Optimization of Sorghum Spike Recognition Algorithm and Yield Estimation

In the natural field environment, the high planting density of sorghum and severe occlusion among spikes substantially increases the difficulty of sorghum spike recognition, resulting in frequent false positives and false negatives. The target detection model suitable for this environment requires h...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengyao Han, Jian Gao, Cuiqing Wu, Qingliang Cui, Xiangyang Yuan, Shujin Qiu
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Agronomy
Subjects:
Online Access:https://www.mdpi.com/2073-4395/15/7/1526
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849304084206583808
author Mengyao Han
Jian Gao
Cuiqing Wu
Qingliang Cui
Xiangyang Yuan
Shujin Qiu
author_facet Mengyao Han
Jian Gao
Cuiqing Wu
Qingliang Cui
Xiangyang Yuan
Shujin Qiu
author_sort Mengyao Han
collection DOAJ
description In the natural field environment, the high planting density of sorghum and severe occlusion among spikes substantially increases the difficulty of sorghum spike recognition, resulting in frequent false positives and false negatives. The target detection model suitable for this environment requires high computational power, and it is difficult to realize real-time detection of sorghum spikes on mobile devices. This study proposes a detection-tracking scheme based on improved YOLOv8s-GOLD-LSKA with optimized DeepSort, aiming to enhance yield estimation accuracy in complex agricultural field scenarios. By integrating the GOLD module’s dual-branch multi-scale feature fusion and the LSKA attention mechanism, a lightweight detection model is developed. The improved DeepSort algorithm enhances tracking robustness in occlusion scenarios by optimizing the confidence threshold filtering (0.46), frame-skipping count, and cascading matching strategy (n = 3, max_age = 40). Combined with the five-point sampling method, the average dry weight of sorghum spikes (0.12 kg) was used to enable rapid yield estimation. The results demonstrate that the improved model achieved a mAP of 85.86% (a 6.63% increase over the original YOLOv8), an F1 score of 81.19%, and a model size reduced to 7.48 MB, with a detection speed of 0.0168 s per frame. The optimized tracking system attained a MOTA of 67.96% and ran at 42 FPS. Image- and video-based yield estimation accuracies reached 89–96% and 75–93%, respectively, with single-frame latency as low as 0.047 s. By optimizing the full detection–tracking–yield pipeline, this solution overcomes challenges in small object missed detections, ID switches under occlusion, and real-time processing in complex scenarios. Its lightweight, high-efficiency design is well suited for deployment on UAVs and mobile terminals, providing robust technical support for intelligent sorghum monitoring and precision agriculture management, and thereby playing a crucial role in driving agricultural digital transformation.
format Article
id doaj-art-0d5b7afbcd954a1f9f27a60a7f6a1143
institution Kabale University
issn 2073-4395
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Agronomy
spelling doaj-art-0d5b7afbcd954a1f9f27a60a7f6a11432025-08-20T03:55:49ZengMDPI AGAgronomy2073-43952025-06-01157152610.3390/agronomy15071526Optimization of Sorghum Spike Recognition Algorithm and Yield EstimationMengyao Han0Jian Gao1Cuiqing Wu2Qingliang Cui3Xiangyang Yuan4Shujin Qiu5College of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030801, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030801, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030801, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030801, ChinaCollege of Agriculture, Shanxi Agricultural University, Jinzhong 030801, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030801, ChinaIn the natural field environment, the high planting density of sorghum and severe occlusion among spikes substantially increases the difficulty of sorghum spike recognition, resulting in frequent false positives and false negatives. The target detection model suitable for this environment requires high computational power, and it is difficult to realize real-time detection of sorghum spikes on mobile devices. This study proposes a detection-tracking scheme based on improved YOLOv8s-GOLD-LSKA with optimized DeepSort, aiming to enhance yield estimation accuracy in complex agricultural field scenarios. By integrating the GOLD module’s dual-branch multi-scale feature fusion and the LSKA attention mechanism, a lightweight detection model is developed. The improved DeepSort algorithm enhances tracking robustness in occlusion scenarios by optimizing the confidence threshold filtering (0.46), frame-skipping count, and cascading matching strategy (n = 3, max_age = 40). Combined with the five-point sampling method, the average dry weight of sorghum spikes (0.12 kg) was used to enable rapid yield estimation. The results demonstrate that the improved model achieved a mAP of 85.86% (a 6.63% increase over the original YOLOv8), an F1 score of 81.19%, and a model size reduced to 7.48 MB, with a detection speed of 0.0168 s per frame. The optimized tracking system attained a MOTA of 67.96% and ran at 42 FPS. Image- and video-based yield estimation accuracies reached 89–96% and 75–93%, respectively, with single-frame latency as low as 0.047 s. By optimizing the full detection–tracking–yield pipeline, this solution overcomes challenges in small object missed detections, ID switches under occlusion, and real-time processing in complex scenarios. Its lightweight, high-efficiency design is well suited for deployment on UAVs and mobile terminals, providing robust technical support for intelligent sorghum monitoring and precision agriculture management, and thereby playing a crucial role in driving agricultural digital transformation.https://www.mdpi.com/2073-4395/15/7/1526sorghum spikeYOLOv8sDeepSortalgorithm optimizationtarget detectionyield estimation
spellingShingle Mengyao Han
Jian Gao
Cuiqing Wu
Qingliang Cui
Xiangyang Yuan
Shujin Qiu
Optimization of Sorghum Spike Recognition Algorithm and Yield Estimation
Agronomy
sorghum spike
YOLOv8s
DeepSort
algorithm optimization
target detection
yield estimation
title Optimization of Sorghum Spike Recognition Algorithm and Yield Estimation
title_full Optimization of Sorghum Spike Recognition Algorithm and Yield Estimation
title_fullStr Optimization of Sorghum Spike Recognition Algorithm and Yield Estimation
title_full_unstemmed Optimization of Sorghum Spike Recognition Algorithm and Yield Estimation
title_short Optimization of Sorghum Spike Recognition Algorithm and Yield Estimation
title_sort optimization of sorghum spike recognition algorithm and yield estimation
topic sorghum spike
YOLOv8s
DeepSort
algorithm optimization
target detection
yield estimation
url https://www.mdpi.com/2073-4395/15/7/1526
work_keys_str_mv AT mengyaohan optimizationofsorghumspikerecognitionalgorithmandyieldestimation
AT jiangao optimizationofsorghumspikerecognitionalgorithmandyieldestimation
AT cuiqingwu optimizationofsorghumspikerecognitionalgorithmandyieldestimation
AT qingliangcui optimizationofsorghumspikerecognitionalgorithmandyieldestimation
AT xiangyangyuan optimizationofsorghumspikerecognitionalgorithmandyieldestimation
AT shujinqiu optimizationofsorghumspikerecognitionalgorithmandyieldestimation