CN-121091987-B - Scenic spot interaction system based on virtual reality

CN121091987BCN 121091987 BCN121091987 BCN 121091987BCN-121091987-B

Abstract

The invention discloses a scenic spot interaction system based on virtual reality, which relates to the technical field of electronic business card manufacturing, and comprises an interaction terminal and a server, wherein the interaction terminal is provided with an NFC read-write chip, the NFC read-write chip is used for establishing wireless communication with target terminal equipment and receiving an image synthesis instruction of the target terminal equipment based on the wireless communication, the server responds to the image synthesis instruction to acquire a first target image containing landscapes and multiple tourists and generate an acquisition instruction of a second target image, the NFC read-write chip responds to the acquisition instruction to acquire the second target image containing target characters in the target terminal equipment, the server synthesizes the target characters in the second target image into the first target image to obtain a third target image, and generates an electronic business card according to the third target image, and the NFC read-write chip sends the electronic business card to the target terminal equipment to enable the target terminal equipment to access the third target image through the electronic business card. The invention improves the group photo efficiency and convenience of tourists.

Inventors

ZHAO ZHANGANG
XU LIPING
QU YUAN
YAN LIXIA

Assignees

内蒙古自由极客数字科技有限公司

Dates

Publication Date: 20260512
Application Date: 20250815

Claims (8)

1. The scenic spot interaction system based on virtual reality is characterized by comprising an interaction terminal and a server; the interaction terminal is provided with an NFC read-write chip, and the NFC read-write chip is used for establishing wireless communication with target terminal equipment and receiving an image synthesis instruction of the target terminal equipment based on the wireless communication; The server is used for responding to the image synthesis instruction, acquiring a first target image containing landscapes and multiple tourists and generating an acquisition instruction of a second target image; the NFC read-write chip is used for responding to the acquisition instruction and acquiring a second target image containing a target person from the target terminal equipment based on the wireless communication; The server is used for synthesizing the target person in the second target image into the first target image to obtain a third target image, and generating an electronic business card according to the third target image; the NFC read-write chip is used for sending the electronic business card to the target terminal equipment so that the target terminal equipment accesses the third target image through the electronic business card; The synthesizing the target person in the second target image into the first target image to obtain a third target image includes: identifying gesture information and position information of a plurality of tourists in the first target image, wherein the gesture information comprises tourist orientation information of the tourists; identifying intention information of the tourist according to the gesture information; Determining tourists of which intention information is shooting intention as preselected tourists; identifying the orientation information of the target person in the second target image to obtain target orientation information; determining first orientation consistency data corresponding to the preselected tourist according to the similarity of the tourist orientation information of the preselected tourist and the target orientation information; Identifying landscape orientation information in the first target image to obtain landscape orientation information; determining second orientation consistency data corresponding to the preselected tourist according to the similarity of the tourist orientation information of the preselected tourist and the landscape orientation information; determining a heading consistency score from the first heading consistency data and the second heading consistency data; determining a target guest from the preselected guests according to the orientation consistency score; Determining space anchor point information according to the position information of the target tourist; And fusing the target person in the second target image into the position corresponding to the space anchor point information in the first target image to obtain a third target image.
2. The system of claim 1, wherein said determining a target guest from said preselected guests based on said orientation uniformity score comprises: According to the intention information and the position information, carrying out cluster analysis on tourists in the first target image to obtain a spatially aggregated tourist cluster; identifying a guest cluster to which the preselected guest belongs, and determining the number of guests in a unit area in the guest cluster; determining the real space density obtained after perspective transformation correction as the guest density corresponding to the preselected guest according to the number of guests in a unit area in the guest cluster; And determining target tourists from the preselected tourists according to the orientation consistency score and the tourist density corresponding to the preselected tourists.
3. The system of claim 2, wherein said determining a target guest from among the preselected guests based on the orientation uniformity score and a guest density corresponding to the preselected guest comprises: Determining tourists, of which gesture information is classified as standing gestures, as pre-enclosed tourists; Identifying the pre-enclosed tourists in a tourist cluster where the pre-selected tourists are located, and analyzing tourist orientation information of the pre-enclosed tourists; if the extension line of the direction vector corresponding to the guest direction information of the pre-enclosed guest points to any pre-selected guest in the guest cluster, determining the pre-enclosed guest as an enclosed guest; determining the ratio of the number of the enclosed tourists to the total number of the pre-enclosed tourists in the tourist cluster as the occupation ratio of the enclosed tourists; And determining target tourists from the preselected tourists according to the surrounding tourist duty ratio, the orientation consistency score and the tourist density corresponding to the preselected tourists.
4. The system of claim 3, wherein the fusing the target person in the second target image into the first target image at the location corresponding to the spatial anchor information to obtain a third target image comprises: cutting the original tourist image block at the position corresponding to the space anchor point information; carrying out affine transformation alignment on the foreground image of the target person, and filling the foreground image into the space anchor point; And carrying out edge fusion by using a Poisson image editing technology to obtain a third target image.
5. The system of claim 4, wherein prior to said edge blending using poisson image editing techniques to obtain a third target image, the server is further configured to perform: Extracting the ground normal vector at the space anchor point; Determining an included angle between a sole support plane of the target person and the ground normal vector; and if the included angle is larger than a preset included angle threshold, adjusting the leg joint point gesture of the target person so that the foot of the target person completely fits the ground surface at the space anchor point.
6. The system of claim 5, wherein the identifying pose information and location information for a plurality of guests in the first target image further comprises: If the upper half of the tourist is visible and the lower half of the tourist is blocked, determining the gesture information of the tourist through the included angle between the shoulder-hip connecting line and the gravity direction.
7. The system of claim 6, wherein the identifying orientation information of the landscape in the first target image to obtain landscape orientation information comprises: Acquiring the type of the landscape in the first target image; When the landscape type is a building type, extracting a main structural line through Canny edge detection and probability Hough transformation, and fitting vanishing points as dominant directions to obtain landscape orientation information; when the landscape type is natural, applying a frequency domain significance detection algorithm, and taking the centroid direction of the maximum connected region as the visual focus direction to obtain landscape orientation information; When the landscape type is a mixed type, building edges and natural significance characteristics are fused, and a composite orientation vector is generated to obtain landscape orientation information.
8. The system of claim 7, wherein the fusing the target person in the second target image into the first target image at the location corresponding to the spatial anchor information comprises: If the angle difference between the target orientation information and the guest orientation information of the target guest is detected to be larger than a preset angle difference threshold value, camera viewpoint parameters of the first target image and the second target image are obtained; constructing a homography transformation matrix according to the camera viewpoint parameters and the relative pose difference of the target tourist and the target person; performing perspective correction deformation on the target person in the second target image according to the homography transformation matrix to generate a correction image, wherein the correction image comprises a correction target person matched with the view angle of the target tourist; and fusing the corrected target person in the corrected image into the position corresponding to the space anchor point information in the first target image.

Description

Scenic spot interaction system based on virtual reality Technical Field The invention relates to the technical field of electronic business card manufacturing, in particular to a scenic spot interaction system based on virtual reality. Background With the vigorous development of the travel industry, tourist attractions become a popular choice for people to enjoy leisure and entertainment. Tourists often want to take a group photo containing themselves and a scenic spot logo as a reminder during the tour. However, in order to find the best scenic spot, adjust the composition, wait for a proper time or ask others to assist in shooting, the tourist often needs to spend a lot of time on the shooting link. The frequent framing and shooting process breaks up continuity of the tour, disperses immersive viewing experience of tourists on scenic spots and scenic spots, causes a plurality of tourists to 'carelessly view' real appearances of scenic spots in the journey, and excessively invests precious time in the shooting. The tourist group photo needs are usually solved by firstly photographing the tourists mutually, but the tourists depend on the technology of other people and possibly influence the sightseeing of other people, secondly using a self-timer rod or a tripod to perform self-photographing, but the operation is inconvenient in a crowded scenic spot of people and large scene group photos are difficult to photograph, and thirdly, the scenic spot provides professional photographic service for paying, but the cost is high, the tourists are required to wait in line, and the efficiency is low and the method is not convenient. Although the methods meet the group photo requirements of tourists to a certain extent, the problems of low efficiency, inconvenient operation or influence on the sightseeing of other people exist generally, and the contradiction that the tourists consume a great deal of time and influence the viewing experience due to photographing cannot be solved fundamentally. Therefore, how to improve the group photo efficiency and convenience of tourists becomes a technical problem to be solved. Disclosure of Invention The invention solves the technical problems that the group photo efficiency is low and the convenience is poor when tourists play scenic spots. In order to solve the technical problems, the invention provides the technical scheme that the scenic spot interaction system based on virtual reality comprises an interaction terminal and a server; the interaction terminal is provided with an NFC read-write chip, and the NFC read-write chip is used for establishing wireless communication with target terminal equipment and receiving an image synthesis instruction of the target terminal equipment based on the wireless communication; The server is used for responding to the image synthesis instruction, acquiring a first target image containing landscapes and multiple tourists and generating an acquisition instruction of a second target image; the NFC read-write chip is used for responding to the acquisition instruction and acquiring a second target image containing a target person from the target terminal equipment based on the wireless communication; The server is used for synthesizing the target person in the second target image into the first target image to obtain a third target image, and generating an electronic business card according to the third target image; The NFC read-write chip is used for sending the electronic business card to the target terminal equipment so that the target terminal equipment can access the third target image through the electronic business card. Preferably, the synthesizing the target person in the second target image with the first target image to obtain a third target image includes: identifying gesture information and position information of a plurality of tourists in the first target image, wherein the gesture information comprises tourist orientation information of the tourists; identifying intention information of the tourist according to the gesture information; Determining tourists of which intention information is shooting intention as preselected tourists; identifying the orientation information of the target person in the second target image to obtain target orientation information; determining first orientation consistency data corresponding to the preselected tourist according to the similarity of the tourist orientation information of the preselected tourist and the target orientation information; Identifying landscape orientation information in the first target image to obtain landscape orientation information; determining second orientation consistency data corresponding to the preselected tourist according to the similarity of the tourist orientation information of the preselected tourist and the landscape orientation information; determining a heading consistency score from the first heading consistency data and the second heading consistency d