动对象通常指代车辆,本申请实施例对此不进行具体限定。
402、轨迹数据处理设备对获取到的轨迹数据进行预处理。
由于从真实世界收集的轨迹数据的数据质量与挖掘目标所需的质量要求存在一定差距,因此在收集到轨迹数据后,还需先剔除轨迹数据中的异常轨迹与噪声,以提高数据质量。
在一种可能的实现方式中,如图2所示,预处理包括以下至少一种:
(a)、数据转换
该步骤用于对收集到的轨迹数据按照网格区域进行划分。即是将目标地理范围按照空间切分为多个网格区域,其中,网格区域如图3中左图所示;之后,将收集到的轨迹数据中包括的各条移动轨迹按照所属网格区域进行划分。
在一种可能的实现方式中,以全国范围为例,可对收集到的针对全国范围的轨迹数据,按照空间切分为多个矩形进行网格化管理。作为一个示例,本申请实施例以宽为12公里,高为8公里的矩形为切分网格的单位,相应地,全国约切分为30多万个网格区域;之后,将收集到的所有轨迹数据按照所属网格区域进行划分。
另外,对于跨多个网格区域的轨迹,可按照网格区域边界进行截断,以便于数据解耦和后续并行处理分析,本申请实施例对此不进行具体限定。
(b)、数据过滤
本步骤即是按照设置的过滤条件,对收集到的轨迹数据中包括各条移动轨迹进行过滤处理。
在一种可能的实现方式中,本申请实施例可以过滤掉满足以下至少一种条件的轨迹:行程时间低于某一时间阈值、行程距离低于某一距离阈值、该轨迹中包括的采样点数少于一定数目阈值。其中,采样点数通常为GPS点数。
(c)、异常剔除
本步骤即是对收集到的轨迹数据中包括的各条移动轨迹进行异常剔除处理。换一种表达方式,该步骤用于对轨迹进行二次截断。
作为一个示例,对于一条轨迹,当该条轨迹中相邻两个采样点(假设这两个采样点分别为Pm和Pn))在数值上偏离较远时,则以Pm为断点对整条轨迹进 行截断,以此类推,递归执行前述步骤,直至将整条轨迹遍历一遍。其中,在数值上偏离较远,通常指代的是在空间或时间上偏离较远,本申请实施例对此不进行具体限定。
(d)、数据插值
本步骤即是对收集到的轨迹数据中包括的各条移动轨迹进行数据插值处理。
在一种可能的实现方式中,可以使用线性插值算法来估计移动对象在未采样位置处的数值,其中,此处的数值可指代位置信息、时间信息或速度等,本申请实施例对此不进行具体限定。
403、轨迹数据处理设备对经过预处理的轨迹数据进行特征建模。
本步骤用于将收集到的轨迹数据在如图3所示的新的三维空间进行建模,通过构建出轨迹流量图层、通行方向图层以及移动速度图层,对历史海量轨迹进行建模,该种轨迹建模方式能够有效保留轨迹数据的重要特征信息。
在一种可能的实现方式中,如图5所示,轨迹建模过程可以分为区域成图、坐标系转换、特征图层构建、多通道图层融合以及可视化分析等五个步骤。换一种表达方式,参见图6,轨迹建模包括下述处理流程:
区域成图
4031、将每一个网格区域分别缩放为一个统一尺寸大小的轨迹图像。
其中,区域成图用于将前述切分出来的各个网格区域分别缩放至某一固定尺寸大小的图像。比如,将对全国范围的地理区域进行空间切分得到的网格区域分别缩放到某一固定尺大小的图像。
作为一个示例,轨迹图像的尺寸大小可为高8192个像素、宽9420个像素,对应到实际地理区域,每个像素的分辨率大约为1.5米的精度,本申请实施例对此不进行具体限定。
坐标系转换
4032、对于任意一个网格区域,对该网格区域包括的每条移动轨迹进行坐标系转换,得到每条移动轨迹上各个采样点在相应轨迹图像中的像素位置。
本步骤用于将地理坐标转换为图像坐标。其中,地理坐标通常指代GPS坐标,即经纬度坐标。在本申请实施例中,对于落入任意一个网格区域内的全部轨迹数据,会计算每条移动轨迹上的各个采样点落入相应轨迹图像中的像素位置,该像素位置在本文中可以(loc_x,loc_y)指代。
在一种可能的实现方式中,对于任意一个网格区域,对该网格区域包括的每条移动轨迹进行坐标系转换,得到每条移动轨迹上各个采样点在相应轨迹图像中的像素位置,可采取下述方式实现:
(a)、获取该网格区域的区域边界的经纬度坐标。
其中,区域边界包括左边界、右边界、上边界以及下边界,相应地,上述区域边界的经纬度坐标包括:左边界的经纬度坐标、右边界的经纬度坐标、上边界的经纬度坐标以及下边界的经纬度坐标。
(b)、获取该网格区域的轨迹图像的高度数值和宽度数值。
其中,高度数值和宽度数值分别表示轨迹图像的高度和宽度的像素大小。
(c)、对于任意一个采样点,根据该网格区域的左右边界的经纬度坐标、该 轨迹图像的宽度数值和该采样点的经度坐标,获取该采样点在该轨迹图像中的横轴坐标;根据该网格区域的上下边界的经纬度坐标、该轨迹图像的高度数值和该采样点的纬度坐标,获取该采样点在该轨迹图像中的纵轴坐标。
在本申请实施例中,横轴指代x轴,纵轴指代y轴,本步骤即是计算loc_x和loc_y。
作为一个示例,假设一个网格区域为Rect(left,right,top,bottom),其中,各参数分别表示该网格区域的区域边界的经纬度坐标,参数left指代左边界的经纬度坐标,参数right指代右边界的经纬度坐标,参数top指代上边界的经纬度坐标,参数bottom指代下边界的经纬度坐标;相应地,该网格区域缩放到的轨迹图像为img(height,width),其中,参数height表示该轨迹图像的高度,参数width表示该轨迹图像的宽度;以每个采样点为Pt(x,y)为例,其中,参数x和y分别表示采样点的经度和纬度,则在进行坐标系转换时,上述横轴坐标和纵轴坐标的计算公式分别如下:
loc_x=(Pt.x-Rect.left)/(Rect.right-Rect.left)*img.width
loc_y=(Rect.top-Pt.y)/(Rect.top-Rect.bottom)*img.height
需要说明的是,上述仅是给出进行坐标系转换的一种可能实现方式,除此之外,还可采取其他方式进行坐标系转换,本申请实施例对此不进行具体限定。
特征图层构建
在一种可能的实现方式中,针对特征图层构建,本申请实施例会构建轨迹流量图层、移动速度图层以及通行方向图层。其中,轨迹流量图层的物理意义为:表示相应地理空间区域内实际流量数值,其中,流量能够反映道路的重要性,流量越大代表道路越重要;速度图层的物理意义为:表示相应地理空间区域内的实际通行速度;方向图层的物理意义为:表示相应地理空间区域内的实际通行方向。
4033、对经过预处理的轨迹数据进行特征图层构建,得到轨迹流量图层、移动速度图层和通行方向图层。
需要说明的是,本申请实施例是以网格区域为单位进行特征图层构建的。换一种表达方式,一个网格区域缩放至的轨迹图像分别对应一个轨迹流量图层、一个移动速度图层和一个通行方向图层。参见图7,特征图层构建包括:
4033-1、对经过预处理的轨迹数据进行轨迹流量图层构建。
在一种可能的实现方式中,轨迹流量图层的构建过程,包括:
对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在该轨迹图像中的相应像素位置,检测每一条移动轨迹是否与该像素位置相交;每当一条移动轨迹与该像素位置相交,将该像素位置处的轨迹流量进行加一处理;重复执行上述过程,直至遍历完每条移动轨迹上的每个采样点。其中,该轨迹流量图层中包括该轨迹图像中每一个像素位置的轨迹流量。
作为一个示例,轨迹流量图层的构建方式为遍历每条移动轨迹的每个GPS点,基于前述步骤4032计算其在对应轨迹图像中的像素位置,并对该像素位置处进行流量累积。在一种可能的实现方式中,可按照下述公式计算轨迹流量:
其中,Flow
ij表示轨迹流量图层中C
ij处的流量值,其初始值为0,Traj∩C
ij表示当前移动轨迹是否与C
ij相交;因此,最后输出的轨迹流量图层的取值空间为[0,+∞]。
在本申请实施例中,轨迹流量图层的直观效果如图8所示,其中,图8中左图为真实世界的轨迹图像,右图为该轨迹图像的轨迹流量图层。另外,轨迹流量图层中越亮的位置代表该位置处的车流量越大,道路通行能力越强。
需要说明的是,上述仅是给出进行轨迹流量图层构建的一种可能实现方式,除此之外,还可采取其他方式进行轨迹流量图层构建,本申请实施例对此不进行具体限定。
4033-2、对经过预处理的轨迹数据进行移动速度图层构建。
在一种可能的实现方式中,移动速度图层的构建过程,包括:
对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在该轨迹图像中的相应像素位置,检测每一条移动轨迹是否与该像素位置相交;每当一条移动轨迹与该像素位置相交,将该移动轨迹上具有该像素位置的采样点的移动速度进行累积;基于累积的采样点总数和累积的移动速度总和,获取平均移动速度;响应于平均移动速度小于速度阈值,将该平均移动速度作为移动对象在该像素位置处的移动速度;重复执行上述过程,直至遍历完每条移动轨迹上的每个采样点。其中,移动速度图层包括移动对象在该轨迹图像中每一个像素位置处的移动速度。
需要说明的第一点是,本文中提及的移动速度通常均指代平均移动速度。
作为一个示例,移动速度图层的构建方式为遍历每条移动轨迹的每个GPS点,基于前述步骤4032计算其在对应轨迹图像中的像素位置,并对该像素位置处进行速度累积。
在一种可能的实现方式中,本申请采用的速度单位是米/秒,并设置了速度阈值;其中,速度阈值的取值可为50米/秒,本申请实施例对此不进行具体限定。假设默认移动速度的最大值为50米/秒,则本申请实施例对超过速度最大值的采样点进行数值截断(置为50),以减少异常干扰。
即,可按照下述公式计算移动流量:
其中,Speed
ij表示移动速度流量图层中C
ij处的速度值,其初始值为0,Traj∩C
ij表示当前移动轨迹是否与C
ij相交;另外,由于每个像素位置处累积了多个采样点的移动速度,因此,为了计算得到该位置C
ij处的平均移动速度,需要除以采样点个数之和,得到Traj.speed。此外,若得到的Traj.speed大于速度阈值,则将Traj.speed的大小置为50,即最后输出的移动速度图层的取值空间为[0,50]。
在本申请实施例中,移动速度图层的直观效果如图9所示,其中,图9中左图为真实世界的轨迹图像,右图为该轨迹图像的移动速度图层。另外,移动速度图层中越亮的位置代表该位置处的通行速度越大。
需要说明的是,上述仅是给出进行移动速度图层构建的一种可能实现方式,除此之外,还可采取其他方式进行移动速度图层构建,本申请实施例对此不进行具体限定。
4033-3、对经过预处理的轨迹数据进行通行方向图层构建。
在一种可能的实现方式中,通行方向图层的构建过程,包括:
对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在该轨迹图像中的相应像素位置,检测每一条移动轨迹是否与该像素位置相交;每当一条移动轨迹与该像素位置相交,将该移动轨迹上具有该像素位置的采样点的矢量方向进行累积;重复执行上述过程,直至遍历完每条移动轨迹上的每个采样点。其中,通行方向图层包括该轨迹图像中每一个像素位置的矢量方向,而矢量方向表征了移动对象的通行方向。
作为一个示例,通行方向图层的构建方式为遍历每条移动轨迹的每个GPS点,基于前述步骤4032计算其在对应轨迹图像中的像素位置,并对该像素位置处进行矢量方向累积。在一种可能的实现方式中,本申请实施例采用的方向坐标系是基于正北夹角的,可按照下述公式计算矢量方向:
在本申请实施例中,通行方向图层的直观效果如图10所示。其中,图10中左图为真实世界的轨迹图像,中图为该轨迹图像的通行方向图层,右图用于表示象限颜色。
在一种可能的实现方式中,本申请实施例将方向划分为四个象限,而落在不同象限的矢量方向采样不同的颜色展示;比如,第一象限采用红色、第二象限采用绿色、第三象限采用白色、第四象限采用蓝色。相应地,通行方向图层显示为红绿白蓝四种颜色。
需要说明的是,上述仅是给出进行通行方向图层构建的一种可能实现方式,除此之外,还可采取其他方式进行通行方向图层构建,本申请实施例对此不进行具体限定。
多通道图层融合
4034、对构建出来的轨迹流量图层、移动速度图层和通行方向图层进行融合处理。
本步骤用于将步骤4033构建的轨迹流量图层、移动速度图层和通行方向图层进行拼接合并,这里主要解决两个问题:
(1)、不同图层的取值空间差异较大,需要对其进行统一转换。
如前文所述,轨迹流量图层的取值空间为[0,+∞],移动速度图层的取值空间为[0,50],通行方向图层的取值空间为[0,360]。
(2)、方向特征属于类别型特征,在进行图层融合时本身数值大小没有意 义,直接进行拼接合并效果并不理想。
针对以上两个问题,本申请实施例会通过如下方式进行流量特征处理、速度和方向特征处理,过程如下:
流量特征处理
4034-1、对轨迹流量图层进行流量特征处理,得到第一特征图层。
在一种可能的实现方式中,对轨迹流量图层进行流量特征处理,包括:
获取流量取值上限值、第一平滑因子和第二平滑因子;获取轨迹流量图层中的最大轨迹流量;对于轨迹流量图层中的任意一个像素位置,基于该像素位置的轨迹流量、第一平滑因子、第二平滑因子、最大轨迹流量和流量取值上限值,重新获取该像素位置的轨迹流量。
作为一个示例,流量取值上限值设置为255,本申请实施例对此不进行具体限定。另外,为了能够有效表示流量变化趋势,有效分割不同道路对应的移动轨迹,本申请实施例可以通过以下公式对轨迹流量图层做进一步地优化处理:
其中,Flow
ij表示轨迹流量图层某像素位置处的流量值,max(Flow)指代该轨迹流量图层中的最大流量值,α,β表示平滑因子,取值均为常数,在本文中,α也称之为第一平滑因子,β也称之为第二平滑因子,作为一个示例,本申请实施例可分别采用α=1,β=2。
需要说明的是,上述公式将整个轨迹流量图层的取值空间从[0,+∞]转换为[0,255],同时该公式也保留了海量轨迹流量变化的纹理,区分性更强,处理后的轨迹流量图层可如图11中右图所示,其中,图11中左图示出了进行特征处理之前的轨迹流量图层。
方向、速度特征处理
4034-2、基于通行方向图层,将移动速度图层的移动速度在方向空间进行投影,得到第二特征图层。
在本申请实施例中,为了有效处理通行方向图层,本申请实施例会综合考虑速度特征和方向特征,将速度在方向空间进行投影,在保留速度特征细节的前提下,能够更合理有效地将方向特征融入到多图层通道中。
在一种可能的实现方式中,基于通行方向图层,将移动速度图层的移动速度在方向空间进行投影,包括:
对于任意一个像素位置,在移动速度图层中获取移动对象在该像素位置处的移动速度;在通行方向图层中获取移动对象在该像素位置处的矢量方向;将该移动速度与该矢量方向正弦值的乘积,作为该像素位置在纵轴方向的移动速度;将该移动速度与该矢量方向余弦值的乘积,作为该像素位置在横轴方向的移动速度。
作为一个示例,针对前述步骤4033-2得到的每个像素位置处的移动速度Speed
ij,以及前述步骤4033-2得到的每个像素位置处累积的矢量方向θ,本申请实施例将方向角度正弦值与移动速度的乘积,作为该像素位置在y轴方向的平均移动速度,将方向角度余弦值与移动速度的乘积,作为该像素位置在x轴方 向的平均移动速度;至此,x轴和y轴的计算结果的取值空间分别为[-50,50],其中,正负代表方向相反;最后,通过线性变换将该取值区间由[-50,50]转换为
[0,255],详细计算公式如下所示:
其中,Sy
ij表示y轴方向的平均移动速度,Sx
ij表示x轴方向的平均移动速度。
需要说明的第一点是,上述仅是给出进行速度和方向特征处理的一种可能实现方式,除此之外,还可采取其他方式进行速度和方向特征处理,本申请实施例对此不进行具体限定。
需要说明的第二点是,在完成上述特征处理后,不同的特征图层之间具有相同的取值空间,即均为[0,255]。
4034-3、将第一特征图层和第二特征图层进行拼接合并,得到融合特征图层。
如图12所示,其中,图12中左图为进行特征处理之前进行多通道图层融合后的效果图,右图为进行特征处理之后再进行多通道图层融合后的效果图。
可视化分析
4035、对融合特征图层进行可视化分析。
该步骤主要用于直观分析每个像素位置处的历史轨迹流量、历史平均移动速度以及历史通行方向。
其中,在显示融合特征图层时,界面元素提供的功能包括但不限于位置、尺寸、颜色、特征展示等。位置,指代响应于获取到针对融合特征图层中任一个像素位置的选中指令时,可直观展示该像素位置处的流量特征、速度特征和方向特征;尺寸,指代一个融合特征图层对应于前述的一个网格区域;颜色,指代RGB分量,如前文所述,R代表流量,G代表速度,B代表方向;特征展示,即是展示相应像素位置处的流量特征、速度特征和方向特征。
在一种可能的实现方式中,在通过可视化工具显示融合特征图层后,响应于获取到对该融合特征图层上任意一个像素位置的选中指令时,如图13所示,显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向。作为一个示例,可以显示该像素位置对应的实际地理位置处的历史轨迹流量、历史平均移动速度和历史通行方向。
另外,在展示轨迹流量、移动速度和通行方向等特征时,可在融合特征图层上叠加显示一个特征展示图层;在特征展示图层上显示轨迹流量、移动速度和通行方向,本申请实施例对此不进行具体限定。
以图13为例,对于每个像素位置来说,不同的颜色直观地表示了真实世界的不同通行情况,如图13所示,光标所在的指定像素位置处的R、G、B三分量分别为207、96、155,对应该像素位置的历史轨迹流量为3970条轨迹、车辆的历史平均移动速度为59.03公里/每小时,通行方向与正北夹角为311.13度。
本申请实施例提供的方法,至少具有以下有益效果:
本申请实施例在收集到轨迹数据后,首先对收集到的轨迹数据进行预处理;之后,对经过预处理的轨迹数据在新的三维空间内进行建模,构建出轨迹流量 图层、通行方向图层以及移动速度图层,即将轨迹数据建模为在流量、速度和方向三个维度上的新的三维空间数据,最后通过多通道图层融合技术形成一个融合特征图层,并通过可视化工具对融合特征图层进行可视化分析。
即,本申请实施例通过多通道特征融合的方式,在轨迹建模阶段综合考虑了轨迹流量、移动速度、通行方向等多个维度的属性信息,有效地保留了轨迹数据中隐含的重要特征信息,比如时空信息和相关属性,基于此,一方面,本申请实施例能够准确地挖掘移动对象在真实世界中的运动模式,即能够提取到真实运动属性,可以形成一个集时间、地理和轨迹三维可视化等为一体的轨迹建模框架,能够用于研究移动对象理论和作为真实数据集;另一方面,本申请实施例便于扩展和丰富后续轨迹挖掘,为在路网更新系统中实现简单、高效、主动式且智能化的轨迹情报挖掘提供了技术基础。
换一种表达方式,本申请实施例在海量轨迹数据的基础上,不但能够精确挖掘移动对象在真实世界中的运动模式,而且还能够提升后续的轨迹挖掘效果。
另外,在通过可视化工具显示融合特征图层后,响应于获取到对该融合特征图层上任意一个像素位置的选中指令,还能够直观显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向,即本申请实施例能够直观分析每个像素位置处的历史轨迹流量、历史平均移动速度以及历史通行方向,效果较佳。
在另一个实施例中,如图2所示,在通过轨迹建模阶段处理后,经过轨迹建模的轨迹数据会用于轨迹挖掘。即,参见图14,本申请实施例提供的方法流程还包括:
1401、获取目标地理范围内的轨迹数据。
1402、对获取到的轨迹数据进行预处理。
1403、对经过预处理的轨迹数据进行特征建模。
1404、基于得到的特征建模结果进行轨迹挖掘。
其中,本步骤用于结合不同的挖掘模式实现对轨迹数据的深入研究。
在一种可能的实现方式中,如图2所示,轨迹挖掘模式包括但不限于频繁模式、分类模式、回归模式以及聚类模式,本申请实施例对此不进行具体限定。
其中,频繁模式来源于频繁项集,频繁项集的含义为:有一系列集合,这些集合有些相同的元素,集合中同时出现频率高的元素形成一个子集,满足一定阈值条件,即是频繁项集。该种模式可基于轨迹数据进行用户画像刻画。
分类模式用于对轨迹数据进行诸如分类或基于轨迹数据对用户进行分类等,本申请实施例对此不进行具体限定。回归模式可用于基于轨迹数据对道路的拥挤程度进行预测等。聚类模式可基于轨迹数据对用户经常出现的场所进行预测。
本申请实施例提供的方法,在收集到轨迹数据后,首先对收集到的轨迹数据进行预处理;之后,对经过预处理的轨迹数据在新的三维空间内进行建模,构建出轨迹流量图层、通行方向图层以及移动速度图层,即将轨迹数据建模为在流量、速度和方向三个维度上的新的三维空间数据,并通过多通道图层融合技术形成一个融合特征图层,且通过可视化工具对融合特征图层进行可视化分析。
即,本申请实施例通过多通道特征融合的方式,在轨迹建模阶段综合考虑了 轨迹流量、移动速度、通行方向等多个维度的属性信息,有效地保留了轨迹数据中隐含的重要特征信息,比如时空信息和相关属性,基于此,一方面,本申请实施例能够准确地挖掘移动对象在真实世界中的运动模式,即能够提取到真实运动属性,可以形成一个集时间、地理和轨迹三维可视化等为一体的轨迹建模框架,能够用于研究移动对象理论和作为真实数据集;另一方面,本申请实施例便于扩展和丰富后续轨迹挖掘,为在路网更新系统中实现简单、高效、主动式且智能化的轨迹情报挖掘提供了技术基础。
换一种表达方式,本申请实施例在海量轨迹数据的基础上,不但能够精确挖掘移动对象在真实世界中的运动模式,而且还能够提升后续的轨迹挖掘效果。
在另一个实施例中,本申请实施例提供的轨迹数据处理方法,还包括如图15所示的方法流程:
1501、获取目标地理范围内的轨迹数据,其中,轨迹数据中包括多个移动对象的移动轨迹。
1502、生成轨迹数据的特征图层,特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层。
1503、对特征图层中包括的各个图层进行融合处理,得到融合特征图层。
1504、显示融合特征图层;响应于获取到对融合特征图层上任意一个像素位置的选中指令,显示像素位置处的轨迹流量、移动对象的移动速度和通行方向。
本申请实施例在收集到轨迹数据后,对轨迹数据在新的三维空间内进行建模,构建出轨迹流量图层、通行方向图层以及移动速度图层,即将轨迹数据建模为在流量、速度和方向三个维度上的新的三维空间数据,之后,通过多通道图层融合技术形成一个融合特征图层,并对融合特征图层进行可视化分析。
即,本申请实施例通过多通道特征融合的方式,在轨迹建模阶段综合考虑了轨迹流量、移动速度、通行方向等多个维度的属性信息,有效地保留了轨迹数据中隐含的重要特征信息,比如时空信息和相关属性,基于此,一方面,本申请实施例能够准确地挖掘移动对象在真实世界中的运动模式,即能够提取到真实运动属性,可以形成一个集时间、地理和轨迹三维可视化等为一体的轨迹建模框架,能够用于研究移动对象理论和作为真实数据集;另一方面,本申请实施例便于扩展和丰富后续轨迹挖掘,为在路网更新系统中实现简单、高效、主动式且智能化的轨迹情报挖掘提供了技术基础。
换一种表达方式,本申请实施例在海量轨迹数据的基础上,不但能够精确挖掘移动对象在真实世界中的运动模式,而且还能够提升后续的轨迹挖掘效果。
另外,在显示融合特征图层后,响应于获取到对该融合特征图层上任意一个像素位置的选中指令,还能够直观显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向,即本申请实施例能够直观分析每个像素位置处的轨迹流量、移动对象的移动速度以及通行方向,效果较佳。
在另一个实施例中,本申请实施例提供的轨迹数据处理方法,还包括如图16 所示的方法流程:
1601、显示特征图层,该特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层。
其中,本步骤中的特征图层即指代前述的融合特征图层,即本步骤中的特征图层为轨迹数据的特征图层中包括的各个图层的融合结果,各个图层至少包括轨迹流量图层、移动速度图层和通行方向图层。
需要说明的是,本步骤的融合特征图层即为经过前述步骤4034的融合处理得到的特征图层。即,在获取到目标地理范围内的轨迹数据后,生成该轨迹数据的特征图层,通过对该特征图层中包括的各个图层进行融合处理,得到上述融合特征图层。其中,该轨迹数据中包括多个移动对象的移动轨迹。
在一种可能的实现方式中,参加图3和图13,在融合特征图层中会以不同的颜色来分别表征轨迹流量、移动速度和通行方向。比如,本申请实施例采用红色R代表轨迹流量,采用绿色G代表移动速度,采用蓝色B代表通行方向。
其中,在显示融合特征图层时,界面元素提供的功能包括但不限于位置、尺寸、颜色、特征展示等。位置,指代响应于获取到针对融合特征图层中任一个像素位置的选中指令时,可直观展示该像素位置处的流量特征、速度特征和方向特征;尺寸,指代一个融合特征图层对应于前述的一个网格区域;颜色,指代RGB分量,如前文所述,R代表流量,G代表速度,B代表方向;特征展示,即是展示相应像素位置处的流量特征、速度特征和方向特征。
1602、响应于获取到对融合特征图层上任意一个像素位置的选中指令,显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向。
在一种可能的实现方式中,显示像素位置处的轨迹流量、移动对象的移动速度和通行方向,包括但不限于:在融合特征图层上叠加显示一个特征展示图层;在特征展示图层上显示像素位置处的轨迹流量、移动对象的移动速度和通行方向。
以图13为例,对于每个像素位置来说,不同的颜色直观地表示了真实世界的不同通行情况,如图13所示,光标所在的指定像素位置处的R、G、B三分量分别为207、96、155,对应该像素位置的历史轨迹流量为3970条轨迹、车辆的历史平均移动速度为59.03公里/每小时,通行方向与正北夹角为311.13度。
另外,对比图12中右图和图13,还可对融合特征图层进行局部放大显示,本申请实施例对此不进行具体限定。
本申请实施例提供的方法,可以对轨迹数据进行可视化分析,比如在显示特征图层后,响应于获取到对该特征图层上任意一个像素位置的选中指令,能够直观显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向,即本申请实施例能够直观分析每个像素位置处的流量特征、速度特征以及方向特征,效果较佳。
在另一个实施例中,除了采取上述描述的方式对轨迹数据进行处理外,还可采取下述的处理方式:
(a)、本申请实施例是直接针对收集到的海量移动轨迹进行建模,而在某些 轨迹挖掘应用场景下可能需要关注特定时段,因此本申请实施例也可以扩展为对海量移动轨迹进行分时段分别建模。
(b)、本申请实施例在对海量移动轨迹进行建模时,考虑了流量、速度和方向等特征,后续也可以加入更多特征,本申请实施例对此不进行具体限定。
(c)、在多特征图层构建时,对各个特征图层的特征处理方式并不唯一,除了前述的特征处理方式之外,可以根据实际应用需求做自适应调整,采取其他的特征处理方式,本申请实施例对此不进行具体限定。
(d)、在进行可视化分析过程中,还可以加入更多的分析元素,比如周围平均速度分布、流量分布等等,本申请实施例对此不进行具体限定。
图17是本申请实施例提供的一种轨迹数据处理装置的结构示意图。参见图17,该装置包括:
获取模块1701,用于获取目标地理范围内的轨迹数据,所述轨迹数据中包括多个移动对象的移动轨迹;
生成模块1702,用于生成所述轨迹数据的特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;
处理模块1703,用于对所述特征图层中包括的各个图层进行融合处理,得到融合特征图层;
显示模块1704,用于显示融合特征图层;响应于获取到对融合特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
本申请实施例提供的装置,在获取到目标地理范围内的轨迹数据后,首先生成该轨迹数据的特征图层,其中,该特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层,即本申请实施例是对轨迹数据在新的三维空间内进行建模,也即将轨迹数据建模为在流量、速度和方向三个维度上的新的三维空间数据,之后,对各个图层进行融合处理形成一个融合特征图层,并对融合特征图层进行可视化分析。
即,本申请实施例通过多通道特征融合的方式,在轨迹建模阶段综合考虑了轨迹流量、移动速度、通行方向等多个维度的属性信息,有效地保留了轨迹数据中隐含的重要特征信息,因此,本申请实施例不但能够精确挖掘移动对象在真实世界中的运动模式,而且还能够提升后续的轨迹挖掘效果。
另外,在显示融合特征图层后,响应于获取到对该融合特征图层上任意一个像素位置的选中指令时,还能够直观显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向,即本申请实施例能够直观分析每个像素位置处的轨迹流量、移动对象的移动速度以及通行方向,效果较佳。
在一种可能的实现方式中,显示模块,用于在所述融合特征图层上叠加显示一个特征展示图层;在所述特征展示图层上显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
在一种可能的实现方式中,该装置还包括:
缩放模块,用于将每一个网格区域分别缩放为一个统一尺寸大小的轨迹图 像;
转换模块,用于对于任意一个网格区域,对所述网格区域包括的每条移动轨迹进行坐标系转换,得到每条移动轨迹上各个采样点在相应轨迹图像中的像素位置;
其中,所述网格区域是对所述目标地理范围按照空间进行切分得到的,所述轨迹数据中包括的各条移动轨迹是按照所属网格区域进行划分的。
在一种可能的实现方式中,转换模块,用于获取所述网格区域的区域边界的经纬度坐标;获取相应轨迹图像的高度数值和宽度数值;对于任意一个采样点,根据所述网格区域的左右边界的经纬度坐标、所述宽度数值和所述采样点的经度坐标,获取所述采样点在相应轨迹图像中的横轴坐标;根据所述网格区域的上下边界的经纬度坐标、所述高度数值和所述采样点的纬度坐标,获取所述采样点在相应轨迹图像中的纵轴坐标。
在一种可能的实现方式中,响应于所述特征图层包括所述轨迹流量图层,生成模块,用于对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述像素位置处的轨迹流量进行加一处理;
其中,所述轨迹流量图层包括所述轨迹图像中每一个像素位置的轨迹流量。
在一种可能的实现方式中,响应于所述特征图层包括所述移动速度图层,生成模块,还用于对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的移动速度进行累积;基于累积的采样点总数和累积的移动速度总和,获取平均移动速度;响应于所述平均移动速度小于速度阈值,将所述平均移动速度作为移动对象在所述像素位置处的移动速度;
其中,所述移动速度图层包括移动对象在所述轨迹图像中每一个像素位置处的移动速度。
在一种可能的实现方式中,响应于所述特征图层包括所述通行方向图层,生成模块,还用于对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的矢量方向进行累积;
其中,所述通行方向图层包括所述轨迹图像中每一个像素位置的矢量方向,所述矢量方向表征了移动对象的通行方向。
在一种可能的实现方式中,所述显示模块,用于显示所述像素位置对应的地理位置处的历史轨迹流量、移动对象的历史平均移动速度和历史通行方向。
在一种可能的实现方式中,处理模块,用于对所述轨迹流量图层进行流量特征处理,得到第一特征图层;基于所述通行方向图层,将所述移动速度图层 的移动速度在方向空间进行投影,得到第二特征图层;将所述第一特征图层和所述第二特征图层进行拼接合并,得到所述融合特征图层;
其中,所述第一特征图层和所述第二特征图层具有相同的取值空间。
在一种可能的实现方式中,处理模块,用于获取流量取值上限值、第一平滑因子和第二平滑因子;获取所述轨迹流量图层中的最大轨迹流量;对于所述轨迹流量图层中的任意一个像素位置,基于所述像素位置的轨迹流量、所述第一平滑因子、所述第二平滑因子、所述最大轨迹流量和所述流量取值上限值,重新获取所述像素位置的轨迹流量。
在一种可能的实现方式中,处理模块,用于对于任意一个像素位置,在所述移动速度图层中获取移动对象在所述像素位置处的移动速度;在所述通行方向图层中获取移动对象在所述像素位置处的矢量方向;将所述移动速度与所述矢量方向正弦值的乘积,作为所述像素位置在纵轴方向的移动速度;将所述移动速度与所述矢量方向余弦值的乘积,作为所述像素位置在横轴方向的移动速度。
在一种可能的实现方式中,该装置还包括:
预处理模块,用于在获取到所述轨迹数据后,对所述轨迹数据进行预处理;
其中,预处理包括以下至少一种:
将所述目标地理范围按照空间切分为多个网格区域,将所述轨迹数据中包括的各条移动轨迹按照所属网格区域进行划分;
按照设置的过滤条件,对所述轨迹数据中包括各条移动轨迹进行过滤处理;
对所述轨迹数据中包括的各条移动轨迹进行异常剔除处理;
对所述轨迹数据中包括的各条移动轨迹进行数据插值处理。
上述所有可选技术方案,可以采用任意结合形成本公开的可选实施例,在此不再一一赘述。
图18是本申请实施例提供的一种轨迹数据处理装置的结构示意图。参见图18,该装置包括:
第一显示模块1801,用于显示特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;
第二显示模块1802,用于响应于获取到对所述特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
本申请实施例提供的装置,可以对轨迹数据进行可视化分析,比如在显示特征图层后,响应于获取到对该特征图层上任意一个像素位置的选中指令,能够直观显示该像素位置处的轨迹流量、移动对象的移动速度和通行方向,即本申请实施例能够直观分析每个像素位置处的流量特征、速度特征以及方向特征,效果较佳。
在一种可能的实现方式中,所述特征图层中以不同的颜色分别表征轨迹流量、移动速度和通行方向。
在一种可能的实现方式中,第二显示模块,还用于在所述特征图层上叠加 显示一个特征展示图层;在所述特征展示图层上显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
上述所有可选技术方案,可以采用任意结合形成本公开的可选实施例,在此不再一一赘述。
需要说明的是:上述实施例提供的轨迹数据处理装置在处理轨迹数据时,仅以上述各功能模块的划分进行举例说明,实际应用中,可以根据需要而将上述功能分配由不同的功能模块完成,即将装置的内部结构划分成不同的功能模块,以完成以上描述的全部或者部分功能。另外,上述实施例提供的轨迹数据处理装置与轨迹数据处理方法实施例属于同一构思,其具体实现过程详见方法实施例,这里不再赘述。
图19是本申请实施例提供的一种轨迹数据处理设备的结构示意图。
该设备1900可因配置或性能不同而产生比较大的差异,可以包括一个或一个以上处理器(Central Processing Units,CPU)1901和一个或一个以上的存储器1902,其中,所述存储器1902中存储有至少一条指令,所述至少一条指令由所述处理器1901加载并执行以实现上述各个方法实施例提供的轨迹数据处理方法。当然,该设备还可以具有有线或无线网络接口、键盘以及输入输出接口等部件,以便进行输入输出,该设备还可以包括其他用于实现设备功能的部件,在此不做赘述。
在示例性实施例中,还提供了一种计算机可读存储介质,例如包括指令的存储器,上述指令可由终端中的处理器执行以完成上述实施例中的轨迹数据处理方法。例如,所述计算机可读存储介质可以是ROM(Read Only Memory,只读存储器)、RAM(Random Access Memory,随机存取存储器)、CD-ROM(Compact Disc Read-Only Memory,光盘只读存储器)、磁带、软盘和光数据存储设备等。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成,也可以通过程序来指令相关的硬件完成,所述的程序可以存储于一种计算机可读存储介质中,上述提到的存储介质可以是只读存储器,磁盘或光盘等。
以上所述仅为本申请的较佳实施例,并不用以限制本申请,凡在本申请的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本申请的保护范围之内。
Claims (25)
- 一种轨迹数据处理方法,其特征在于,所述方法应用于轨迹数据处理设备,所述方法包括:获取目标地理范围内的轨迹数据,所述轨迹数据中包括多个移动对象的移动轨迹;生成所述轨迹数据的特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;对所述特征图层中包括的各个图层进行融合处理,得到融合特征图层;显示所述融合特征图层;响应于获取到对所述融合特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
- 根据权利要求1所述的方法,其特征在于,所述显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向,包括:在所述融合特征图层上叠加显示一个特征展示图层;在所述特征展示图层上显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
- 根据权利要求1所述的方法,其特征在于,所述方法还包括:将每一个网格区域分别缩放为一个统一尺寸大小的轨迹图像;对于任意一个网格区域,对所述网格区域包括的每条移动轨迹进行坐标系转换,得到每条移动轨迹上各个采样点在相应轨迹图像中的像素位置;其中,所述网格区域是对所述目标地理范围按照空间进行切分得到的,所述轨迹数据中包括的各条移动轨迹是按照所属网格区域进行划分的。
- 根据权利要求3所述的方法,其特征在于,所述对所述网格区域包括的每条移动轨迹进行坐标系转换,得到每条移动轨迹上各个采样点在相应轨迹图像中的像素位置,包括:获取所述网格区域的区域边界的经纬度坐标;获取相应轨迹图像的高度数值和宽度数值;对于任意一个采样点,根据所述网格区域的左右边界的经纬度坐标、所述宽度数值和所述采样点的经度坐标,获取所述采样点在相应轨迹图像中的横轴坐标;根据所述网格区域的上下边界的经纬度坐标、所述高度数值和所述采样点的纬度坐标,获取所述采样点在相应轨迹图像中的纵轴坐标。
- 根据权利要求3所述的方法,其特征在于,响应于所述特征图层包括所述轨迹流量图层,所述生成所述轨迹数据的特征图层,包括:对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述像素位置处的轨迹流量进行加一处理;其中,所述轨迹流量图层包括所述轨迹图像中每一个像素位置的轨迹流量。
- 根据权利要求3所述的方法,其特征在于,响应于所述特征图层包括所述移动速度图层,所述生成所述轨迹数据的特征图层,包括:对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的移动速度进行累积;基于累积的采样点总数和累积的移动速度总和,获取平均移动速度;响应于所述平均移动速度小于速度阈值,将所述平均移动速度作为移动对象在所述像素位置处的移动速度;其中,所述移动速度图层包括移动对象在所述轨迹图像中每一个像素位置处的移动速度。
- 根据权利要求3所述的方法,其特征在于,响应于所述特征图层包括所述通行方向图层,所述生成所述轨迹数据的特征图层,包括:对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的矢量方向进行累积;其中,所述通行方向图层包括所述轨迹图像中每一个像素位置的矢量方向,所述矢量方向表征了移动对象的通行方向。
- 根据权利要求1所述的方法,其特征在于,所述对所述特征图层中包括的各个图层进行融合处理,得到融合特征图层,包括:对所述轨迹流量图层进行流量特征处理,得到第一特征图层;基于所述通行方向图层,将所述移动速度图层的移动速度在方向空间进行投影,得到第二特征图层;将所述第一特征图层和所述第二特征图层进行拼接合并,得到所述融合特征图层;其中,所述第一特征图层和所述第二特征图层具有相同的取值空间。
- 根据权利要求1所述的方法,其特征在于,所述显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向,包括:显示所述像素位置对应的地理位置处的历史轨迹流量、移动对象的历史平均移动速度和历史通行方向。
- 根据权利要求8所述的方法,其特征在于,所述对所述轨迹流量图层进行流量特征处理,包括:获取流量取值上限值、第一平滑因子和第二平滑因子;获取所述轨迹流量图层中的最大轨迹流量;对于所述轨迹流量图层中的任意一个像素位置,基于所述像素位置的轨迹流量、所述第一平滑因子、所述第二平滑因子、所述最大轨迹流量和所述流量取值上限值,重新获取所述像素位置的轨迹流量。
- 根据权利要求8所述的方法,其特征在于,所述基于所述通行方向图层,将所述移动速度图层的移动速度在方向空间进行投影,包括:对于任意一个像素位置,在所述移动速度图层中获取移动对象在所述像素位置处的移动速度;在所述通行方向图层中获取移动对象在所述像素位置处的矢量方向;将所述移动速度与所述矢量方向正弦值的乘积,作为所述像素位置在纵轴方向的移动速度;将所述移动速度与所述矢量方向余弦值的乘积,作为所述像素位置在横轴方向的移动速度。
- 根据权利要求1至11中任一个权利要求所述的方法,其特征在于,所述方法还包括:在获取到所述轨迹数据后,对所述轨迹数据进行预处理;其中,预处理包括以下至少一种:将所述目标地理范围按照空间切分为多个网格区域,将所述轨迹数据中包括的各条移动轨迹按照所属网格区域进行划分;按照过滤条件,对所述轨迹数据中包括的各条移动轨迹进行过滤处理;对所述轨迹数据中包括的各条移动轨迹进行异常剔除处理;对所述轨迹数据中包括的各条移动轨迹进行数据插值处理。
- 一种轨迹数据处理设备,其特征在于,所述设备包括处理器和存储器,所述存储器中存储有至少一条指令,所述至少一条指令由所述处理器加载并执行以实现如下步骤:获取目标地理范围内的轨迹数据,所述轨迹数据中包括多个移动对象的移动轨迹;生成所述轨迹数据的特征图层,所述特征图层至少包括轨迹流量图层、移动速度图层和通行方向图层;对所述特征图层中包括的各个图层进行融合处理,得到融合特征图层;显示所述融合特征图层;响应于获取到对所述融合特征图层上任意一个像素位置的选中指令,显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
- 根据权利要求13所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:在所述融合特征图层上叠加显示一个特征展示图层;在所述特征展示图层上显示所述像素位置处的轨迹流量、移动对象的移动速度和通行方向。
- 根据权利要求13所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:将每一个网格区域分别缩放为一个统一尺寸大小的轨迹图像;对于任意一个网格区域,对所述网格区域包括的每条移动轨迹进行坐标系转换,得到每条移动轨迹上各个采样点在相应轨迹图像中的像素位置;其中,所述网格区域是对所述目标地理范围按照空间进行切分得到的,所述轨迹数据中包括的各条移动轨迹是按照所属网格区域进行划分的。
- 根据权利要求15所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:获取所述网格区域的区域边界的经纬度坐标;获取相应轨迹图像的高度数值和宽度数值;对于任意一个采样点,根据所述网格区域的左右边界的经纬度坐标、所述宽度数值和所述采样点的经度坐标,获取所述采样点在相应轨迹图像中的横轴坐标;根据所述网格区域的上下边界的经纬度坐标、所述高度数值和所述采样点的纬度坐标,获取所述采样点在相应轨迹图像中的纵轴坐标。
- 根据权利要求15所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述像素位置处的轨迹流量进行加一处理;其中,所述轨迹流量图层包括所述轨迹图像中每一个像素位置的轨迹流量。
- 根据权利要求15所述的设备,其特征在于,响应于所述特征图层包括 所述移动速度图层,所述至少一条指令由所述处理器加载并执行以实现如下步骤:对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的移动速度进行累积;基于累积的采样点总数和累积的移动速度总和,获取平均移动速度;响应于所述平均移动速度小于速度阈值,将所述平均移动速度作为移动对象在所述像素位置处的移动速度;其中,所述移动速度图层包括移动对象在所述轨迹图像中每一个像素位置处的移动速度。
- 根据权利要求15所述的设备,其特征在于,响应于所述特征图层包括所述通行方向图层,所述至少一条指令由所述处理器加载并执行以实现如下步骤:对于任意一个轨迹图像中的每一个像素位置,基于相应网格区域包括的每条移动轨迹上各个采样点在所述轨迹图像中的相应像素位置,检测每一条所述移动轨迹是否与所述像素位置相交;每当一条所述移动轨迹与所述像素位置相交,将所述移动轨迹上具有所述像素位置的采样点的矢量方向进行累积;其中,所述通行方向图层包括所述轨迹图像中每一个像素位置的矢量方向,所述矢量方向表征了移动对象的通行方向。
- 根据权利要求13所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:对所述轨迹流量图层进行流量特征处理,得到第一特征图层;基于所述通行方向图层,将所述移动速度图层的移动速度在方向空间进行投影,得到第二特征图层;将所述第一特征图层和所述第二特征图层进行拼接合并,得到所述融合特征图层;其中,所述第一特征图层和所述第二特征图层具有相同的取值空间。
- 根据权利要求13所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:显示所述像素位置对应的地理位置处的历史轨迹流量、移动对象的历史平均移动速度和历史通行方向。
- 根据权利要求20所述的设备,其特征在于,所述至少一条指令由所述 处理器加载并执行以实现如下步骤:获取流量取值上限值、第一平滑因子和第二平滑因子;获取所述轨迹流量图层中的最大轨迹流量;对于所述轨迹流量图层中的任意一个像素位置,基于所述像素位置的轨迹流量、所述第一平滑因子、所述第二平滑因子、所述最大轨迹流量和所述流量取值上限值,重新获取所述像素位置的轨迹流量。
- 根据权利要求20所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:对于任意一个像素位置,在所述移动速度图层中获取移动对象在所述像素位置处的移动速度;在所述通行方向图层中获取移动对象在所述像素位置处的矢量方向;将所述移动速度与所述矢量方向正弦值的乘积,作为所述像素位置在纵轴方向的移动速度;将所述移动速度与所述矢量方向余弦值的乘积,作为所述像素位置在横轴方向的移动速度。
- 根据权利要求13至23中任一个权利要求所述的设备,其特征在于,所述至少一条指令由所述处理器加载并执行以实现如下步骤:在获取到所述轨迹数据后,对所述轨迹数据进行预处理;其中,预处理包括以下至少一种:将所述目标地理范围按照空间切分为多个网格区域,将所述轨迹数据中包括的各条移动轨迹按照所属网格区域进行划分;按照过滤条件,对所述轨迹数据中包括的各条移动轨迹进行过滤处理;对所述轨迹数据中包括的各条移动轨迹进行异常剔除处理;对所述轨迹数据中包括的各条移动轨迹进行数据插值处理。
- 一种存储介质,其特征在于,所述存储介质中存储有至少一条指令,所述至少一条指令由处理器加载并执行以实现如权利要求1至12中任一个权利要求所述的轨迹数据处理方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP20848404.8A EP4006846A4 (en) | 2019-07-29 | 2020-06-18 | TRAJECTORY DATA PROCESSING METHOD, DEVICE AND STORAGE MEDIA |
US17/394,389 US20210366130A1 (en) | 2019-07-29 | 2021-08-04 | Trajectory data processing method and device, and storage medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910690467.6 | 2019-07-29 | ||
CN201910690467.6A CN110428500B (zh) | 2019-07-29 | 2019-07-29 | 轨迹数据处理方法、装置、存储介质以及设备 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US17/394,389 Continuation US20210366130A1 (en) | 2019-07-29 | 2021-08-04 | Trajectory data processing method and device, and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021017675A1 true WO2021017675A1 (zh) | 2021-02-04 |
Family
ID=68413019
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/096747 WO2021017675A1 (zh) | 2019-07-29 | 2020-06-18 | 轨迹数据处理方法、设备以及存储介质 |
Country Status (4)
Country | Link |
---|---|
US (1) | US20210366130A1 (zh) |
EP (1) | EP4006846A4 (zh) |
CN (1) | CN110428500B (zh) |
WO (1) | WO2021017675A1 (zh) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110428500B (zh) * | 2019-07-29 | 2022-07-12 | 腾讯科技(深圳)有限公司 | 轨迹数据处理方法、装置、存储介质以及设备 |
CN113811830B (zh) * | 2019-12-30 | 2022-05-10 | 深圳元戎启行科技有限公司 | 轨迹预测方法、装置、计算机设备和存储介质 |
CN112017436B (zh) * | 2020-09-09 | 2021-09-28 | 中国科学院自动化研究所 | 城市市内交通旅行时间的预测方法及系统 |
CN112632208B (zh) * | 2020-12-25 | 2022-12-16 | 际络科技(上海)有限公司 | 一种交通流轨迹变形方法及装置 |
CN112598767B (zh) * | 2020-12-29 | 2024-05-10 | 厦门市美亚柏科信息股份有限公司 | 基于时空大数据的轨迹行为分析方法、终端设备及存储介质 |
CN112633602B (zh) * | 2020-12-31 | 2023-03-03 | 深圳云天励飞技术股份有限公司 | 一种基于gis地图信息的交通拥堵指数预测方法及装置 |
CN112783967A (zh) * | 2021-02-01 | 2021-05-11 | 北京海天维景科技有限公司 | 一种水下目标运动动态的可视化方法和装置 |
CN114490601A (zh) * | 2021-12-31 | 2022-05-13 | 北京中交兴路信息科技有限公司 | 一种车辆轨迹测试方法和系统 |
CN114661839B (zh) * | 2022-01-12 | 2024-02-27 | 中国人民解放军32200部队 | 一种基于瓦片的轨迹地图的展示方法、装置、介质和设备 |
CN114638912A (zh) * | 2022-03-16 | 2022-06-17 | 广州博进信息技术有限公司 | 海洋科学矢量数据的时空可视化与交互方法、介质及设备 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102332210A (zh) * | 2011-08-04 | 2012-01-25 | 东南大学 | 基于手机定位数据的实时城市道路交通流数据提取方法 |
US8694241B1 (en) * | 2010-10-05 | 2014-04-08 | Google Inc. | Visualization of traffic patterns using GPS data |
CN105808912A (zh) * | 2014-12-31 | 2016-07-27 | 中国科学院深圳先进技术研究院 | 一种交通轨迹数据的可视分析方法、系统 |
CN106157624A (zh) * | 2016-08-04 | 2016-11-23 | 浙江工业大学 | 基于交通定位数据的多粒度道路分流可视分析方法 |
CN110428500A (zh) * | 2019-07-29 | 2019-11-08 | 腾讯科技(深圳)有限公司 | 轨迹数据处理方法、装置、存储介质以及设备 |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
AU785021C (en) * | 2001-07-03 | 2007-06-28 | Mark Thomas Dawson | Improvements in and relating to method and apparatus for producing anaglyphic 3-D images |
AU2013200609A1 (en) * | 2008-04-11 | 2013-02-21 | Terraspark Geosciences, Llc | Visulation of geologic features using data representations thereof |
US20100033482A1 (en) * | 2008-08-11 | 2010-02-11 | Interactive Relighting of Dynamic Refractive Objects | Interactive Relighting of Dynamic Refractive Objects |
US9818211B1 (en) * | 2013-04-25 | 2017-11-14 | Domo, Inc. | Automated combination of multiple data visualizations |
CN104700624B (zh) * | 2015-03-16 | 2017-07-07 | 电子科技大学 | 基于相敏光时域反射仪的车流量在线监测系统的监测方法 |
CN106251620B (zh) * | 2016-09-26 | 2019-01-25 | 北京东土科技股份有限公司 | 基于智能交通云控制系统的中心系统 |
CN108133611A (zh) * | 2016-12-01 | 2018-06-08 | 中兴通讯股份有限公司 | 车辆行驶轨迹监测方法及系统 |
CN108507586A (zh) * | 2017-02-28 | 2018-09-07 | 北京嘀嘀无限科技发展有限公司 | 显示对象的移动展示方法、装置及存储介质 |
US10782694B2 (en) * | 2017-09-07 | 2020-09-22 | Tusimple, Inc. | Prediction-based system and method for trajectory planning of autonomous vehicles |
CN107770721A (zh) * | 2017-10-10 | 2018-03-06 | 东南大学 | 一种旅游交通客流大数据可视化方法 |
FR3090544B1 (fr) * | 2018-12-21 | 2021-01-15 | Renault Sas | Dispositif et procédé de contrôle de trajectoire d’un véhicule automobile |
-
2019
- 2019-07-29 CN CN201910690467.6A patent/CN110428500B/zh active Active
-
2020
- 2020-06-18 WO PCT/CN2020/096747 patent/WO2021017675A1/zh unknown
- 2020-06-18 EP EP20848404.8A patent/EP4006846A4/en active Pending
-
2021
- 2021-08-04 US US17/394,389 patent/US20210366130A1/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8694241B1 (en) * | 2010-10-05 | 2014-04-08 | Google Inc. | Visualization of traffic patterns using GPS data |
CN102332210A (zh) * | 2011-08-04 | 2012-01-25 | 东南大学 | 基于手机定位数据的实时城市道路交通流数据提取方法 |
CN105808912A (zh) * | 2014-12-31 | 2016-07-27 | 中国科学院深圳先进技术研究院 | 一种交通轨迹数据的可视分析方法、系统 |
CN106157624A (zh) * | 2016-08-04 | 2016-11-23 | 浙江工业大学 | 基于交通定位数据的多粒度道路分流可视分析方法 |
CN110428500A (zh) * | 2019-07-29 | 2019-11-08 | 腾讯科技(深圳)有限公司 | 轨迹数据处理方法、装置、存储介质以及设备 |
Non-Patent Citations (2)
Title |
---|
See also references of EP4006846A4 * |
WANG, SHENGCE: "Design of a Visualization Solution for Analyzing Spatio-Temporal Cycling Trajectory Data: Case Study of Fietstelweek Data and BikePRINT", MASTER THESIS, 15 February 2018 (2018-02-15), pages 1 - 96, XP009525718, ISSN: 1674-0246 * |
Also Published As
Publication number | Publication date |
---|---|
EP4006846A4 (en) | 2022-10-26 |
EP4006846A1 (en) | 2022-06-01 |
CN110428500A (zh) | 2019-11-08 |
US20210366130A1 (en) | 2021-11-25 |
CN110428500B (zh) | 2022-07-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021017675A1 (zh) | 轨迹数据处理方法、设备以及存储介质 | |
US11222465B2 (en) | Embedded urban design scene emulation method and system | |
US10545029B2 (en) | Lane network construction using high definition maps for autonomous vehicles | |
CN108921943B (zh) | 一种基于车道级高精度地图的道路三维模型建模方法 | |
Richter et al. | Concepts and techniques for integration, analysis and visualization of massive 3D point clouds | |
CN110274602A (zh) | 室内地图自动构建方法及系统 | |
CN112115198A (zh) | 一种城市遥感智能服务平台 | |
CN109086286B (zh) | 一种彩色地形图生产与发布方法 | |
Zhao et al. | Statistical analysis on the evolution of OpenStreetMap road networks in Beijing | |
CN110716935A (zh) | 基于网约车出行的轨迹数据分析与可视化方法及系统 | |
CN104134121A (zh) | 一种电网信息系统业务数据可视化的实现方法 | |
CN105427210A (zh) | 一种旅游资源信息采集与评价网络系统 | |
CN114238542A (zh) | 多源交通gis路网多层级实时融合更新方法 | |
Khayyal et al. | Creation and spatial analysis of 3D city modeling based on GIS data | |
CN110660125A (zh) | 用于电力配网系统的三维建模装置 | |
CN112066997A (zh) | 高清航线地图的导出方法及系统 | |
Doulamis et al. | EasyTracker: An Android application for capturing mobility behavior | |
Schoier et al. | Spatial data mining for highlighting hotspots in personal navigation routes | |
CN112017287A (zh) | 一种地下电缆设施三维参数化建模及高效渲染方法 | |
CN114661744A (zh) | 一种基于深度学习的地形数据库更新方法及系统 | |
CN112097776A (zh) | 用于提高航线地图渲染效率的方法及系统 | |
CN112287425A (zh) | 一种市政工程测量定位方法及系统 | |
CN111382165A (zh) | 一种移动国土管理系统 | |
Zou et al. | An overview of geospatial information visualization | |
Kim et al. | Stinuum: A holistic visual analysis of moving objects with open source software |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20848404 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2020848404 Country of ref document: EP Effective date: 20220228 |