Yu-En Yang

楊宇恩

博士(資源工程)

國立成功大學 衛星資訊中心

具備「Machine Learning × Computer Vision × 遙測影像分析 × WebGIS 系統落地」之整合能力, 以 Python 為主要開發語言,使用 TensorFlow / Keras 進行深度學習模型訓練 (CNN/YOLO/GAN),並結合 GPU(RTX 4090)進行大尺度資料運算。 可自資料流程建置、模型訓練與驗證, 至系統部署與平台化,完成政府與產業場域之端到端 AI 解決方案交付。

📍 臺南,臺灣 ✉️ [email protected] ORCID iD

Yu-En Yang Lab

核心技術能力

  • Machine Learning / AI: CNN、YOLO、GAN(cGAN)、Random Forest、XGBoost |Python、TensorFlow / Keras(PyTorch experience)
  • 程式與資料處理: Python、SQL、C#, 大尺度影像處理(40,000+ images)、特徵工程、GPU 加速運算
  • 遙測與 GIS: Google Earth Engine、QGIS、ArcGIS、UAV 攝影測量
  • 系統開發與部署: ASP.NET MVC、WebGIS、REST API、模型推論服務整合

代表性專案

  • 含石綿建材 AI 辨識與管理系統(WebGIS + YOLO)
    建立 YOLO 物件偵測模型,處理 4 萬筆以上 UAV、街景與手機影像, 並整合至 WebGIS 平台,支援政府大尺度盤查與決策應用。
  • cGAN 地形變遷(DoD)與崩塌量體推估
    以多時期遙測與地形資料建立生成對抗網路模型, 支援山崩體積變化分析與災害評估流程。
  • GBSAR / ArcSAR 邊坡變形監測
    分析毫米級地表變形資料,支援邊坡穩定監測與預警判斷。

現職經歷

  • 專案工程師/GIS 顧問(政府委託專案)
    國立成功大學 衛星資訊中心
Yu-En Yang

Yu-En Yang

Ph.D. in Resource Engineering (Jan 2026)

Satellite Information Center
National Cheng Kung University (NCKU)

Ph.D. in Resource Engineering with strong end-to-end expertise in Machine Learning, Computer Vision, remote sensing analytics, and WebGIS system deployment. Primary programming language is Python, with hands-on experience in TensorFlow / Keras (and PyTorch familiarity), GPU-accelerated model training (RTX 4090), and large-scale image datasets. Experienced in building data pipelines, training and validating deep learning models (CNN, YOLO, GAN), and delivering production-ready AI systems for real-world government and industry applications.

📍 Tainan, Taiwan ✉️ [email protected] ORCID iD

Yu-En Yang Lab

Core Technical Skills

  • Machine Learning / AI: CNN, YOLO, GAN, Random Forest, XGBoost
  • Programming & Data Processing: Python, SQL, C#, large-scale image processing, feature engineering
  • Remote Sensing & GIS: Google Earth Engine, QGIS, ArcGIS, UAV photogrammetry
  • Systems & Deployment: ASP.NET MVC, WebGIS, REST APIs

Selected Projects

  • AI-based Asbestos Mapping and Management System (WebGIS + YOLO)
    Developed YOLO-based object detection models using 40,000+ labeled images from UAV, street-view, and mobile sources. Integrated AI inference results into a WebGIS platform to support large-scale screening and inspection planning for local governments.
  • Landslide Volume Estimation Using Conditional GAN
    Applied conditional GAN models to estimate DEM of Difference (DoD) from multi-temporal remote sensing and terrain data, supporting landslide volume analysis and hazard assessment workflows.
  • GBSAR / ArcSAR Slope Deformation Monitoring
    Analyzed millimeter-level ground deformation data to support slope stability monitoring and early warning decision-making.

Professional Experience

  • Project Engineer & GIS Consultant (Government-funded Projects)
    Satellite Information Center, National Cheng Kung University