WO2020181517A1 - 基于多媒体数字化技术的传统古村落传播系统及方法 - Google Patents

基于多媒体数字化技术的传统古村落传播系统及方法 Download PDF

Info

Publication number
WO2020181517A1
WO2020181517A1 PCT/CN2019/077953 CN2019077953W WO2020181517A1 WO 2020181517 A1 WO2020181517 A1 WO 2020181517A1 CN 2019077953 W CN2019077953 W CN 2019077953W WO 2020181517 A1 WO2020181517 A1 WO 2020181517A1
Authority
WO
WIPO (PCT)
Prior art keywords
village
villages
user
traditional ancient
traditional
Prior art date
Application number
PCT/CN2019/077953
Other languages
English (en)
French (fr)
Inventor
李本建
桂宇晖
白凌婷
杨浩
Original Assignee
桂林理工大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 桂林理工大学 filed Critical 桂林理工大学
Priority to PCT/CN2019/077953 priority Critical patent/WO2020181517A1/zh
Publication of WO2020181517A1 publication Critical patent/WO2020181517A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • the invention relates to the field of ancient village communication, in particular to a traditional ancient village communication system and method based on multimedia digital technology.
  • the present invention provides a traditional ancient village propagation system and method based on multimedia digital technology. It can efficiently and intuitively disseminate traditional ancient villages to users and accurately recommend them.
  • the technical problem to be solved by the present invention is the technical problem of low propagation efficiency in the prior art.
  • the traditional ancient village communication system based on multimedia digital technology has the characteristics of high communication efficiency.
  • a traditional ancient village communication system based on multimedia digital technology including a VR system, a key feature extraction system, a recommendation system, and a network system for data interconnection;
  • the VR system is used to collect data on the implementation of traditional ancient villages, perform VR modeling, and aggregate the VR models of all traditional ancient villages into a VR model library;
  • the key feature extraction system is used to extract and match the inherent features of traditional ancient villages from VR models and user evaluations, and establish the mapping relationship between feature words and corresponding traditional ancient villages;
  • the recommendation system is used to customize or automatically acquire user characteristics, recommend ancient villages with user hobby characteristics, call VR models for users to preview, and at the same time disseminate user preview results to traditional ancient villages or other users.
  • the working principle of the present invention performs VR modeling of the traditional ancient village and establishes a model library, so that the user can experience the actual landscape of the traditional ancient village in advance.
  • the VR model can be updated to improve the efficiency of the traditional ancient village.
  • the dissemination of traditional ancient villages through VR can also reduce the deliberate destruction of traditional ancient villages because the ancient villages do not conform to the will.
  • the present invention performs feature extraction and grouping of traditional ancient villages, establishes a mapping relationship, and spreads according to user similarity and village similarity, thereby improving the accuracy and efficiency of the communication.
  • the key feature extraction system stores a village evaluation function construction program
  • the village evaluation function construction program includes the following steps:
  • Step 1 According to the VR model, define the number of characteristic words in a certain evaluation dimension of traditional ancient villages as N, and arrange the characteristic words in descending order of their appearance frequency. Define the frequency of appearance of the characteristic words with sequence number i in the arrangement as F i , F i -1 ⁇ F i , In the arrangement, select the first g feature words among the N feature words as the dimension feature dictionary to construct the village dimension description matrix;
  • Step 2 Carry out inclusive optimization on the village dimension description matrix of step 1 to obtain the village dimension description optimization matrix
  • i and j represent two feature words, P(i) represents the probability of feature vocabulary i appearing independently, P(j) represents the probability of feature vocabulary j appearing independently, P(i,j) represents feature vocabulary i and feature vocabulary the probability of j appearing at the same time;
  • Step 3 According to the user evaluation data, repeat steps 1 to 2 to construct an optimization matrix of user evaluation degree description;
  • Step 4. Match and filter the village dimension description optimization matrix of the same evaluation dimension of the same village with the user evaluation degree description optimization matrix to obtain the actual characteristics of the traditional ancient village.
  • the recommendation system stores a user recommendation program adapted to the VR model of the traditional ancient village to the user, and the user recommendation program includes the following steps:
  • Step A Collect the user's stay time in the traditional ancient village VR model, and the stay time will represent the user's preference for a traditional ancient village VR model.
  • Step B Establish interrelated village similarity functions, extract common factors of similar features among villages, and establish a transfer model
  • Step C using the user's preference value for the VR model of the traditional ancient village as the input number, and calculating the direct similarity of each village through the cosine similarity method;
  • Step D Calculate the indirect similarity of any village relative to other villages through the transfer model in step B;
  • Step E weighted fusion of the direct similarity and loudness of a certain village with the indirect similarity of any other village to calculate the final similarity of a certain village;
  • Step F calculate the final similarity of all villages, and group the villages according to the final similarity of the villages;
  • step G other villages in the unified group in the group of step F are recommended to the user.
  • calculating the direct similarity of each village by the cosine similarity method includes:
  • Step a define the direct similarity as:
  • rui is the preference value of user u for village i
  • I is the set of villages whose preference value reaches the predetermined threshold in the village model by users u and v at the same time. Is the preference vector in the village model of the user u towards the village;
  • Step b subtract the average value of preference values in all village models from step a, and correct the direct similarity.
  • the corrected direct similarity is:
  • the user can manually set the preference value between 0-1.
  • the present invention also provides a traditional ancient village communication method based on multimedia digital technology.
  • the traditional ancient village communication method based on multimedia digital technology is based on the aforementioned traditional ancient village communication system based on multimedia digital technology. include:
  • Step 1 Collect parameters of traditional ancient villages, and use VR modeling method to establish a VR model of traditional ancient villages;
  • Step 2 Collect the characteristic parameters of the VR model of traditional ancient villages, and group the traditional ancient villages into characteristic groups;
  • Step 3 Collect the user's staying time in the traditional ancient village VR model as the scoring coefficient, and define the user's score for the traditional ancient village VR model;
  • Step 4 For similar users, group VR models of traditional ancient villages with features in the same group to recommend similar users or value users to search for similar traditional ancient villages, and recommend VR models of traditional ancient villages with features in the same group to users , To complete the spread of traditional ancient villages.
  • the VR modeling method is completed by the VR-GIS method.
  • the beneficial effects of the present invention uses VR modeling and use methods, as well as traditional ancient villages for modeling, feature extraction, similar staging, user similarity grouping, etc., and the user’s browsing time for the traditional ancient villages is used as the preference value Characterizes the extroversion, and realizes efficient and accurate push of traditional ancient villages to users. Realize the high efficiency of communication.
  • Fig. 1 a flowchart of the procedure for constructing a village evaluation function in embodiment 1.
  • FIG. 2 is a flowchart of the user recommendation program in Embodiment 1.
  • FIG. 3 the flow chart of the traditional ancient village propagation method.
  • This embodiment provides a traditional ancient village communication system based on multimedia digital technology, which is characterized in that: the traditional ancient village communication system includes a VR system, a key feature extraction system, a recommendation system, and a network system for data interconnection;
  • the VR system is used to collect data on the implementation of traditional ancient villages, perform VR modeling, and aggregate the VR models of all traditional ancient villages into a VR model library;
  • the key feature extraction system is used to extract and match the inherent features of traditional ancient villages from VR models and user evaluations, and establish the mapping relationship between feature words and corresponding traditional ancient villages;
  • the recommendation system is used to customize or automatically acquire user characteristics, recommend ancient villages with user hobby characteristics, call VR models for users to preview, and at the same time disseminate user preview results to traditional ancient villages or other users.
  • the present invention performs VR modeling of the traditional ancient village and establishes a model library, so that the user can experience the actual landscape of the traditional ancient village in advance.
  • the VR model can be updated to improve the efficiency of the traditional ancient village.
  • the dissemination of traditional ancient villages through VR can also reduce the deliberate destruction of traditional ancient villages because the ancient villages do not conform to the will.
  • the present invention performs feature extraction and grouping of traditional ancient villages, establishes a mapping relationship, and spreads according to user similarity and village similarity, thereby improving the accuracy and efficiency of the communication.
  • the key feature extraction system stores a village evaluation function construction program, as shown in Figure 1.
  • the village evaluation function construction program includes the following steps:
  • Step 1 According to the VR model, define the number of characteristic words in a certain evaluation dimension of traditional ancient villages as N, and arrange the characteristic words in descending order of their appearance frequency. Define the frequency of appearance of the characteristic words with sequence number i in the arrangement as F i , F i -1 ⁇ F i , In the arrangement, select the first g feature words among the N feature words as the dimension feature dictionary to construct the village dimension description matrix;
  • Step 2 Carry out inclusive optimization on the village dimension description matrix of step 1 to obtain the village dimension description optimization matrix
  • i and j represent two feature words, P(i) represents the probability of feature vocabulary i appearing independently, P(j) represents the probability of feature vocabulary j appearing independently, P(i,j) represents feature vocabulary i and feature vocabulary the probability of j appearing at the same time;
  • Step 3 According to the user evaluation data, repeat steps 1 to 2 to construct an optimization matrix of user evaluation degree description;
  • Step 4. Match and filter the village dimension description optimization matrix of the same evaluation dimension of the same village with the user evaluation degree description optimization matrix to obtain the actual characteristics of the traditional ancient village.
  • the recommendation system stores a user recommendation program adapted to the traditional VR model of the ancient village, as shown in Figure 2.
  • the user recommendation program includes the following steps:
  • Step A Collect the user's stay time in the traditional ancient village VR model, and the stay time will represent the user's preference for a traditional ancient village VR model.
  • Step B Establish interrelated village similarity functions, extract common factors of similar features among villages, and establish a transfer model
  • Step C using the user's preference value for the VR model of the traditional ancient village as the input number, and calculating the direct similarity of each village through the cosine similarity method;
  • Step D Calculate the indirect similarity of any village relative to other villages through the transfer model in step B;
  • Step E weighted fusion of the direct similarity and loudness of a certain village with the indirect similarity of any other village to calculate the final similarity of a certain village;
  • Step F calculate the final similarity of all villages, and group the villages according to the final similarity of the villages;
  • step G other villages in the unified group in the group of step F are recommended to the user.
  • calculating the direct similarity of each village through the cosine similarity method includes:
  • Step a define the direct similarity as:
  • rui is the preference value of user u for village i
  • I is the set of villages whose preference value reaches the predetermined threshold in the village model by users u and v at the same time. Is the preference vector in the village model of the user u towards the village;
  • Step b subtract the average value of preference values in all village models from step a, and correct the direct similarity.
  • the corrected direct similarity is:
  • the maximum value of the user’s stay time in the traditional ancient village VR model can be collected, the maximum value is taken as 1, the remaining stay time is normalized, and the normalized calculation result is defined as the automatically collected preference value ;
  • the user can manually set the preference value between 0-1.
  • This embodiment only expresses one kind of mapping between stay duration and preference value.
  • a variety of mathematical methods can be used for mapping and association to realize the corresponding relationship between stay duration and preference value. Of course, the corresponding relationship should be the longer the stay duration, The greater the preference value.
  • the present invention also provides a traditional ancient village communication method based on multimedia digitization technology.
  • the traditional ancient village communication method based on multimedia digitization technology is based on the aforementioned traditional ancient village communication system based on multimedia digitization technology, as shown in Figure 3.
  • the propagation methods of ancient villages include:
  • Step 1 Collect parameters of traditional ancient villages, and use VR modeling method to establish a VR model of traditional ancient villages;
  • Step 2 Collect the characteristic parameters of the VR model of traditional ancient villages, and group the traditional ancient villages into characteristic groups;
  • Step 3 Collect the user's staying time in the traditional ancient village VR model as the scoring coefficient, and define the user's score for the traditional ancient village VR model;
  • Step 4 For similar users, group VR models of traditional ancient villages with features in the same group to recommend similar users or value users to search for similar traditional ancient villages, and recommend VR models of traditional ancient villages with features in the same group to users , To complete the spread of traditional ancient villages.
  • the VR modeling method can be completed by using a VR-GIS method.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本发明涉及一种基于多媒体数字化技术的传统古村落传播系统及方法，解决的是传播效率低的技术问题，通过采用包括VR系统，关键特征抽取系统、推荐系统以及用于数据互联的网络系统；VR系统用于对传统古村落实景进行数据采集，进行VR建模并将所有传统古村落的VR模型聚合为VR模型库；关键特征抽取系统用于自VR模型及用户评价对传统古村落的固有特征进行抽取匹配，建立特征词汇与对应传统古村落的映射关系；推荐系统用于自定义或自动获取的用户特征，推荐具备用户爱好特征的古村落，调取VR模型供用户进行预览，同时将用户预览的结果向传统古村落或其他用户传播的技术方案，较好的解决了该问题，可用于传统古村落传播。

Description

基于多媒体数字化技术的传统古村落传播系统及方法 技术领域
本发明涉及古村落传播领域，具体涉及一种基于多媒体数字化技术的传统古村落传播系统及方法。
背景技术
古村落是指历史悠久建村的村落，保留了较大的历史沿革，即建筑环境、建筑风貌、村落选址未有大的变动，具有独特民俗民风，虽经历久远年代，但至今仍为人们服务的村落。
现有的对于传统古村落的传播主要是借住互联网平台，旅游APP等。鉴于该方式中用户对于传统古村落的反馈效比低，不能有效的给新用户推荐其实际爱好或喜欢的传统古村落，从另一程度降低了传统古村落传播的扩展效率。此外，目前主要通过的图片等方式，不能给用户提前感受传统古村落的意境，拉低效率。
为解决上述技术问题，本发明提供一种基于多媒体数字化技术的传统古村落传播系统及方法。能够高效、直观的向用户传播传统古村落并精准的进行推荐。
发明内容
本发明所要解决的技术问题是现有技术中存在的传播效率低的技术问题。提供一种新的基于多媒体数字化技术的传统古村落传播系统，该基于多媒体数字化技术的传统古村落传播系统具有传播效率高的特点。
为解决上述技术问题，采用的技术方案如下：
一种基于多媒体数字化技术的传统古村落传播系统，所述传统古村落传播系统包括VR系统，关键特征抽取系统、推荐系统以及用于数据互联的网络系统；
VR系统用于对传统古村落实景进行数据采集，进行VR建模并将所有传统古村落的VR模型聚合为VR模型库；
关键特征抽取系统用于自VR模型及用户评价对传统古村落的固有特征进行抽取匹配，建立特征词汇与对应传统古村落的映射关系；
推荐系统用于自定义或自动获取的用户特征，推荐具备用户爱好特征的古村落，调取VR模型供用户进行预览，同时将用户预览的结果向传统古村落或其他用户传播。
本发明的工作原理：本发明将传统古村落进行VR建模，并建立模型库，从而使得用户可以提前感受传统古村落的实际景观。同时，根据传统古村落的参数更新，能够更新VR模型，提高传统古村落的效率。此外，通过VR进行传统古村落的传播也可以减少因古村落不符合心意而蓄意破坏传统古村落的情况。此外，本发明通过对传统古村落进行特征提取分组，建立映射关系，并根据用户相似以及村落相似进行扩散传播，提高了传播的准确性和效率。
上述方案中，为优化，进一步地，关键特征抽取系统存储有村落评价函数构建程序，村落评价函数构建程序包括如下步骤：
步骤1，根据VR模型，定义传统古村落某一个评价维度中特征词汇的数量为N，特征词汇按照出现的频率降序排列，定义排列中序号为i的特征词汇出现的频率为F i，F i-1≥F i，
Figure PCTCN2019077953-appb-000001
在排列中选择N个特征词汇中前g个特征词汇作为维度特征词典来构建村落维度描述矩阵；
步骤2，对步骤1的村落维度描述矩阵进行包容化优化得到村落维度描述优化矩阵，
Figure PCTCN2019077953-appb-000002
其中，i和j表示两个特征词汇，P(i)表示特征词汇i独立出现的概率，P(j)表示特征词汇j独立出现的概率，P(i，j)表示特征词汇i和特征词汇j同时出现的概率；
步骤3，根据用户评价数据，重复重复步骤1-步骤2，构建用户评价度描述优化矩阵；
步骤4，将同一村落的同一评价维度的村落维度描述优化矩阵与用户评价度描述优化矩阵进行匹配筛选，得到传统古村落的实际特征。
进一步地，推荐系统存储有向用户推荐适配传统古村落VR模型的用户推荐程序，用户推荐程序包括如下步骤：
步骤A，采集用户在传统古村落VR模型中的停留时长，将停留时长表征用户对于某一传统古村落VR模型喜好的参数值，停留时长越长，则喜好值越高；用户可手动更新喜好值，手动更新的喜好值优先级高级自动根据停留时长计算的喜好值；
步骤B，建立互相关联的村落相似性函数，将各村落间相似的部分特征提取公因式，并建立传递模型；
步骤C，将用户对传统古村落VR模型喜好值作为输入数，通过余弦相似度方法计算各村落的直接相似度；
步骤D，通过步骤B中的传递模型计算任一村落相对于其他村落的间接相似度；
步骤E，将某一村落的直接相似度与响度与其他任一村落的间接相似度进行加权融合，计算出某一村落的终相似度；
步骤F，计算出所有村落的终相似度，依照村落的终相似度对村落进行分群；
步骤G，将步骤F的分群在统一群内的其他村落推荐给用户。
进一步地，通过余弦相似度方法计算各村落的直接相似度包括：
步骤a，定义直接相似度为：
Figure PCTCN2019077953-appb-000003
其中，r ui为用户u对村落i的喜好值，I为用户u、v同时进行村落模型中喜好值达到预定阀值的村落集合，
Figure PCTCN2019077953-appb-000004
为用户u对村落的村落模型中偏好向量；
步骤b，在步骤a的基础上减去所有村落模型中喜好值的平均值，对直接相似度进行修正，修正后的直接相似度为：
Figure PCTCN2019077953-appb-000005
其中，
Figure PCTCN2019077953-appb-000006
为用户u对所有村落的喜好值的平均值。
进一步地，采集用户在传统古村落VR模型中的停留时长最大值，将最大值作为1，将其余停留时长进行归一化计算，并将归一化计算结果定义为自动采集的喜好值；
用户可在0-1之间进行手动设置喜好值。
本发明还提供一种基于多媒体数字化技术的传统古村落传播方法，所述基于多媒体数字化技术的传统古村落传播方法基于前述的基于多媒体数字化技术的传统古村落传播系统，所述传统古村落传播方法包括：
步骤一，采集传统古村落的参数，使用VR建模方法建立传统古村落VR模型；
步骤二，采集传统古村落VR模型的特征参数，对传统古村落进行特征分群；
步骤三，采集用户在传统古村落VR模型中停留时长作为评分系数，定义用 户对于传统古村落VR模型的评分；
步骤四，对相似用户，将特征分群在同一群内的传统古村落的VR模型推荐相似用户或值用户搜索相似传统古村落，将特征分群在同一群内的传统古村落的VR模型推荐给用户，完成传统古村落的传播。
具体地，所述VR建模方法采用VR-GIS方法完成。
本发明的有益效果：本发明通过VR建模及使用的方法，以及将传统古村落进行建模、特征提取、相似分期、用户相似分群等，并将用户对于传统古村落的浏览时间作为喜好值表征外向，实现了高效率准确推送传统古村落给用户。实现了传播的高效。
附图说明
下面结合附图和实施例对本发明进一步说明。
图1，实施例1中村落评价函数构建程序流程图。
图2，实施例1中的用户推荐程序流程图。
图3，传统古村落传播方法流程图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本发明，并不用于限定本发明。
实施例1
本实施例提供一种基于多媒体数字化技术的传统古村落传播系统，其特征在于：所述传统古村落传播系统包括VR系统，关键特征抽取系统、推荐系统以及用于数据互联的网络系统；
VR系统用于对传统古村落实景进行数据采集，进行VR建模并将所有传统古村落的VR模型聚合为VR模型库；
关键特征抽取系统用于自VR模型及用户评价对传统古村落的固有特征进行抽取匹配，建立特征词汇与对应传统古村落的映射关系；
推荐系统用于自定义或自动获取的用户特征，推荐具备用户爱好特征的古村落，调取VR模型供用户进行预览，同时将用户预览的结果向传统古村落或其他用户传播。
本实施例的工作原理：本发明将传统古村落进行VR建模，并建立模型库，从而使得用户可以提前感受传统古村落的实际景观。同时，根据传统古村落的参数更新，能够更新VR模型，提高传统古村落的效率。此外，通过VR进行传统古村落的传播也可以减少因古村落不符合心意而蓄意破坏传统古村落的情况。此外，本发明通过对传统古村落进行特征提取分组，建立映射关系，并根据用户相似以及村落相似进行扩散传播，提高了传播的准确性和效率。
具体地，关键特征抽取系统存储有村落评价函数构建程序，如图1，村落评价函数构建程序包括如下步骤：
步骤1，根据VR模型，定义传统古村落某一个评价维度中特征词汇的数量为N，特征词汇按照出现的频率降序排列，定义排列中序号为i的特征词汇出现的频率为F i，F i-1≥F i，
Figure PCTCN2019077953-appb-000007
在排列中选择N个特征词汇中前g个特征词汇作为维度特征词典来构建村落维度描述矩阵；
步骤2，对步骤1的村落维度描述矩阵进行包容化优化得到村落维度描述优化矩阵，
Figure PCTCN2019077953-appb-000008
其中，i和j表示两个特征词汇，P(i)表示特征词汇i独立出现的概率，P(j)表示特征词汇j独立出现的概率，P(i，j)表示特征词汇i和特征词汇j同时出现的概率；
步骤3，根据用户评价数据，重复重复步骤1-步骤2，构建用户评价度描述优化矩阵；
步骤4，将同一村落的同一评价维度的村落维度描述优化矩阵与用户评价度描述优化矩阵进行匹配筛选，得到传统古村落的实际特征。
具体地，推荐系统存储有向用户推荐适配传统古村落VR模型的用户推荐程序，如图2，用户推荐程序包括如下步骤：
步骤A，采集用户在传统古村落VR模型中的停留时长，将停留时长表征用户对于某一传统古村落VR模型喜好的参数值，停留时长越长，则喜好值越高；用户可手动更新喜好值，手动更新的喜好值优先级高级自动根据停留时长计算的喜好值；
步骤B，建立互相关联的村落相似性函数，将各村落间相似的部分特征提取公因式，并建立传递模型；
步骤C，将用户对传统古村落VR模型喜好值作为输入数，通过余弦相似度方法计算各村落的直接相似度；
步骤D，通过步骤B中的传递模型计算任一村落相对于其他村落的间接相似度；
步骤E，将某一村落的直接相似度与响度与其他任一村落的间接相似度进行加权融合，计算出某一村落的终相似度；
步骤F，计算出所有村落的终相似度，依照村落的终相似度对村落进行分群；
步骤G，将步骤F的分群在统一群内的其他村落推荐给用户。
其中，通过余弦相似度方法计算各村落的直接相似度包括：
步骤a，定义直接相似度为：
Figure PCTCN2019077953-appb-000009
其中，r ui为用户u对村落i的喜好值，I为用户u、v同时进行村落模型中喜好值达到预定阀值的村落集合，
Figure PCTCN2019077953-appb-000010
为用户u对村落的村落模型中偏好向量；
步骤b，在步骤a的基础上减去所有村落模型中喜好值的平均值，对直接相似度进行修正，修正后的直接相似度为：
Figure PCTCN2019077953-appb-000011
其中，
Figure PCTCN2019077953-appb-000012
为用户u对所有村落的喜好值的平均值。
具体地，可将采集用户在传统古村落VR模型中的停留时长最大值，将最大值作为1，将其余停留时长进行归一化计算，并将归一化计算结果定义为自动采集的喜好值；用户可在0-1之间进行手动设置喜好值。本实施例对于停留时长与喜好值的映射仅表达了一种，实际上可采用多种数学方法进行映射关联，实现停留时长与喜好值的对应关系，当然该对应关系应是停留时长越长，喜好值越大。
本发明还提供一种基于多媒体数字化技术的传统古村落传播方法，所述基于多媒体数字化技术的传统古村落传播方法基于前述的基于多媒体数字化技术的传统古村落传播系统，如图3，所述传统古村落传播方法包括：
步骤一，采集传统古村落的参数，使用VR建模方法建立传统古村落VR模型；
步骤二，采集传统古村落VR模型的特征参数，对传统古村落进行特征分群；
步骤三，采集用户在传统古村落VR模型中停留时长作为评分系数，定义用 户对于传统古村落VR模型的评分；
步骤四，对相似用户，将特征分群在同一群内的传统古村落的VR模型推荐相似用户或值用户搜索相似传统古村落，将特征分群在同一群内的传统古村落的VR模型推荐给用户，完成传统古村落的传播。
具体地，所述VR建模方法可采用VR-GIS方法完成。
本实施例未披露的部分均可采用现有技术的对应方案。
尽管上面对本发明说明性的具体实施方式进行了描述，以便于本技术领域的技术人员能够理解本发明，但是本发明不仅限于具体实施方式的范围，对本技术领域的普通技术人员而言，只要各种变化只要在所附的权利要求限定和确定的本发明精神和范围内，一切利用本发明构思的发明创造均在保护之列。

Claims (7)

  1. 一种基于多媒体数字化技术的传统古村落传播系统，其特征在于：所述传统古村落传播系统包括VR系统，关键特征抽取系统、推荐系统以及用于数据互联的网络系统；
    VR系统用于对传统古村落实景进行数据采集，进行VR建模并将所有传统古村落的VR模型聚合为VR模型库；
    关键特征抽取系统用于自VR模型及用户评价对传统古村落的固有特征进行抽取匹配，建立特征词汇与对应传统古村落的映射关系；
    推荐系统用于自定义或自动获取的用户特征，推荐具备用户爱好特征的古村落，调取VR模型供用户进行预览，同时将用户预览的结果向传统古村落或其他用户传播。
  2. 根据权利要求1所述的基于多媒体数字化技术的传统古村落传播系统，其特征在于：
    关键特征抽取系统存储有村落评价函数构建程序，村落评价函数构建程序包括如下步骤：
    步骤1，根据VR模型，定义传统古村落某一个评价维度中特征词汇的数量为N，特征词汇按照出现的频率降序排列，定义排列中序号为i的特征词汇出现的频率为F i，F i-1≥F i，
    Figure PCTCN2019077953-appb-100001
    在排列中选择N个特征词汇中前g个特征词汇作为维度特征词典来构建村落维度描述矩阵；
    步骤2，对步骤1的村落维度描述矩阵进行包容化优化得到村落维度描述优化矩阵：
    Figure PCTCN2019077953-appb-100002
    其中，i和j表示两个特征词汇，P(i)表示特征词汇i独立出现的概率，P(j)表 示特征词汇j独立出现的概率，P(i，j)表示特征词汇i和特征词汇j同时出现的概率；
    步骤3，根据用户评价数据，重复重复步骤1-步骤2，构建用户评价度描述优化矩阵；
    步骤4，将同一村落的同一评价维度的村落维度描述优化矩阵与用户评价度描述优化矩阵进行匹配筛选，得到传统古村落的实际特征。
  3. 根据权利要求2所述的基于多媒体数字化技术的传统古村落传播系统，其特征在于：
    推荐系统存储有向用户推荐适配传统古村落VR模型的用户推荐程序，用户推荐程序包括如下步骤：
    步骤A，采集用户在传统古村落VR模型中的停留时长，将停留时长表征用户对于某一传统古村落VR模型喜好的参数值，停留时长越长，则喜好值越高；用户可手动更新喜好值，手动更新的喜好值优先级高级自动根据停留时长计算的喜好值；
    步骤B，建立互相关联的村落相似性函数，将各村落间相似的部分特征提取公因式，并建立传递模型；
    步骤C，将用户对传统古村落VR模型喜好值作为输入数，通过余弦相似度方法计算各村落的直接相似度；
    步骤D，通过步骤B中的传递模型计算任一村落相对于其他村落的间接相似度；
    步骤E，将某一村落的直接相似度与响度与其他任一村落的间接相似度进行加权融合，计算出某一村落的终相似度；
    步骤F，计算出所有村落的终相似度，依照村落的终相似度对村落进行分群；
    步骤G，将步骤F的分群在统一群内的其他村落推荐给用户。
  4. 根据权利要求3所述的基于多媒体数字化技术的传统古村落传播系统，其特征在于：
    通过余弦相似度方法计算各村落的直接相似度包括：
    步骤a，定义直接相似度为：
    Figure PCTCN2019077953-appb-100003
    其中，r ui为用户u对村落i的喜好值，I为用户u、v同时进行村落模型中喜好值达到预定阀值的村落集合，
    Figure PCTCN2019077953-appb-100004
    为用户u对村落的村落模型中偏好向量；
    步骤b，在步骤a的基础上减去所有村落模型中喜好值的平均值，对直接相似度进行修正，修正后的直接相似度为：
    Figure PCTCN2019077953-appb-100005
    其中，
    Figure PCTCN2019077953-appb-100006
    为用户u对所有村落的喜好值的平均值。
  5. 根据权利要求4所述的基于多媒体数字化技术的传统古村落传播系统，其特征在于：
    采集用户在传统古村落VR模型中的停留时长最大值，将最大值作为1，将其余停留时长进行归一化计算，并将归一化计算结果定义为自动采集的喜好值；
    用户可在0-1之间进行手动设置喜好值。
  6. 一种基于多媒体数字化技术的传统古村落传播方法，其特征在于：所述基于多媒体数字化技术的传统古村落传播方法基于权利要求1-5任一所述的基于多媒体数字化技术的传统古村落传播系统，所述传统古村落传播方法包括：
    步骤一，采集传统古村落的参数，使用VR建模方法建立传统古村落VR模型；
    步骤二，采集传统古村落VR模型的特征参数，对传统古村落进行特征分群；
    步骤三，采集用户在传统古村落VR模型中停留时长作为评分系数，定义用户对于传统古村落VR模型的评分；
    步骤四，对相似用户，将特征分群在同一群内的传统古村落的VR模型推荐相似用户或值用户搜索相似传统古村落，将特征分群在同一群内的传统古村落的VR模型推荐给用户，完成传统古村落的传播。
  7. 根据权利要求6所述的基于多媒体数字化技术的传统古村落传播方法，其特征在于：所述VR建模方法采用VR-GIS方法完成。
PCT/CN2019/077953 2019-03-13 2019-03-13 基于多媒体数字化技术的传统古村落传播系统及方法 WO2020181517A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/077953 WO2020181517A1 (zh) 2019-03-13 2019-03-13 基于多媒体数字化技术的传统古村落传播系统及方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/077953 WO2020181517A1 (zh) 2019-03-13 2019-03-13 基于多媒体数字化技术的传统古村落传播系统及方法

Publications (1)

Publication Number Publication Date
WO2020181517A1 true WO2020181517A1 (zh) 2020-09-17

Family

ID=72427731

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/077953 WO2020181517A1 (zh) 2019-03-13 2019-03-13 基于多媒体数字化技术的传统古村落传播系统及方法

Country Status (1)

Country Link
WO (1) WO2020181517A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106650989A (zh) * 2016-09-26 2017-05-10 成都奥科睿科技有限公司 景区旅游服务平台及其工作方法
CN106777119A (zh) * 2016-12-16 2017-05-31 福建福光股份有限公司 在线式同步旅游方法
CN107169791A (zh) * 2017-05-03 2017-09-15 吴天重 一种旅行社用行前预览营销系统及方法
EP3340177A1 (en) * 2016-12-23 2018-06-27 Yu-Hsien Li Method and system for creating virtual message onto a moving object and searching the same

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106650989A (zh) * 2016-09-26 2017-05-10 成都奥科睿科技有限公司 景区旅游服务平台及其工作方法
CN106777119A (zh) * 2016-12-16 2017-05-31 福建福光股份有限公司 在线式同步旅游方法
EP3340177A1 (en) * 2016-12-23 2018-06-27 Yu-Hsien Li Method and system for creating virtual message onto a moving object and searching the same
CN107169791A (zh) * 2017-05-03 2017-09-15 吴天重 一种旅行社用行前预览营销系统及方法

Similar Documents

Publication Publication Date Title
CN104717124B (zh) 一种好友推荐方法、装置及服务器
CN104809117B (zh) 视频数据聚合处理方法、聚合系统及视频搜索平台
WO2017181612A1 (zh) 个性化视频推荐方法及装置
US20150293972A1 (en) Method and device used for providing input candidate items corresponding to an input character string
WO2020063092A1 (zh) 知识图谱的处理方法及装置
WO2019041521A1 (zh) 用户关键词提取装置、方法及计算机可读存储介质
WO2018000281A1 (zh) 一种基于深度神经网络的用户画像表示学习系统及方法
CN111885399B (zh) 内容分发方法、装置、电子设备以及存储介质
CN108399180B (zh) 一种知识图谱构建方法、装置及服务器
CN104050245B (zh) 一种基于活跃度的社交网络影响力最大化方法
WO2018166115A1 (zh) 客服问答数据的处理方法、电子装置及计算机可读存储介质
CN108920648A (zh) 一种基于音乐-图像语义关系的跨模态匹配方法
CN113986933A (zh) 物化视图的创建方法、装置、存储介质及电子设备
CN112084373A (zh) 一种基于图嵌入的多源异构网络用户对齐方法
CN105956158B (zh) 基于海量微博文本和用户信息的网络新词自动提取的方法
CN107133567B (zh) 一种创可贴广告点位选取方法及装置
CN108319888A (zh) 视频类型的识别方法及装置、计算机终端
CN110442977A (zh) 基于建筑施工工序网络推荐的移动端bim模型智能缓存方法
CN112000763A (zh) 兴趣点竞争关系确定方法、装置、设备和介质
WO2021135104A1 (zh) 基于多源数据的对象推送方法、装置、设备及存储介质
CN102184201B (zh) 一种用于选取查询序列的推荐序列的设备和方法
CN115510249A (zh) 一种知识图谱的构建方法及装置、电子设备、存储介质
CN115374189A (zh) 基于区块链的食品安全溯源方法、装置及设备
CN110247805B (zh) 一种基于k壳分解的识别传播关键节点的方法及装置
WO2020181517A1 (zh) 基于多媒体数字化技术的传统古村落传播系统及方法

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: 19918667

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19918667

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19918667

Country of ref document: EP

Kind code of ref document: A1