Using Convolutional Neural Networks to Count Palm Trees in Satellite Images

TL;DR

This paper presents a CNN-based supervised system for accurately counting and localizing palm trees in high-resolution satellite images, achieving over 99% accuracy with a small training dataset.

Contribution

The study introduces a novel CNN approach for palm tree counting in satellite imagery that requires only a small dataset for training.

Findings

Achieves over 99% accuracy in tree counting

Effective localization of palm trees in satellite images

Uses a small training dataset of 500 images

Abstract

In this paper we propose a supervised learning system for counting and localizing palm trees in high-resolution, panchromatic satellite imagery (40cm/pixel to 1.5m/pixel). A convolutional neural network classifier trained on a set of palm and no-palm images is applied across a satellite image scene in a sliding window fashion. The resultant confidence map is smoothed with a uniform filter. A non-maximal suppression is applied onto the smoothed confidence map to obtain peaks. Trained with a small dataset of 500 images of size 40x40 cropped from satellite images, the system manages to achieve a tree count accuracy of over 99%.