一种改进的基于像素的热力图渲染算法

An improved heat map rendering algorithm based on pixel

针对格网聚类后不同像素点的数据密度可能存在较大差异,导致生成的热力图不能准确表达数据源区域分布特征的问题,文章在研究基于像素的热力图渲染算法基础上,在中心影响力计算步骤引入非线性变换的图像增强算法予以改进,同时使用改进后的热力图渲染算法、改进前基于像素的热力图渲染算法、主流地图引擎Mapbox进行了热力图渲染比较实验。实验结果表明:改进后算法与改进前算法相比,能够很好地处理由于中心灰度值分化导致热力图渲染失真的问题;改进后算法与使用Mapbox进行热力图渲染的效果基本一致,甚至在边缘区域的分布特征细节上表现更优;针对不同数量级的数据源进行热力图渲染时,改进后算法的渲染效率相比于Mapbox都有不同程度的提升。

In view of the fact that the data density of different pixels may differ greatly after grid clustering,resulting that the generated heat map cannot accurately express the regional distribution characteristics of the data source,this paper introduces the nonlinear transformation image enhancement algorithm in the central influence calculation step on the basis of studying the pixel-based heat map rendering algorithm,and the improved algorithm is compared with the original algorithm and the map engine Mapbox.The experimental results show that compared with the previous algorithm,the improved algorithm can deal with the distortion of heat map rendering due to the differentiation of central gray value;the rendering effect of the improved algorithm is basically consistent with that of using Mapbox,and it is even better in the distribution feature details of edge area;as for different data sources of different levels in the rendering of heat map,the rendering efficiency of the improved algorithm is improved to a different extent compared with Mapbox.