Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree Crowns

Instance segmentation techniques based on convolutional neural networks (CNNs) is a vital tool for accurately identifying and segmenting individual tree crowns, which plays an essential role in environmental monitoring and forest management. In varied rural landscapes, canopy imagery often includes...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziyi Sun, Bing Xue, Mengjie Zhang, Jan Schindler
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11024197/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849472986103414784
author Ziyi Sun
Bing Xue
Mengjie Zhang
Jan Schindler
author_facet Ziyi Sun
Bing Xue
Mengjie Zhang
Jan Schindler
author_sort Ziyi Sun
collection DOAJ
description Instance segmentation techniques based on convolutional neural networks (CNNs) is a vital tool for accurately identifying and segmenting individual tree crowns, which plays an essential role in environmental monitoring and forest management. In varied rural landscapes, canopy imagery often includes a mix of tiny, small, and medium tree objects scattered across diverse terrains, from standalone trees to densely clustered forest stands. This variability poses significant challenges to traditional instance segmentation methods. To achieve this, we introduce a new method named YOLOv8-FF, which incorporates a feature fusion (FF) technique based on the YOLOv8 architecture. We first design a network architecture based on YOLOv8 that is optimized for the characteristics of our dataset, enabling effective segmentation of densely distributed tiny and small tree crowns. Moreover, YOLOv8-FF incorporates a FF mechanism that includes both cross-scale and same-scale fusion methods, enhancing the model’s ability to integrate information across different layers and scales, thereby improving segmentation performance. We incorporate Sparse Large Kernel Network, whose large convolution kernel can effectively extract key features, helping the model capture richer and deeper global information in the image. Experimental results on the tree crown dataset demonstrate that YOLOv8-FF outperforms several recent peer competitors, making it a promising tool for accurate and efficient tree crown instance segmentation.
format Article
id doaj-art-c35e585f574342629da9a47babb82f62
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-c35e585f574342629da9a47babb82f622025-08-20T03:24:21ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118147521476510.1109/JSTARS.2025.357678011024197Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree CrownsZiyi Sun0https://orcid.org/0000-0002-2011-4553Bing Xue1https://orcid.org/0000-0002-4865-8026Mengjie Zhang2https://orcid.org/0000-0003-4463-9538Jan Schindler3Center for Data Science and Artificial Intelligence & School of Engineering and Computer Science, Victoria University of Wellington, Wellington, New ZealandCenter for Data Science and Artificial Intelligence & School of Engineering and Computer Science, Victoria University of Wellington, Wellington, New ZealandCenter for Data Science and Artificial Intelligence & School of Engineering and Computer Science, Victoria University of Wellington, Wellington, New ZealandManaaki Whenua-Landcare Research, Wellington, New ZealandInstance segmentation techniques based on convolutional neural networks (CNNs) is a vital tool for accurately identifying and segmenting individual tree crowns, which plays an essential role in environmental monitoring and forest management. In varied rural landscapes, canopy imagery often includes a mix of tiny, small, and medium tree objects scattered across diverse terrains, from standalone trees to densely clustered forest stands. This variability poses significant challenges to traditional instance segmentation methods. To achieve this, we introduce a new method named YOLOv8-FF, which incorporates a feature fusion (FF) technique based on the YOLOv8 architecture. We first design a network architecture based on YOLOv8 that is optimized for the characteristics of our dataset, enabling effective segmentation of densely distributed tiny and small tree crowns. Moreover, YOLOv8-FF incorporates a FF mechanism that includes both cross-scale and same-scale fusion methods, enhancing the model’s ability to integrate information across different layers and scales, thereby improving segmentation performance. We incorporate Sparse Large Kernel Network, whose large convolution kernel can effectively extract key features, helping the model capture richer and deeper global information in the image. Experimental results on the tree crown dataset demonstrate that YOLOv8-FF outperforms several recent peer competitors, making it a promising tool for accurate and efficient tree crown instance segmentation.https://ieeexplore.ieee.org/document/11024197/Instance segmentationremote sensingtree crownstree speciesYOLOv8
spellingShingle Ziyi Sun
Bing Xue
Mengjie Zhang
Jan Schindler
Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree Crowns
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Instance segmentation
remote sensing
tree crowns
tree species
YOLOv8
title Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree Crowns
title_full Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree Crowns
title_fullStr Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree Crowns
title_full_unstemmed Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree Crowns
title_short Feature Fusion to Improve YOLOv8 for Segmenting and Classifying Aerial Images of Tree Crowns
title_sort feature fusion to improve yolov8 for segmenting and classifying aerial images of tree crowns
topic Instance segmentation
remote sensing
tree crowns
tree species
YOLOv8
url https://ieeexplore.ieee.org/document/11024197/
work_keys_str_mv AT ziyisun featurefusiontoimproveyolov8forsegmentingandclassifyingaerialimagesoftreecrowns
AT bingxue featurefusiontoimproveyolov8forsegmentingandclassifyingaerialimagesoftreecrowns
AT mengjiezhang featurefusiontoimproveyolov8forsegmentingandclassifyingaerialimagesoftreecrowns
AT janschindler featurefusiontoimproveyolov8forsegmentingandclassifyingaerialimagesoftreecrowns