本文二维码信息
二维码(扫一下试试看!)
Application Research of Regional Feature Correlation Packaging Color Based on Multi-source Data and K-means Clustering
Received:November 21, 2022  
DOI:DOI:10.11980/j.issn.0254-508X.2023.03.013
Key Words:Zhongyang Street  regional characteristic  historic district  paper packaging
Fund Project:2021黑龙江社哲学社会科学研究规划项目(21YSC235)。
Author NameAffiliationPostcode
MIAO Mingrui* College of Landscape Architecture, Northeast Forest University, Harbin, Heilongjiang Province, 150060 150060
XIONG Jinghua School of Art, Heilongjiang University, Harbin, Heilongjiang Province, 150080 150080
FENG Lyuhang* School of Art, Heilongjiang University, Harbin, Heilongjiang Province, 150080 150080
XU Dawei College of Landscape Architecture, Northeast Forest University, Harbin, Heilongjiang Province, 150060 150060
Hits: 1897
Download times: 1952
Abstract:Taking the scenic spot of Harbin Zhongyang Street as the research object, OSM open source map platform, Baidu API platform, ArcGIS10.6 platform and Python platform were used to obtain data. DeepLabV3+ model under Matlab platform was used to perform semantic segmentation on the obtained street images. Using the HSB color space as the model benchmark, the K-means clustering algorithm drew the color scatter clustering model, color continuity simulation model and fragment color clustering model to visualize and analyze the Zhongyang Street scenic spot, and put forward the color application suggestions of regional characteristics associated paper packaging design. The multi-source data combined with K-means clustering method could complete the paper packaging color extraction and application of strategy formulation relatively fast, accurately and scientifically in a short period of time. The results reflected the regional cultural packaging local characteristics, enhanced the recognition of regional characteristics packaging which helped to achieve regional characteristic associated paper packaging color scheme optimization, and provided reference for paper packaging color application.
View Full Text  HTML  View/Add Comment  Download reader