摘要
仪表板是支持制造企业开展数据可视化分析和经营决策的重要手段。为解决可视化仪表板着色过程中对用户专业知识依赖强及色彩不协调等问题,提出了一种基于深度学习的可视化仪表板着色方法,该方法基于半自动仪表板着色法进行改进,并针对颜色提取过程中出现K-means算法无法准确提取图表颜色等问题,引入目标检测算法Grounding Dino和区域分割法Segment Anything对仪表板图表进行精准提取与分割后进行颜色提取;随后针对着色过程中出现背景色与图表颜色不协调等问题,引入全连接神经网络生成与背景色相协调的着色方案。为了将该方法应用于实际应用中,设计并开发了制造企业可视化仪表板着色系统,并将该方法引入作为仪表板着色工具。实验结果表明,该方法的应用能够提高制造企业仪表板着色的质量和效率。
Dashboard is an important means to support manufacturing enterprises to carry out data visual analysis and busi⁃ness decision-making.In order to solve the problems of strong dependence on user's professional knowledge and color incoordination in the process of visual dashboard coloring,a visual dashboard coloring method based on deep learning is proposed.This method is based on the semi-automatic dashboard coloring method,and the K-means algorithm can not accurately extract the color of the chart in the color extraction process.Grounding Dino,an object detection algorithm,and Segment Anything,a region segmentation method,are introduced to accurately extract and segment the dashboard chart for color extraction.Then,in order to solve the prob⁃lem that the background color is not consistent with the color of the chart in the coloring process,a fully connected neural network is introduced to generate a coloring scheme that is consistent with the background color.In order to apply the method to practical appli⁃cations,a visual dashboard coloring system for manufacturing enterprises is designed and developed,and the method is introduced as a dashboard coloring tool.Experimental results show that the application of this method can improve the quality and efficiency of dashboard coloring in manufacturing enterprises.
作者
王旭
陈亮
刘晋宇
WANG Xu;CHEN Liang;LIU Jinyu(School of Computer Science,Xi'an Polytechnic University,Xi'an 710600;Chongqing Vocational Institute of Tourism,Chongqing 409000)
出处
《舰船电子工程》
2025年第10期100-104,共5页
Ship Electronic Engineering
关键词
可视化
深度学习
颜色提取
仪表板着色
visualization
deep learning
color extraction
dashboard coloring
