Analysis and Optimization of Green Rural Cultural Tourism Image A Case Study Utilizing Big Data from Tourists' Photographs

The objective of this study is to explore the fundamental attributes of rural cultural tourism with a focus on green landscapes and to propose strategies for optimizing the rural tourism image. This study centers on rural tourism destinations across China, leveraging publicly available datasets comprising a large volume of tourist photographs sourced from online platforms. Using an enhanced Convolutional Neural Network (CNN) model, the study conducts image recognition to identify elements such as natural scenery and traditional cultural motifs depicted in the photographs, followed by their classification and organization. The improvement process integrates GoogleNet and implements a range of techniques, including the utilization of ReLU 6 activation parameters, batch normalization algorithms, and the principles of transfer learning. Subsequently, several key variables are established, encompassing site characteristics, recommendations for image optimization, tourist satisfaction levels, and levels of tourist participation, among others. Finally, analytical methods such as correlation analysis and multivariate regression analysis are employed to investigate the relationships between these variables and the enhancement of the rural cultural tourism image. The results reveal a correlation coefficient of 0.75 between site characteristics and image optimization recommendations, and 0.68 between site characteristics and tourist satisfaction. Moreover, image optimization recommendations and tourist satisfaction demonstrate a positive correlation with the enhancement of the tourism image, with regression coefficients of 0.42 and 0.38, respectively. These findings underscore a significant correlation between the distinctive attributes of tourist sites and the recommendations for image optimization, as well as the close association between tourist satisfaction and site characteristics. It is suggested that professional recommendations for image optimization and heightened levels of tourist satisfaction can effectively contribute to the enhancement of the rural cultural tourism image.

本研究旨在探索以绿色景观为核心的乡村文化旅游核心属性，并提出优化乡村旅游形象的策略。本研究以中国境内乡村旅游目的地为研究对象，依托公开数据集，该数据集包含大量从在线平台采集的旅游摄影作品。本研究采用改进型卷积神经网络（Convolutional Neural Network, CNN）模型开展图像识别工作，以识别照片中呈现的自然景观、传统文化符号等元素，并对其进行分类与整理。该改进过程融合了GoogleNet架构，并整合了多项技术手段，包括使用ReLU 6激活参数、批量归一化算法以及迁移学习原理。随后，本研究确立了多项关键变量，涵盖景区特征、形象优化建议、游客满意度以及游客参与度等。最后，本研究采用相关分析、多元回归分析等分析方法，探究上述变量与乡村文化旅游形象提升之间的关联关系。研究结果显示，景区特征与形象优化建议之间的相关系数为0.75，景区特征与游客满意度之间的相关系数为0.68。此外，形象优化建议与游客满意度均与旅游形象提升呈正相关，对应的回归系数分别为0.42与0.38。上述研究结果表明，旅游景区的独特属性与形象优化建议之间存在显著关联，同时游客满意度与景区特征也紧密相关。研究建议，专业的形象优化建议与更高水平的游客满意度可有效助力乡村文化旅游形象的提升。