Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environments

This research addresses challenges in capsicum peduncle detection in night-time greenhouse environments, including low light, uneven illumination, and shadows, using advanced computer vision models. A dataset of 200 images was curated, capturing diverse distances, heights, occlusion levels, and ligh...

Full description

Saved in:
Bibliographic Details
Main Authors: Ayan Paul, Rajendra Machavaram
Format: Article
Language:English
Published: Taylor & Francis Group 2024-12-01
Series:Systems Science & Control Engineering
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/21642583.2024.2437162
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850115186040504320
author Ayan Paul
Rajendra Machavaram
author_facet Ayan Paul
Rajendra Machavaram
author_sort Ayan Paul
collection DOAJ
description This research addresses challenges in capsicum peduncle detection in night-time greenhouse environments, including low light, uneven illumination, and shadows, using advanced computer vision models. A dataset of 200 images was curated, capturing diverse distances, heights, occlusion levels, and lighting conditions, and was rigorously pre-processed and augmented. Two YOLOv9 instance segmentation variants, YOLOv9c-seg and YOLOv9e-seg, were custom-trained and fine-tuned using Google Colaboratory. YOLOv9c-seg (56.3 MB) achieved superior mean Average Precision (mAP) scores of 0.751 (box) and 0.725 (mask), outperforming YOLOv9e-seg (121.9 MB) with mAP scores of 0.674 (box) and 0.658 (mask). Grounded SAM, a zero-shot segmentation model, achieved maximum peduncle detection confidences of 59% and 49% with positional prompts. Comparative testing on 50 images containing 70 capsicums showed YOLOv9c-seg achieving mean precision, recall, and F1-scores of 0.93, 0.86, and 0.89, respectively, outperforming Grounded SAM (0.86, 0.70, and 0.77). This study highlights the efficacy of single-shot versus zero-shot segmentation models for automated capsicum peduncle detection in controlled agricultural environments, offering insights into model performance and future research directions for model optimization and dataset expansion.
format Article
id doaj-art-17e6e23069ba4c3eb352f4c5730c2da9
institution OA Journals
issn 2164-2583
language English
publishDate 2024-12-01
publisher Taylor & Francis Group
record_format Article
series Systems Science & Control Engineering
spelling doaj-art-17e6e23069ba4c3eb352f4c5730c2da92025-08-20T02:36:39ZengTaylor & Francis GroupSystems Science & Control Engineering2164-25832024-12-0112110.1080/21642583.2024.2437162Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environmentsAyan Paul0Rajendra Machavaram1Indian Institute of Technology Kharagpur, Kharagpur, West Bengal, IndiaIndian Institute of Technology Kharagpur, Kharagpur, West Bengal, IndiaThis research addresses challenges in capsicum peduncle detection in night-time greenhouse environments, including low light, uneven illumination, and shadows, using advanced computer vision models. A dataset of 200 images was curated, capturing diverse distances, heights, occlusion levels, and lighting conditions, and was rigorously pre-processed and augmented. Two YOLOv9 instance segmentation variants, YOLOv9c-seg and YOLOv9e-seg, were custom-trained and fine-tuned using Google Colaboratory. YOLOv9c-seg (56.3 MB) achieved superior mean Average Precision (mAP) scores of 0.751 (box) and 0.725 (mask), outperforming YOLOv9e-seg (121.9 MB) with mAP scores of 0.674 (box) and 0.658 (mask). Grounded SAM, a zero-shot segmentation model, achieved maximum peduncle detection confidences of 59% and 49% with positional prompts. Comparative testing on 50 images containing 70 capsicums showed YOLOv9c-seg achieving mean precision, recall, and F1-scores of 0.93, 0.86, and 0.89, respectively, outperforming Grounded SAM (0.86, 0.70, and 0.77). This study highlights the efficacy of single-shot versus zero-shot segmentation models for automated capsicum peduncle detection in controlled agricultural environments, offering insights into model performance and future research directions for model optimization and dataset expansion.https://www.tandfonline.com/doi/10.1080/21642583.2024.2437162Capsicum peduncle detectionnight-time greenhouseYOLOv9 instance segmentationzero-shot segmentationperformance evaluation
spellingShingle Ayan Paul
Rajendra Machavaram
Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environments
Systems Science & Control Engineering
Capsicum peduncle detection
night-time greenhouse
YOLOv9 instance segmentation
zero-shot segmentation
performance evaluation
title Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environments
title_full Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environments
title_fullStr Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environments
title_full_unstemmed Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environments
title_short Advanced segmentation models for automated capsicum peduncle detection in night-time greenhouse environments
title_sort advanced segmentation models for automated capsicum peduncle detection in night time greenhouse environments
topic Capsicum peduncle detection
night-time greenhouse
YOLOv9 instance segmentation
zero-shot segmentation
performance evaluation
url https://www.tandfonline.com/doi/10.1080/21642583.2024.2437162
work_keys_str_mv AT ayanpaul advancedsegmentationmodelsforautomatedcapsicumpeduncledetectioninnighttimegreenhouseenvironments
AT rajendramachavaram advancedsegmentationmodelsforautomatedcapsicumpeduncledetectioninnighttimegreenhouseenvironments