CORTEXA
← Browse
crossrefAgriculture2025-04-10Cited by 24

Enhancing Autonomous Orchard Navigation: A Real-Time Convolutional Neural Network-Based Obstacle Classification System for Distinguishing ‘Real’ and ‘Fake’ Obstacles in Agricultural Robotics

Tabinda Naz Syed, Jun Zhou, Imran Ali Lakhiar, Francesco Marinello, Tamiru Tesfaye Gemechu, Luke Toroitich Rottok, Zhizhen Jiang

Autonomous navigation in agricultural environments requires precise obstacle classification to ensure collision-free movement. This study proposes a convolutional neural network (CNN)-based model designed to enhance obstacle classification for agricultural robots, particularly in orchards. Building upon a previously developed YOLOv8n-based real-time detection system, the model incorporates Ghost Modules and Squeeze-and-Excitation (SE) blocks to enhance feature extraction while maintaining computational efficiency. Obstacles are categorized as “Real”—those that physically impact navigation, such as tree trunks and persons—and “Fake”—those that do not, such as tall weeds and tree branches—allowing for precise navigation decisions. The model was trained on separate orchard and campus datasets and fine-tuned using Hyperband optimization and evaluated on an external test set to assess generalization to unseen obstacles. The model’s robustness was tested under varied lighting conditions, including low-light scenarios, to ensure real-world applicability. Computational efficiency was analyzed based on inference speed, memory consumption, and hardware requirements. Comparative analysis against state-of-the-art classification models (VGG16, ResNet50, MobileNetV3, DenseNet121, EfficientNetB0, and InceptionV3) confirmed the proposed model’s superior precision (p), recall (r), and F1-score, particularly in complex orchard scenarios. The model maintained strong generalization across diverse environmental conditions, including varying illumination and previously unseen obstacles. Furthermore, computational analysis revealed that the orchard-combined model achieved the highest inference speed at 2.31 FPS while maintaining a strong balance between accuracy and efficiency. When deployed in real-time, the model achieved 95.0% classification accuracy in orchards and 92.0% in campus environments. The real-time system demonstrated a false positive rate of 8.0% in the campus environment and 2.0% in the orchard, with a consistent false negative rate of 8.0% across both environments. These results validate the model’s effectiveness for real-time obstacle differentiation in agricultural settings. Its strong generalization, robustness to unseen obstacles, and computational efficiency make it well-suited for deployment in precision agriculture. Future work will focus on enhancing inference speed, improving performance under occlusion, and expanding dataset diversity to further strengthen real-world applicability.

View free PDFSource page

Related papers

openalexAgriculture2026-07-24

Recognition of Posture Transition Behavior in Sows Approaching Parturition Based on YOLOv11 and a Multi-Scale RGB–Flow Cross-Modal Temporal Network

Runhe Xue, Rui Ye, Yingjun Xiong, Yu Ding

Posture transition behavior in sows approaching parturition provides an important physiological cue for farrowing prediction. However, manual monitoring is time-consuming, labor-intensive and difficult to sustain under nighttime production conditions, while existing machine visio…

View free PDFSource page
openalexAgriculture2026-07-23

Integrating GIS-MCDA and Machine Learning Approach to Identify Marginal and Underutilized Lands in Leon County, Florida

Tewodros A. Simret, Sewunet A. Natae, Gang Chen, Victor Ibeanusi, Hubert Hirwa

This study developed an integrated Geographical Information System (GIS) based Multi-Criteria Decision Analysis (GIS-MCDA) framework based on hydrological marginality, soil suitability, land cover marginality, slope, social marginality, and contaminated land indicators to identif…

View free PDFSource page
crossrefAgriculture2026-07-23

UAV Multispectral Estimation of Citrus Leaf Nitrogen Content by Integrating Object-Based Canopy Extraction and PSO-Optimized Machine Learning

Hongmei Gu, Weiqi Zhang, Yuliang Fu, Yun Zhong, Songlin Wang

Leaf nitrogen content (LNC) is an important physiological indicator for evaluating citrus nutritional status, photosynthetic capacity, and fertilization demand. However, conventional LNC determination mainly relies on field sampling and laboratory chemical analysis, which are des…

View free PDFSource page
openalexAgriculture2026-07-23

Does the County-Level Business Environment Improve the Quality of Agriculture-Related Market Entry? Evidence from China

Xingyuan Yao, Jingyi Huang, Yufen Zhong, Qingfan Lin, Weiming Lin

Improving the quality of agriculture-related market entry is central to rural industrial upgrading, yet existing studies mainly measure entry quantity. The primary objective of this study is to assess whether the county-level business environment is associated with the structural…

View free PDFSource page
crossrefAgriculture2026-07-19

An Operation-Centered Review of Deep-Learning Computer Vision for Dairy Cow Management

Chi Zhang, Ziwen Yu, Jin Wang

Dairy cow management depends on repeated observations of behavior and physical condition to support health, welfare, and operational decisions, but these observations remain labor-intensive. Deep learning (DL)-based computer vision can automate parts of this work, although deploy…

View free PDFSource page
crossrefAgriculture2026-06-10

Unveiling the Effects of Digital Transformation on Agribusiness Green Innovation in China: An Explainable Machine Learning-Based Approach

Wanqi Liang, Xin Feng

Digital transformation is a key driver of green innovation in agribusiness. While the positive impact of digital transformation on firm innovation has been well documented, its multidimensional nature and heterogeneous associations on agribusiness green innovation remain underexp…

View free PDFSource page