# Automatic segmentation of trees in dynamic outdoor environments

## TL;DR

This paper presents a superpixel-based segmentation method tailored for dynamic outdoor environments like orchards, effectively distinguishing trees from backgrounds despite variable lighting conditions, aiding in applications like tree reconstruction and flower detection.

## Contribution

The paper introduces a novel segmentation approach combining superpixels and color analysis to improve accuracy in challenging outdoor scenes.

## Key findings

- Effective segmentation in variable outdoor lighting conditions
- Improved tree and flower detection accuracy
- Suitable for orchard automation applications

## Abstract

Segmentation in dynamic outdoor environments can be difficult when the illumination levels and other aspects of the scene cannot be controlled. Specifically in orchard and vineyard automation contexts, a background material is often used to shield a camera's field of view from other rows of crops. In this paper, we describe a method that uses superpixels to determine low texture regions of the image that correspond to the background material, and then show how this information can be integrated with the color distribution of the image to compute optimal segmentation parameters to segment objects of interest. Quantitative and qualitative experiments demonstrate the suitability of this approach for dynamic outdoor environments, specifically for tree reconstruction and apple flower detection applications.