Scholarlink

Population estimation is a critical component in urban planning, resource management, and socio-economic analysis. This paper introduces CNNPL, a convolutional neural network-based model that incorporates optimized Point of Interest weights and various geospatial data. The study utilizes a dataset of spatial patches to evaluate the model’s performance against both a pixel-level counterpart (CNNPxL) and established baseline models, including LandScan and GPWv4. Results indicate that CNNPL significantly improves predictive performance, achieving a coefficient of determination ( \(R^2\) ) of 0.8377 and a lower Mean Squared Error (MSE) of \(1.07 \times 10^{10}\) , thereby demonstrating superior efficacy in capturing population distribution patterns while maintaining computational efficiency.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Efficient Population Estimation Using Patch-Based Deep Learning

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Issa Nasralli,
  • Imen Masmoudi,
  • Hassen Drira,
  • Mohamed Ali Hadj Taieb

摘要

Population estimation is a critical component in urban planning, resource management, and socio-economic analysis. This paper introduces CNNPL, a convolutional neural network-based model that incorporates optimized Point of Interest weights and various geospatial data. The study utilizes a dataset of spatial patches to evaluate the model’s performance against both a pixel-level counterpart (CNNPxL) and established baseline models, including LandScan and GPWv4. Results indicate that CNNPL significantly improves predictive performance, achieving a coefficient of determination ( \(R^2\) ) of 0.8377 and a lower Mean Squared Error (MSE) of \(1.07 \times 10^{10}\) , thereby demonstrating superior efficacy in capturing population distribution patterns while maintaining computational efficiency.