Tourism scene modeling interaction method and system based on virtual reality

Abstract

The invention relates to the technical field of virtual scene building, and particularly discloses a tourism scene modeling interaction method and system based on virtual reality, and the method comprises the steps: obtaining a scene video in real time, carrying out the recognition of the scene video, and determining the visitor flow of each position in a tourism scene; determining the tourism value of each position according to the visitor flow of each position; combining the positions according to the tourism value, synchronously calculating an average tourism value, and generating a tourism area containing the average tourism value; and establishing a VR scene of the tourism scene, marking a tourism area containing the average tourism value in the VR scene, and sending the tourism area to the user side. According to the invention, the visitor flow information is obtained through the monitoring system in the tourism scene, the tourism value of each area in the tourism scene is determined according to the visitor flow information, and the tourism value of each area is displayed based on the request of the user in the virtual scene building process, thereby providing guidance for the virtual tourism process.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual scene construction, and in particular to a tourism scene modeling interaction method and system based on virtual reality. Background Art

[0002] The application of virtual reality (VR) technology in the tourism industry is developing rapidly, especially in the display of scenic spots, virtual tourism, and support for tourism decision-making. Tourism scene modeling based on virtual reality can not only provide tourists with an immersive experience, but also in many cases, help tourists understand the attractions in advance before the actual travel, thereby enhancing the travel experience.

[0003] Existing virtual reality technology can replicate tourist scenes, but some additional information will not be displayed, making it difficult for users to fully understand the tourist scenes and plan their travel routes. How to add some scene information to the tourist scenes to provide guidance for users is a technical problem that the technical solution of the present invention aims to solve. Summary of the invention

[0004] The purpose of the present invention is to provide a tourism scene modeling interaction method and system based on virtual reality to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A tourism scene modeling interaction method based on virtual reality, the method comprising:

[0007] Establishing a connection channel with the monitoring system of the tourist scene, acquiring the scene video in real time, identifying the scene video, and determining the flow of people at various locations in the tourist scene;

[0008] Determine the tourism value of each location according to the flow of people at each location; the tourism value is used to represent the tourism income of a certain location;

[0009] Merge each location according to the tourism value, calculate the average tourism value simultaneously, and generate the tourism area with the average tourism value;

[0010] A VR scene of a tourism scene is established, and tourist areas with average tourism value are marked in the VR scene, and sent to a user terminal; wherein the user terminal is a VR device.

[0011] As a further solution of the present invention, the steps of establishing a connection channel with a monitoring system of a tourist scene, acquiring a scene video in real time, identifying the scene video, and determining the flow of people at various locations in the tourist scene include:

[0012] Establish a connection channel with the monitoring system of the tourism scene; the monitoring system is a camera network, and each camera records the time and collection area in real time when collecting data;

[0013] Receive images containing time and acquisition area uploaded by the monitoring system, group images of the same acquisition area together, arrange them in chronological order, and obtain scene videos of each acquisition area;

[0014] Identify scene videos and determine the flow of people at various locations within the tourism scene;

[0015] Among them, in the process of arranging according to time sequence to obtain scene videos of each acquisition area, when the time difference between two images is less than a preset difference threshold and the two images are acquired by two different cameras, the images are compared, the similarity is calculated, and the detection accuracy of the two cameras is determined according to the similarity; the similarity adopts histogram similarity, and the detection accuracy is proportional to the similarity; when the detection accuracy is less than the preset accuracy threshold, a maintenance instruction directed to the two cameras is generated.

[0016] As a further solution of the present invention: the step of identifying the scene video and determining the flow of people at each location in the tourist scene includes:

[0017] For each image in the scene video, locate the person in the image and determine the person's position and orientation;

[0018] Determine the instantaneous flow of people at each location in the acquisition area corresponding to the image according to the positions and orientations of all people;

[0019] Accumulate the instantaneous flow of people corresponding to all images in the scene video to obtain the flow of people at each location in the collection area;

[0020] Count the flow of people at each location in all collection areas to obtain the flow of people at each location in the tourism scene;

[0021] The calculation process of the instantaneous flow of people is as follows:

[0022] Where S(x,y) is the instantaneous flow of people at the location (x,y), T i is the characteristic value of the ith person at (x, y), A i The values ​​include A 1 and A 2 , A 1 >A 2 , the observation area of ​​the ith person is determined according to the person's orientation and the preset wide angle. When the position (x, y) is included in the observation area, A i Take A 1, when the position (x, y) is not included in the observation area, A i Take A 2 ;x i is the horizontal coordinate of the i-th person, y i is the vertical coordinate of the i-th person.

[0023] As a further solution of the present invention: the step of determining the tourism value of each location according to the flow of people at each location includes:

[0024] Accept the ratio range pre-set by the tourism management party;

[0025] Calculating a range of passenger flow according to the ratio range, and converting the passenger flow into tourism value based on the passenger flow range;

[0026] The conversion process is:

[0027] Where V is the tourism value, k is the preset correction coefficient, d(x) is the distance between the flow rate and the flow range; a and b are the left and right endpoints of the flow range, respectively; α is the preset power exponent; α∈(0,1), one of the values ​​of α is 0.1; min(|xa|,|xb|) means to find the minimum value of |xa| and |xb|.

[0028] As a further solution of the present invention: the steps of merging various locations according to tourism value, synchronously calculating the average tourism value, and generating a tourism area with average tourism value include:

[0029] Traversing all positions in the tourism scene, calculating the difference between the tourism value of any position and the tourism value of an adjacent position; the adjacent position adopts four neighboring positions;

[0030] When the difference is less than a preset difference threshold, the corresponding position is regarded as a similar position;

[0031] Calculate the mean of tourism value of similar locations to obtain the average tourism value;

[0032] Each type of location is counted as a tourist area, and the average tourism value is used as a parameter of the tourist area.

[0033] As a further solution of the present invention: the method further comprises:

[0034] Tourist areas with average tourism value greater than a preset value threshold are designated as landmark areas;

[0035] A preset number of points are randomly selected in the landmark area, and simulated photos of each point in the preset direction are obtained in the VR scene;

[0036] Input the simulated photos into the trained evaluation model and output the evaluation score for each direction of each point;

[0037] Select points and their directions whose evaluation scores are greater than the preset score threshold and provide feedback to the user.

[0038] The technical solution of the present invention also provides a tourism scene modeling interactive system based on virtual reality, the system comprising:

[0039] A crowd flow determination module is used to establish a connection channel with a monitoring system of a tourist scene, obtain scene videos in real time, identify the scene videos, and determine the crowd flow at various locations in the tourist scene;

[0040] A tourism value determination module, used to determine the tourism value of each location according to the flow of people at each location; the tourism value is used to represent the tourism income of a certain location;

[0041] The location merging module is used to merge various locations according to tourism value, simultaneously calculate the average tourism value, and generate a tourism area with average tourism value;

[0042] The marking display module is used to establish a VR scene of a tourism scene, mark the tourist areas with average tourism value in the VR scene, and send it to the user end; wherein the user end is a VR device.

[0043] As a further solution of the present invention: the human flow determination module includes:

[0044] A channel establishment unit is used to establish a connection channel with a monitoring system of a tourist scene; the monitoring system is a camera network, and each camera records the time and collection area in real time when collecting data;

[0045] An image arrangement unit is used to receive images containing time and acquisition area uploaded by the monitoring system, group the images of the same acquisition area into one group, arrange them according to the time sequence, and obtain scene videos of each acquisition area;

[0046] An identification execution unit is used to identify the scene video and determine the flow of people at various locations in the tourism scene;

[0047] Among them, in the process of arranging according to time sequence to obtain scene videos of each acquisition area, when the time difference between two images is less than a preset difference threshold and the two images are acquired by two different cameras, the images are compared, the similarity is calculated, and the detection accuracy of the two cameras is determined according to the similarity; the similarity adopts histogram similarity, and the detection accuracy is proportional to the similarity; when the detection accuracy is less than the preset accuracy threshold, a maintenance instruction directed to the two cameras is generated.

[0048] As a further solution of the present invention: the tourism value determination module includes:

[0049] A range receiving unit, used for receiving a ratio range preset by a tourism management party;

[0050] A value conversion unit, used for calculating a range of passenger flow according to the ratio range, and converting the passenger flow into tourism value based on the passenger flow range;

[0051] The conversion process is:

[0052] Where V is the tourism value, k is the preset correction coefficient, d(x) is the distance between the flow rate and the flow range; a and b are the left and right endpoints of the flow range, respectively; α is the preset power exponent; α∈(0,1), one of the values ​​of α is 0.1; min(|xa|,|xb|) means to find the minimum value of |xa| and |xb|.

[0053] As a further solution of the present invention: the position merging module includes:

[0054] A traversal unit is used to traverse all positions in the tourism scene and calculate the difference between the tourism value of any position and the tourism value of an adjacent position; the adjacent position adopts four neighboring positions;

[0055] A classification unit, used for treating the corresponding position as a position of the same type when the difference is less than a preset difference threshold;

[0056] A calculation unit, used to calculate the average of the tourism values ​​of similar locations to obtain the average tourism value;

[0057] The statistical unit is used to count each type of location as a tourist area, and the average tourist value is used as a parameter of the tourist area.

[0058] Compared with the prior art, the beneficial effects of the present invention are: the present invention obtains human flow information through the monitoring system in the tourism scene, determines the tourism value of each area in the tourism scene according to the human flow information, and displays the tourism value of each area based on the user's request during the virtual scene construction process, providing guidance for the virtual tourism process. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0060] Figure 1 Flowchart of the interactive method for modeling virtual reality-based tourism scenarios.

[0061] Figure 2 A flowchart of the first sub-process of the interactive method for modeling tourism scenarios based on virtual reality.

[0062] Figure 3 The second sub-process flowchart of the interactive method for modeling tourism scenarios based on virtual reality.

[0063] Figure 4 The third sub-process diagram of the interactive method for modeling tourism scenarios based on virtual reality.

[0064] Figure 5 A structural block diagram of the interactive system for modeling tourism scenes based on virtual reality. DETAILED DESCRIPTION

[0065] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] Figure 1 The flowchart of the interactive method for modeling a tourism scene based on virtual reality is shown in the figure. In an embodiment of the present invention, the interactive method for modeling a tourism scene based on virtual reality includes:

[0067] Step S100: establishing a connection channel with a monitoring system of a tourist scene, acquiring scene video in real time, identifying the scene video, and determining the flow of people at various locations in the tourist scene;

[0068] A tourist scene is generally a scenic spot, which is equipped with multiple image acquisition devices, collectively referred to as a monitoring system, which are used to obtain the real-time situation in the tourist scene, establish a connection channel with the monitoring system of the tourist scene, and obtain information in the tourist scene in real time, which is called a scene video. By identifying the scene video, the flow of people at various locations in the tourist scene can be obtained; the meaning of the flow of people is very simple, that is, how many people pass by each location; in order to simplify the operation process, the scene video obtained in this application is generally a scene video within a period of time, such as a scene video within a week.

[0069] Step S200: determining the tourism value of each location according to the flow of people at each location; the tourism value is used to represent the tourism income of a certain location;

[0070] After obtaining the flow of people at each location, it will be converted into tourism value based on the flow of people at each location. The actual meaning of the tourism value is how worthwhile a certain location is to travel to, which indicates the degree to which it is worthwhile to travel to a certain location. In other words, the tourism value is used to characterize the tourism income of a certain location. It should be noted that the tourism value corresponding to too little or too much flow of people in a place is very large. This is because if the flow of people is too little, it means that the location has not been explored yet and is more worthy of travel. If the flow of people is too much, it means that the location is popular with most people and is also worth traveling.

[0071] Step S300: merging various locations according to tourism value, synchronously calculating the average tourism value, and generating a tourism area with average tourism value;

[0072] After the conversion, each location will have a tourism value. The locations will be merged according to the tourism value, and the locations will be merged into small areas. The average tourism value of each small area will be calculated simultaneously to generate a tourist area with average tourism value. It is worth mentioning that since the tourists' travel process is continuous and will not jump in space or suddenly disappear from a place, the tourism value of adjacent locations does not change much, and the tourism value of each location in the entire tourism scene is smooth as a whole.

[0073] Step S400: establishing a VR scene of a tourism scene, marking a tourist area with average tourism value in the VR scene, and sending it to a user terminal; wherein the user terminal is a VR device;

[0074] Based on the existing VR scene construction technology, a VR scene of a tourist scene is established and sent to the user end, where the user end is a VR device, and the user can remotely visit the scenic spot through the VR device; in the technical solution of the present invention, tourist areas with average tourist value are marked in the VR scene, which intuitively shows whether each area is worth traveling, and provides a reference for the user's remote travel process.

[0075] Figure 2 This is a first sub-flow diagram of a tourism scene modeling interaction method based on virtual reality. The steps of establishing a connection channel with a monitoring system of a tourism scene, acquiring scene video in real time, identifying the scene video, and determining the flow of people at various locations in the tourism scene include:

[0076] Step S101: Establishing a connection channel with a monitoring system of a tourist scene; the monitoring system is a camera network, and each camera records the time and the collection area in real time when collecting data;

[0077] Step S102: receiving images containing time and acquisition area uploaded by the monitoring system, grouping images of the same acquisition area into one group, arranging them in chronological order, and obtaining scene videos of each acquisition area;

[0078] Step S103: Identify the scene video and determine the flow of people at each location in the tourism scene.

[0079] In an example of the technical solution of the present invention, a connection channel with a monitoring system of a tourist scene is established. The monitoring system is a collection of cameras. When collecting information, the cameras collect images containing time in real time. Since the cameras involved in the monitoring system in this application may be fixed cameras, or may be steerable cameras, and there are even some cameras installed on drones, when they collect images, in addition to recording time, they also need to record the area corresponding to the image, which is called the collection area; then, the images of the same collection area are grouped together and arranged in chronological order, and the collection of images is used as a scene video of the collection area; the scene video is identified to determine the flow of people at various locations in the tourist scene.

[0080] It is worth mentioning that for the same collection area, there may be multiple cameras collecting it (for example, two steering cameras collect one collection area at the same time). This means that in the process of arranging according to time sequence to obtain the scene video of each collection area, when the time difference between two images is less than the preset difference threshold and the two images are acquired by two different cameras, there may be differences between the two images. The images are compared, the similarity is calculated, and the detection accuracy of the two cameras is determined based on the similarity; the similarity adopts histogram similarity, and the detection accuracy is proportional to the similarity; when the detection accuracy is less than the preset accuracy threshold, a maintenance instruction is generated to the two cameras.

[0081] The above content provides a functional self-test solution. When the acquisition areas of any two cameras overlap, the acquisition results of the two cameras are compared. If the acquisition areas of the two cameras are similar, it means that the two cameras are working normally. If the acquisition areas of the two cameras are different, it means that there may be some problems in the acquisition process of the two cameras. At this time, maintenance instructions are generated to the two cameras.

[0082] Specifically, the step of identifying the scene video and determining the flow of people at each location in the tourism scene includes:

[0083] For each image in the scene video, locate the person in the image and determine the person's position and orientation;

[0084] Determine the instantaneous flow of people at each location in the acquisition area corresponding to the image according to the positions and orientations of all people;

[0085] Accumulate the instantaneous flow of people corresponding to all images in the scene video to obtain the flow of people at each location in the collection area;

[0086] The flow of people at each location in all collection areas is counted to obtain the flow of people at each location in the tourism scene.

[0087] From the above content, it can be concluded that the scene video is a collection of images of a certain collection area. For each image in the scene video, the image is positioned to determine the position and orientation of the person. The process of positioning the position of the person is common, but the process of determining and applying the orientation of the person is a unique solution in this application; the instantaneous flow of people at each location in the collection area corresponding to the image is determined based on the position and orientation of all people, and the instantaneous flow of people means the flow of people at the time corresponding to the image; the instantaneous flow of people corresponding to all images in the scene video is accumulated to obtain the flow of people at each location in the collection area; in the technical solution of the present invention, the length of the scene video is generally limited, such as within a week or a day.

[0088] The calculation process of the instantaneous flow of people is as follows:

[0089] Where S(x,y) is the instantaneous flow of people at the location (x,y), T i is the characteristic value of the ith person at (x, y), A i The values ​​include A 1 and A 2 , A 1 >A 2 , the observation area of ​​the ith person is determined according to the person's orientation and the preset wide angle. When the position (x, y) is included in the observation area, A i Take A 1 , when the position (x, y) is not included in the observation area, A i Take A 2 ;x i is the horizontal coordinate of the i-th person, y i is the vertical coordinate of the i-th person.

[0090] The calculation process of instantaneous pedestrian flow is to calculate the sum of the influence of all personnel at a certain position, and the influence is inversely proportional to the distance. It should be noted that in the technical solution of the present invention, a parameter of personnel orientation is also introduced. A conical area can be determined by the personnel orientation and the preset wide angle (150 degrees), which is called the observation area. The influence of personnel on the positions within the observation area is greater, so a larger molecule is used. On the contrary, the influence on the positions outside the observation area is smaller, so a smaller molecule is used.

[0091] Figure 3 The second sub-flow diagram of the interactive method for modeling tourism scenes based on virtual reality is shown in the figure. The step of determining the tourism value of each location according to the flow of people at each location includes:

[0092] Step S201: receiving a ratio range preset by a tourism management party;

[0093] Step S202: calculating a range of passenger flow according to the ratio range, and converting the passenger flow into tourism value based on the passenger flow range;

[0094] The proportion range is set by the tourism management party, and the proportion range is a value range, such as 25% to 50%. The range of passenger flow can be calculated from the proportion range, and then the passenger flow is converted into tourism value based on the calculated passenger flow range; this is actually a simple numerical conversion process.

[0095] The conversion process is:

[0096] Where V is the tourism value, k is the preset correction coefficient, d(x) is the distance between the flow rate and the flow range; a and b are the left and right endpoints of the flow range, respectively; α is the preset power exponent; α∈(0,1), one of the values ​​of α is 0.1; min(|xa|,|xb|) means to find the minimum value of |xa| and |xb|.

[0097] The principle of the conversion process is very simple, that is, the closer to the flow of people, the lower the tourism value. When the flow of people does not fall within the flow of people range, too small or too large a flow of people will correspond to a greater tourism value. When the flow of people falls within the flow of people range, the corresponding tourism value is actually similar, but this application also makes a distinction, that is, the closer to the center of the range, the smaller the corresponding value.

[0098] Figure 4 The third sub-flow diagram of the interactive method for modeling tourism scenes based on virtual reality is shown in the figure. The steps of merging various locations according to tourism value, calculating the average tourism value simultaneously, and generating a tourism area with the average tourism value include:

[0099] Step S301: traverse all positions in the tourism scene, and calculate the difference between the tourism value of any position and the tourism value of the adjacent position; the adjacent position adopts four neighborhood positions;

[0100] Step S302: when the difference is less than a preset difference threshold, the corresponding position is regarded as a similar position;

[0101] Step S303: Calculate the average of the tourism values ​​of similar locations to obtain the average tourism value;

[0102] Step S304: Count each type of location as a tourist area, and use the average tourist value as a parameter of the tourist area.

[0103] In an example of the technical solution of the present invention, all positions in the tourism scene are traversed, and the difference between the tourism value of any position and the tourism value of the adjacent position is calculated. The adjacent positions adopt four-neighborhood positions. The meaning of the four-neighborhood positions is very simple, that is, the positions in the four directions of due east, due south, due west and due north. The difference is compared with a preset difference threshold. When the difference is less than the preset difference threshold, the corresponding two positions are classified into one category; all positions in the tourism scene are traversed, and after all positions are processed, multiple categories of positions can be obtained; for each category of positions, the mean of the tourism values ​​of the same category of positions is calculated to obtain the average tourism value. Each category of positions itself constitutes an area, called a tourist area, and the average tourism value is used as a parameter of the tourist area.

[0104] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0105] Tourist areas with average tourism value greater than a preset value threshold are designated as landmark areas;

[0106] A preset number of points are randomly selected in the landmark area, and simulated photos of each point in the preset direction are obtained in the VR scene;

[0107] Input the simulated photos into the trained evaluation model and output the evaluation score for each direction of each point;

[0108] Select points and their directions whose evaluation scores are greater than the preset score threshold and provide feedback to the user.

[0109] In an example of the technical solution of the present invention, the average tourism value is compared with a preset value threshold, and the tourist area with an average tourism value greater than the preset value threshold is taken as a landmark area. A preset number of points are randomly selected in the landmark area. The preset number is generally large, so that the simulation process is more comprehensive; then, the management party pre-sets some directions. The simplest direction is to face the center of the landmark area and obtain the front view direction, the back view direction and two side view directions from the point. Of course, multiple directions can also be set. The more directions, the more comprehensive the detection. The most comprehensive one is to determine some ray clusters with the point as the center. The angles of adjacent rays in the ray clusters are 60 degrees or 30 degrees, and the directions of all rays are used as preset directions.

[0110] A preset number of points are randomly selected in the landmark area, and simulated photos of each point in a preset direction are obtained in the VR scene. The simulated photos are input into a trained evaluation model (an existing photo scoring model can be used, and the accuracy requirement is not high), and the evaluation score of each direction of each point is output. The points and their directions with evaluation scores greater than the preset score threshold are selected and fed back to the user; the obtained points and their directions are actually a kind of shooting guide, which assists users in activities similar to taking photos, and helps users to get better photos at which points and in which directions.

[0111] It should be noted that the evaluation model can be a neural network model. The staff selects photos in advance, scores them, and uses the photos and scores as samples to train the neural network model. After the neural network model is trained, the photos can be evaluated.

[0112] Figure 5 The structure diagram of the tourism scene modeling interaction system based on virtual reality is shown in FIG. 1 . In an embodiment of the present invention, a tourism scene modeling interaction system based on virtual reality is shown in FIG. 1 . The system 10 includes:

[0113] A crowd flow determination module 11 is used to establish a connection channel with a monitoring system of a tourist scene, obtain scene videos in real time, identify the scene videos, and determine the crowd flow at various locations in the tourist scene;

[0114] A tourism value determination module 12 is used to determine the tourism value of each location according to the flow of people at each location; the tourism value is used to represent the tourism income of a certain location;

[0115] The location merging module 13 is used to merge various locations according to tourism value, simultaneously calculate the average tourism value, and generate a tourism area with average tourism value;

[0116] The marking display module 14 is used to establish a VR scene of a tourist scene, mark tourist areas with average tourist value in the VR scene, and send it to the user end; wherein the user end is a VR device.

[0117] Furthermore, the pedestrian flow determination module 11 includes:

[0118] A channel establishment unit is used to establish a connection channel with a monitoring system of a tourist scene; the monitoring system is a camera network, and each camera records the time and collection area in real time when collecting data;

[0119] An image arrangement unit is used to receive images containing time and acquisition area uploaded by the monitoring system, group the images of the same acquisition area into one group, arrange them according to the time sequence, and obtain scene videos of each acquisition area;

[0120] An identification execution unit is used to identify the scene video and determine the flow of people at various locations in the tourism scene;

[0121] Among them, in the process of arranging according to time sequence to obtain scene videos of each acquisition area, when the time difference between two images is less than a preset difference threshold and the two images are acquired by two different cameras, the images are compared, the similarity is calculated, and the detection accuracy of the two cameras is determined according to the similarity; the similarity adopts histogram similarity, and the detection accuracy is proportional to the similarity; when the detection accuracy is less than the preset accuracy threshold, a maintenance instruction directed to the two cameras is generated.

[0122] Specifically, the tourism value determination module 12 includes:

[0123] A range receiving unit, used for receiving a ratio range preset by a tourism management party;

[0124] A value conversion unit, used for calculating a range of passenger flow according to the ratio range, and converting the passenger flow into tourism value based on the passenger flow range;

[0125] The conversion process is:

[0126] Where V is the tourism value, k is the preset correction coefficient, d(x) is the distance between the flow rate and the flow range; a and b are the left and right endpoints of the flow range, respectively; α is the preset power exponent; α∈(0,1), one of the values ​​of α is 0.1; min(|xa|,|xb|) means to find the minimum value of |xa| and |xb|.

[0127] Furthermore, the position merging module 13 includes:

[0128] A traversal unit is used to traverse all positions in the tourism scene and calculate the difference between the tourism value of any position and the tourism value of an adjacent position; the adjacent position adopts four neighboring positions;

[0129] A classification unit, used for treating the corresponding position as a position of the same type when the difference is less than a preset difference threshold;

[0130] A calculation unit, used to calculate the average of the tourism values ​​of similar locations to obtain the average tourism value;

[0131] The statistical unit is used to count each type of location as a tourist area, and the average tourist value is used as a parameter of the tourist area.

[0132] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A tourism scene modeling interaction method based on virtual reality, characterized in that: The method comprises: Establishing a connection channel with the monitoring system of the tourist scene, acquiring the scene video in real time, identifying the scene video, and determining the flow of people at various locations in the tourist scene; Determine the tourism value of each location according to the flow of people at each location; the tourism value is used to represent the tourism income of a certain location; Merge each location according to the tourism value, calculate the average tourism value simultaneously, and generate the tourism area with the average tourism value; A VR scene of a tourism scene is established, and tourist areas with average tourism value are marked in the VR scene, and sent to a user terminal; wherein the user terminal is a VR device.

2. The tourism scene modeling interaction method based on virtual reality according to claim 1 is characterized in that: The steps of establishing a connection channel with the monitoring system of the tourist scene, acquiring the scene video in real time, identifying the scene video, and determining the flow of people at various locations in the tourist scene include: Establish a connection channel with the monitoring system of the tourism scene; the monitoring system is a camera network, and each camera records the time and collection area in real time when collecting data; Receive images containing time and acquisition area uploaded by the monitoring system, group images of the same acquisition area together, arrange them in chronological order, and obtain scene videos of each acquisition area; Identify scene videos and determine the flow of people at various locations within the tourism scene; Among them, in the process of arranging according to time sequence to obtain scene videos of each acquisition area, when the time difference between two images is less than a preset difference threshold and the two images are acquired by two different cameras, the images are compared, the similarity is calculated, and the detection accuracy of the two cameras is determined according to the similarity; the similarity adopts histogram similarity, and the detection accuracy is proportional to the similarity; when the detection accuracy is less than the preset accuracy threshold, a maintenance instruction directed to the two cameras is generated.

3. The tourism scene modeling interaction method based on virtual reality according to claim 2 is characterized in that: The step of identifying the scene video and determining the flow of people at each location in the tourism scene includes: For each image in the scene video, locate the person in the image and determine the person's position and orientation; Determine the instantaneous flow of people at each location in the acquisition area corresponding to the image according to the positions and orientations of all people; Accumulate the instantaneous flow of people corresponding to all images in the scene video to obtain the flow of people at each location in the collection area; Count the flow of people at each location in all collection areas to obtain the flow of people at each location in the tourism scene; The calculation process of the instantaneous flow of people is as follows: Where S(x,y) is the instantaneous flow of people at the location (x,y), T i is the characteristic value of the ith person at (x, y), A i The values ​​of include A1 and A2, A1>A2, and the observation area of ​​the ith person is determined according to the person's orientation and the preset wide angle. When the position (x, y) is included in the observation area, A i Take A1, when the position (x, y) is not included in the observation area, A i Take A2; x i is the horizontal coordinate of the i-th person, y i is the vertical coordinate of the i-th person.

4. The tourism scene modeling interaction method based on virtual reality according to claim 1 is characterized in that: The step of determining the tourism value of each location according to the flow of people at each location includes: Accept the ratio range pre-set by the tourism management party; Calculating a range of passenger flow according to the ratio range, and converting the passenger flow into tourism value based on the passenger flow range; The conversion process is: V=e k·d(x) ; Where V is the tourism value, k is the preset correction coefficient, d(x) is the distance between the flow rate and the flow range; a and b are the left and right endpoints of the flow range, respectively; α is the preset power exponent; α∈(0,1), one of the values ​​of α is 0.1; min(|xa|,|xb|) means to find the minimum value of |xa| and |xb|.

5. The tourism scene modeling interaction method based on virtual reality according to claim 1 is characterized in that: The steps of merging various locations according to tourism value, calculating the average tourism value simultaneously, and generating a tourism area with average tourism value include: Traversing all positions in the tourism scene, calculating the difference between the tourism value of any position and the tourism value of an adjacent position; the adjacent position adopts four neighboring positions; When the difference is less than a preset difference threshold, the corresponding position is regarded as a similar position; Calculate the mean of tourism value of similar locations to obtain the average tourism value; Each type of location is counted as a tourist area, and the average tourism value is used as a parameter of the tourist area.

6. The tourism scene modeling interaction method based on virtual reality according to claim 1 is characterized in that: The method further comprises: Tourist areas with average tourism value greater than a preset value threshold are designated as landmark areas; A preset number of points are randomly selected in the landmark area, and simulated photos of each point in the preset direction are obtained in the VR scene; Input the simulated photos into the trained evaluation model and output the evaluation score for each direction of each point; Select points and their directions whose evaluation scores are greater than the preset score threshold and provide feedback to the user.

7. A tourism scene modeling interactive system based on virtual reality, characterized in that: The system comprises: A crowd flow determination module is used to establish a connection channel with a monitoring system of a tourist scene, obtain scene videos in real time, identify the scene videos, and determine the crowd flow at various locations in the tourist scene; A tourism value determination module, used to determine the tourism value of each location according to the flow of people at each location; the tourism value is used to represent the tourism income of a certain location; The location merging module is used to merge various locations according to tourism value, simultaneously calculate the average tourism value, and generate a tourism area with average tourism value; The marking display module is used to establish a VR scene of a tourism scene, mark the tourist areas with average tourism value in the VR scene, and send it to the user end; wherein the user end is a VR device.

8. The tourism scene modeling interactive system based on virtual reality according to claim 7 is characterized in that: The human flow determination module comprises: A channel establishment unit is used to establish a connection channel with a monitoring system of a tourist scene; the monitoring system is a camera network, and each camera records the time and collection area in real time when collecting data; An image arrangement unit is used to receive images containing time and acquisition area uploaded by the monitoring system, group the images of the same acquisition area into one group, arrange them according to the time sequence, and obtain scene videos of each acquisition area; An identification execution unit is used to identify the scene video and determine the flow of people at various locations in the tourism scene; Among them, in the process of arranging according to time sequence to obtain scene videos of each acquisition area, when the time difference between two images is less than a preset difference threshold and the two images are acquired by two different cameras, the images are compared, the similarity is calculated, and the detection accuracy of the two cameras is determined according to the similarity; the similarity adopts histogram similarity, and the detection accuracy is proportional to the similarity; when the detection accuracy is less than the preset accuracy threshold, a maintenance instruction directed to the two cameras is generated.

9. The tourism scene modeling interactive system based on virtual reality according to claim 7 is characterized in that: The tourism value determination module includes: A range receiving unit, used for receiving a ratio range preset by a tourism management party; A value conversion unit, used for calculating a range of passenger flow according to the ratio range, and converting the passenger flow into tourism value based on the passenger flow range; The conversion process is: V=e k·d(x) ; Where V is the tourism value, k is the preset correction coefficient, d(x) is the distance between the flow rate and the flow range; a and b are the left and right endpoints of the flow range, respectively; α is the preset power exponent; α∈(0,1), one of the values ​​of α is 0.1; min(|xa|,|xb|) means to find the minimum value of |xa| and |xb|.

10. The tourism scene modeling interactive system based on virtual reality according to claim 7 is characterized in that: The position merging module comprises: A traversal unit is used to traverse all positions in the tourism scene and calculate the difference between the tourism value of any position and the tourism value of an adjacent position; the adjacent position adopts four neighboring positions; A classification unit, used for treating the corresponding position as a position of the same type when the difference is less than a preset difference threshold; A calculation unit, used to calculate the average of the tourism values ​​of similar locations to obtain the average tourism value; The statistical unit is used to count each type of location as a tourist area, and the average tourist value is used as a parameter of the tourist area.

Images