This sequence follows an applied computer vision thesis from the first failure at a new flight altitude through dataset repair, model comparison, physical canopy measurement, and a practical deployment framework.

  1. 01

    The Drone That Couldn't See: Why Altitude Breaks Your AI Model

    A model trained at 3cm resolution failed at 6cm. Same drone. Same plantation. Different altitude. Here's the physics and math behind why — and what to do about it.

    4 min read
  2. 02

    Building the Dataset Nobody Had: Annotation, Corruption, and the Split That Kept Everything Honest

    How I built a multi-resolution oil palm segmentation dataset from scratch — including the 301 corrupted label files I had to fix before anything could run.

    5 min read
  3. 03

    Six Models Enter, One Problem Wins: A Head-to-Head Segmentation Showdown

    I trained YOLOv8, YOLOv11, Mask R-CNN, and two SAM hybrid pipelines on oil palm drone data. Here's what the numbers actually showed — including the result that surprised me most.

    4 min read
  4. 04

    The Hybrid Paradox: Why SAM + YOLO Sounds Better Than It Is

    SAM 2.1 was trained on 1 billion masks. My 56MB YOLO model still outperformed the hybrid pipeline in most conditions. Here's the detailed explanation.

    4 min read
  5. 05

    From Pixels to Meters: The Math Behind Canopy Size Estimation

    A segmentation mask is a collection of pixels. Here's how we convert that into a physical canopy area in square metres and an equivalent diameter — with the full derivation and error analysis.

    4 min read
  6. 06

    A Framework for Flying Smarter: The Practical Deployment Guide

    After 8 GSD levels, 6 models, and 3,368 tile-label pairs, here's the condensed operational guide for deploying AI-based oil palm monitoring with a drone.

    4 min read