Explore the vast range of titles available.

[This corrects the article DOI: 10.3389/fpls.2025.1673202.].

Autonomous harvesting shows a promising prospect in the future development of the agriculture industry, while the vision system is one of the most challenging components in the autonomous harvesting technologies. This work proposes a multi-function network to perform the real-time detection and semantic segmentation of apples and branches in orchard environments by using the visual sensor. The developed detection and segmentation network utilises the atrous spatial pyramid pooling and the gate feature pyramid network to enhance feature extraction ability of the network. To improve the real-time computation performance of the network model, a lightweight backbone network based on the residual network architecture is developed. From the experimental results, the detection and segmentation network with ResNet-101 backbone outperformed on the detection and segmentation tasks, achieving an F 1 score of 0.832 on the detection of apples and 87.6% and 77.2% on the semantic segmentation of apples and branches, respectively. The network model with lightweight backbone showed the best computation efficiency in the results. It achieved an F 1 score of 0.827 on the detection of apples and 86.5% and 75.7% on the segmentation of apples and branches, respectively. The weights size and computation time of the network model with lightweight backbone were 12.8 M and 32 ms, respectively. The experimental results show that the detection and segmentation network can effectively perform the real-time detection and segmentation of apples and branches in orchards.

Accurate segmentation of key tobacco structures is essential for enabling automated harvesting. However, complex backgrounds, variable lighting conditions, and blurred boundaries between the stem and petiole significantly hinder segmentation accuracy in field environments. To overcome these challenges, we propose an enhanced instance segmentation approach based on YOLOv8-seg, incorporating depth-based background filtering and architectural improvements. Specifically, depth information from RGB-D images is employed to spatially filter non-target background regions, thereby enhancing foreground clarity. In addition, a Hybrid Dilated Residual Attention Block (HDRAB) is integrated into the YOLOv8-seg backbone to improve boundary discrimination between petioles and stems, while a Lightweight Shared Detail-Enhanced Convolution Detection Head (LSDECD) is designed to efficiently capture fine-grained texture features. Experimental results demonstrate that depth filtering increases mAP50 bb and mAP50 seg by 7.9% and 6.3%, respectively, while the architectural enhancements further raise them to 89.5% and 91.1%, surpassing the YOLOv8-seg baseline by 5.2% and 10.0%. Compared with mainstream models such as Mask R-CNN and SOLOv2, the proposed method achieves superior segmentation accuracy with low computational cost, highlighting its potential for practical deployment in automated tobacco harvesting

Accurate identification of picking points remains a critical challenge for automated marigold harvesting, primarily due to complex backgrounds and significant pose variations of the flowers. To overcome this challenge, this study proposes SCS-YOLO-Seg, a novel method based on a lightweight segmentation model. The approach enhances the baseline YOLOv8n-seg architecture by replacing its backbone with StarNet and introducing C2f-Star, a novel lightweight feature extraction module. These modifications achieve substantial model compression, significantly reducing the model size, parameter count, and computational complexity (GFLOPs). Segmentation efficiency is further optimized through a dual-path collaborative architecture (Seg-Marigold head). Following mask extraction, picking points are determined by intersecting the optimized elliptical mask fitting results with the stem skeleton. Experimental results demonstrate that SCS-YOLO-Seg effectively balances model compression with segmentation performance. Compared to YOLOv8n-seg, it maintains high accuracy while significantly reducing resource requirements, achieving a picking point identification accuracy of 93.36% with an average inference time of 28.66 ms per image. This work provides a robust and efficient solution for vision systems in automated marigold harvesting.

This review explores the transformative role of artificial intelligence (AI) in the entire winemaking process, from viticulture to bottling, with a particular focus on enhancing food safety and traceability. It discusses AI’s applications in optimizing grape cultivation, fermentation, bottling, and quality control, while emphasizing its critical role in managing microbiological risks such as mycotoxins. The review aims to show how AI technologies not only refine operational efficiencies but also raise safety standards and ensure traceability from vineyard to consumer. Challenges in AI implementation and future directions for integrating more advanced AI solutions into the winemaking industry will also be discussed, providing a comprehensive overview of AI’s potential to revolutionize traditional practices.

Accurate perception of tea buds is a fundamental prerequisite for intelligent and precise tea harvesting planning. However, in real tea plantation environments, reliable harvesting-oriented perception at the planning level remains highly challenging due to the small size of tea buds, severe occlusion, complex background clutter, and the lack of accurate three-dimensional spatial information. To address these challenges, we propose TeaNeRF, an integrated three-dimensional visual perception pipeline designed for harvesting-oriented tea bud analysis. Instead of treating detection, segmentation, and spatial analysis as independent tasks, TeaNeRF integrates sequential two-dimensional recognition, monocular depth estimation, and neural radiance field reconstruction into a coherent perception pipeline, allowing accurate spatial understanding of tea buds in complex natural scenes. It should be noted that the proposed integration is conducted at the perception-output level, where multiple modular components are connected through fixed interfaces, rather than through joint optimization or an end-to-end trainable formulation. The proposed framework combines an enhanced YOLO-based detector, prompt-guided segmentation, and monocular depth priors to guide NeRF-based three-dimensional reconstruction. By incorporating depth supervision and semantic-aware neural fields, TeaNeRF generates dense and geometrically consistent point clouds with reliable semantic separation. Quantitative evaluations show consistent improvements in reconstruction fidelity, as reflected by increased PSNR and reduced LPIPS across multiple tea tree scenes. Based on the reconstructed semantic point cloud, a three-dimensional clustering and geometric fitting strategy is further developed to enable tea bud counting and harvesting-oriented candidate point estimation at the perception level. Experiments conducted on a real-world dataset of 4,700 tea plantation images demonstrate that TeaNeRF improves detection accuracy (mAP@50 = 91.7%), segmentation quality (IoU = 0.640), and overall three-dimensional perception performance. Case-level counting results on representative tea trees indicate that the proposed 3D semantic point cloud–based approach can provide feasible tea bud counting behavior and consistent spatial guidance cues for downstream harvesting planning. By providing structured three-dimensional spatial information, including tea bud locations, counts, and harvesting-oriented candidate points, TeaNeRF offers practical perception-level outputs for downstream planning in automated tea harvesting systems.

Mushroom farms in the United States continue to face persistent labor shortages, especially during the harvesting of white button mushrooms (Agaricus bisporus) which requires selective picking by skilled workers. This study addresses this challenge by developing a depth-guided computer vision framework for automated mushroom detection, segmentation, and tracking to support timely harvest decisions, providing the foundation needed to support selective and timely robotic harvesting. The specific objectives of the study were to (1) develop a novel image-processing algorithm (RD-GuideNet) that integrates RGB and depth images for accurate detection and segmentation of mushrooms; (2) implement a custom depth-guided tracking algorithm to preserve mushroom identities across sequential frames; (3) compare the performance of RD-GuideNet against state-of-the-art deep learning models, YOLOv8 and YOLOv11, focusing on segmentation and tracking accuracies. The proposed RD-GuideNet achieved an F1-score of 0.93 for segmentation, outperforming YOLOv8 (0.88) and YOLOv11 (0.86), and produced sharper, more geometrically consistent boundaries that closely followed true mushroom cap contours. Its tracking consistency reached 92.7%, compared to YOLOv8 (95.3%) and YOLOv11 (94.6%). Although slightly lower, RD-GuideNet maintained high temporal consistency across dense mushroom beds. These results suggest that depth-based geometric reasoning and deep learning approaches exhibit complementary strengths in dense production scenes. Combining the two may further enhance detection reliability and shape fidelity, supporting high-precision perception for autonomous mushroom harvesting. A comprehensive quantitative evaluation of such a hybrid framework will be investigated in future work.

Artificial neural networks are one of the most important elements of machine learning and artificial intelligence. They are inspired by the human brain structure and function as if they are based on interconnected nodes in which simple processing operations take place. The spectrum of neural networks application is very wide, and it also includes agriculture. Artificial neural networks are increasingly used by food producers at every stage of agricultural production and in efficient farm management. Examples of their applications include: forecasting of production effects in agriculture on the basis of a wide range of independent variables, verification of diseases and pests, intelligent weed control, and classification of the quality of harvested crops. Artificial intelligence methods support decision-making systems in agriculture, help optimize storage and transport processes, and make it possible to predict the costs incurred depending on the chosen direction of management. The inclusion of machine learning methods in the “life cycle of a farm” requires handling large amounts of data collected during the entire growing season and having the appropriate software. Currently, the visible development of precision farming and digital agriculture is causing more and more farms to turn to tools based on artificial intelligence. The purpose of this Special Issue was to publish high-quality research and review papers that cover the application of various types of artificial neural networks in solving relevant tasks and problems of widely defined agriculture.

The global demand for button mushrooms has surged in recent years, driven by their health benefits, creating a significant challenge for the mushroom industry in meeting this increasing demand. The increasing reliance on human labor, which is becoming unsustainable due to labor shortages and rising wage costs, highlights the urgent need for automated harvesting solutions. This review examines the integration of automated systems in button mushroom harvesting, delving into the key components such as robots, mechanisms, machine elements, programming, and algorithms. It offers a thorough analysis of the performance, design, operational mechanisms, and advantages and limitations of robotic systems, comparing the different methods employed in automated harvesting. This paper compares the performance of all the mushroom harvesters, including the commercially available ones with manual harvesting, and identifies their potential and limitations. The commercial harvesters are shown to pick 2000 mushrooms per hour on average, which is similar to how much a skilled worker picks at the same time. However, commercial automation harvesting has a relatively low success rate, high initial cost, high operating cost, and energy consumption, identifying areas for future research and challenges. This paper serves as a valuable resource for researchers and industry professionals striving to advance automated harvesting technology and improve its efficiency in meeting the rising demand for button mushrooms.

Button mushrooms (Agaricus bisporus) grow in multilayered Dutch shelves with limited space between two shelves. As an alternative to conventional hand-picking, automated harvesting in recent times has gained widespread popularity. However, automated harvesting of mushrooms faces critical challenges in the form of growing environment, limited spaces, picking forces, and efficiency. End effectors for picking button mushrooms are an integral part of the automated harvesting process. The end effectors developed so far are oversized, bulky, and slow and thus are unsuitable for commercial mushroom harvesting applications. This paper introduces a novel three-finger hybrid gripper with rigid and soft parts, specifically designed for harvesting button mushrooms in automated systems even on narrow shelves. It discusses the design, fabrication, force analysis, and picking performance of the gripper in detail for both individual and clustered mushrooms. The results indicate that the gripping force depends on mushroom density and size. The inclusion of textured soft pads on gripper fingertips performs better compared with plain soft pads by reducing force by up to 20% and improving picking time. The gripper achieved a 100% picking success rate for single-grown mushrooms and 64% for clusters, with reduced picking times compared with existing end effectors. However, harvesting clustered mushrooms led to increased damage, suggesting the need for future improvements.