Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems

In this paper, we presents a case study involving the implementation experience and a methodological framework through a comprehensive comparative analysis of the YOLOX and YOLOv12 object detection models for agricultural automation systems deployed in the Jetson AGX Orin edge computing platform. We...

Full description

Saved in:
Bibliographic Details
Main Authors: Joonam Kim, Giryeon Kim, Rena Yoshitoshi, Kenichi Tokuda
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/15/4586
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849406436182851584
author Joonam Kim
Giryeon Kim
Rena Yoshitoshi
Kenichi Tokuda
author_facet Joonam Kim
Giryeon Kim
Rena Yoshitoshi
Kenichi Tokuda
author_sort Joonam Kim
collection DOAJ
description In this paper, we presents a case study involving the implementation experience and a methodological framework through a comprehensive comparative analysis of the YOLOX and YOLOv12 object detection models for agricultural automation systems deployed in the Jetson AGX Orin edge computing platform. We examined the architectural differences between the models and their impact on detection capabilities in data-imbalanced potato-harvesting environments. Both models were trained on identical datasets with images capturing potatoes, soil clods, and stones, and their performances were evaluated through 30 independent trials under controlled conditions. Statistical analysis confirmed that YOLOX achieved a significantly higher throughput (107 vs. 45 FPS, <i>p</i> < 0.01) and superior energy efficiency (0.58 vs. 0.75 J/frame) than YOLOv12, meeting real-time processing requirements for agricultural automation. Although both models achieved an equivalent overall detection accuracy (F1-score, 0.97), YOLOv12 demonstrated specialized capabilities for challenging classes, achieving 42% higher recall for underrepresented soil clod objects (0.725 vs. 0.512, <i>p</i> < 0.01) and superior precision for small objects (0–3000 pixels). Architectural analysis identified a YOLOv12 residual efficient layer aggregation network backbone and area attention mechanism as key enablers of balanced precision–recall characteristics, which were particularly valuable for addressing agricultural data imbalance. However, NVIDIA Nsight profiling revealed implementation inefficiencies in the YOLOv12 multiprocess architecture, which prevented the theoretical advantages from being fully realized in edge computing environments. These findings provide empirically grounded guidelines for model selection in agricultural automation systems, highlighting the critical interplay between architectural design, implementation efficiency, and application-specific requirements.
format Article
id doaj-art-7c1a0ab065f64ec493fb2d7cdbc4fd05
institution Kabale University
issn 1424-8220
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-7c1a0ab065f64ec493fb2d7cdbc4fd052025-08-20T03:36:22ZengMDPI AGSensors1424-82202025-07-012515458610.3390/s25154586Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting SystemsJoonam Kim0Giryeon Kim1Rena Yoshitoshi2Kenichi Tokuda3Research Center for Agricultural Robotics, National Agricultural and Food Research Organization, Tsukuba 3050856, JapanResearch Center for Agricultural Robotics, National Agricultural and Food Research Organization, Tsukuba 3050856, JapanResearch Center for Agricultural Robotics, National Agricultural and Food Research Organization, Tsukuba 3050856, JapanResearch Center for Agricultural Robotics, National Agricultural and Food Research Organization, Tsukuba 3050856, JapanIn this paper, we presents a case study involving the implementation experience and a methodological framework through a comprehensive comparative analysis of the YOLOX and YOLOv12 object detection models for agricultural automation systems deployed in the Jetson AGX Orin edge computing platform. We examined the architectural differences between the models and their impact on detection capabilities in data-imbalanced potato-harvesting environments. Both models were trained on identical datasets with images capturing potatoes, soil clods, and stones, and their performances were evaluated through 30 independent trials under controlled conditions. Statistical analysis confirmed that YOLOX achieved a significantly higher throughput (107 vs. 45 FPS, <i>p</i> < 0.01) and superior energy efficiency (0.58 vs. 0.75 J/frame) than YOLOv12, meeting real-time processing requirements for agricultural automation. Although both models achieved an equivalent overall detection accuracy (F1-score, 0.97), YOLOv12 demonstrated specialized capabilities for challenging classes, achieving 42% higher recall for underrepresented soil clod objects (0.725 vs. 0.512, <i>p</i> < 0.01) and superior precision for small objects (0–3000 pixels). Architectural analysis identified a YOLOv12 residual efficient layer aggregation network backbone and area attention mechanism as key enablers of balanced precision–recall characteristics, which were particularly valuable for addressing agricultural data imbalance. However, NVIDIA Nsight profiling revealed implementation inefficiencies in the YOLOv12 multiprocess architecture, which prevented the theoretical advantages from being fully realized in edge computing environments. These findings provide empirically grounded guidelines for model selection in agricultural automation systems, highlighting the critical interplay between architectural design, implementation efficiency, and application-specific requirements.https://www.mdpi.com/1424-8220/25/15/4586object detectionagricultural automationedge computingYOLOXYOLOv12potato harvesting
spellingShingle Joonam Kim
Giryeon Kim
Rena Yoshitoshi
Kenichi Tokuda
Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems
Sensors
object detection
agricultural automation
edge computing
YOLOX
YOLOv12
potato harvesting
title Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems
title_full Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems
title_fullStr Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems
title_full_unstemmed Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems
title_short Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems
title_sort real time object detection for edge computing based agricultural automation a case study comparing the yolox and yolov12 architectures and their performance in potato harvesting systems
topic object detection
agricultural automation
edge computing
YOLOX
YOLOv12
potato harvesting
url https://www.mdpi.com/1424-8220/25/15/4586
work_keys_str_mv AT joonamkim realtimeobjectdetectionforedgecomputingbasedagriculturalautomationacasestudycomparingtheyoloxandyolov12architecturesandtheirperformanceinpotatoharvestingsystems
AT giryeonkim realtimeobjectdetectionforedgecomputingbasedagriculturalautomationacasestudycomparingtheyoloxandyolov12architecturesandtheirperformanceinpotatoharvestingsystems
AT renayoshitoshi realtimeobjectdetectionforedgecomputingbasedagriculturalautomationacasestudycomparingtheyoloxandyolov12architecturesandtheirperformanceinpotatoharvestingsystems
AT kenichitokuda realtimeobjectdetectionforedgecomputingbasedagriculturalautomationacasestudycomparingtheyoloxandyolov12architecturesandtheirperformanceinpotatoharvestingsystems