Weikuan Jia

Low-light conditions in orchards and facility agriculture hinder accurate fruit recognition. To enhance efficiency in such environments, an optimized semantic contrastive learning-based nighttime fruit image enhancement model, OSCL (Optimized Semantic Contrastive Learning), is proposed. This model integrates semantic contrastive learning for processing low-light fruit images and consists of three modules: image enhancement, semantic segmentation, and contrastive learning. First, the Zero-DCE network is used as the image enhancement network, requiring only input nighttime images for training, resulting in enhanced images through iterative processing. The semantic segmentation module employs a dual attention network (DANet), with a backbone and two parallel attention branches to capture contextual information and meet recognition needs for green fruits. Finally, the VGG-16, known for its regular structure and depth, is selected for feature extraction in the contrastive learning module, effectively analyzing low-light fruit features and preserving information on small fruits. To validate OSCL’s enhancement effect, low-light images of green fruits from apple, persimmon, and pear orchards were collected under active and passive lighting conditions. Both subjective visual assessments and objective numerical evaluations confirmed the model’s efficacy. The generalized peak signal-to-noise ratios (GPSNR) for low-light images processed by OSCL reached 16.23 dB, 15.77 dB, and 15.30 dB under active illumination, and 16.89 dB, 14.38 dB, and 14.17 dB under passive illumination. Overall, OSCL outperforms classical enhancement models, enhancing operational efficiency in low-light conditions. This method offers a theoretical basis for future research on green fruit detection and yield prediction in nighttime environments.

This article proposes GhostConv-Dynamic Snake Convolution Network (GHD Net), a lightweight segmentation model designed to accurately recognize overlapping green tomatoes in complex facility environments. It is optimized using the YOLOv8-n algorithm. GHD Net efficiently recognizes and segments green fruit images affected by overlapping occlusions, varying illumination conditions, and capture angles. Specifically, the model enhances the feature extraction network by introducing a lightweight convolutional technique, GhostConv, to extract features more efficiently while maintaining high performance. In the feature fusion stage, the dynamic snake convolution module adaptively adjusts its position based on the shape and boundary information of the target, accurately capturing target edges and enhancing the representation of overlapping and occluded edge features. The classification and regression subnetworks predict fruit categories and bounding boxes, respectively, while the segmentation subnetwork generates mask prototypes and mask coefficients through Proto branches, ensuring the accuracy and consistency of the segmentation process. The GHD Net model has a parameter count of 3.242 M and a computational complexity of 12.0 GFLOPs, achieving average accuracies of 79.7% and 73.9% in green tomato detection and segmentation tasks, respectively. GHD Net outperforms existing models in terms of accuracy and computational complexity, meeting the accuracy and efficiency requirements for green fruit recognition in smart agricultural equipment.

Accurate detection of apple blossoms is critical for monitoring flowering status and optimizing agricultural management. Traditional methods often fail to address challenges such as overlapping petals and environmental variability, leading to inefficiency and inaccuracy. In this paper, LEAF-Net, a modified YOLOv11-based target detection model, is proposed. The original C3k2 module in YOLOv11 lacks a targeted attention mechanism and exhibits insufficient enhancement of key features such as petal edges. Therefore, we propose our model, LEAF-Net, which incorporates a Multi-scale Attention Enhanced Block (MAEB) that enhances edge feature extraction through a hierarchical attention mechanism and reconstructs the C3k2 module. A Frequency-aware Feature Pyramid Network (Freq-FPN) that optimizes multi-scale feature fusion while preserving high-frequency details; and a comprehensive apple blossom dataset capturing diverse growth stages and environmental conditions. To address the dataset deficiencies, a specialized apple blossom dataset with complex backgrounds is constructed. Experimental results demonstrate state-of-the-art performance, with LEAF-Net achieving 90.4% mAP50 and 70.4% mAP50-95, significantly outperforming existing benchmarks. The framework’s computational efficiency (7.1 GFLOPs) and adaptability make it suitable for real-time deployment in precision agriculture. These advancements provide an extensible framework for precision orchard surveillance, thereby paving the way for their adaptive deployment in diverse agricultural automation contexts.

Fruit detection in an unstructured orchard is a foundational visual task employed in plant phenotyping. This is a challenging task due to the minimal contrast between the foreground target and the background color. This situation seriously affects the accuracy of the target fruit detection model, especially for small pear fruits which are more difficult to recognize. Addressing this challenge, an end-to-end small object fruit detection model based on query vectors, SFD-Net, is proposed to achieve more accurate detection of young pear fruits and thus provide guidance for fruit thinning operations in pear orchards for farmers. Specifically, SFD-Net employs the lightweight MobileNetv2 network as a feature extraction framework to obtain the distinctive features of green fruits, then these features are subsequently fed directly into the decoder, where they are integrated with embedded query vectors for fruit identification and localization prediction. This approach efficiently detects small-scale fruits, benefiting from the global sensing capability of query vectors. Importantly, the model avoids the inclusion of redundant multi-scale feature pyramid networks and feature encoders, achieving an AP value of 51.9% and an APS value of 25.1% on the pear dataset, which escalates to an APS value of 27.9% after undergoing additional training iterations. Experimental results unequivocally demonstrate that SFD-Net outperforms other state-of-the-art methods, offering a significant advantage, particularly when considering its suitability for deployment on agricultural smart devices.

Accurate fruit detection in complex orchard environments remains challenging due to variable lighting conditions and weather factors. This paper proposes an optimized contour segmentation model for green spherical fruits (apples and persimmons) based on the E2EC framework. The model employs DLA34 as the backbone network for feature extraction enhanced by a path aggregation balanced feature pyramid network (PAB FPN) with embedded attention mechanisms to refine feature representation. For contour segmentation, we introduce a Cycle MLP Aggregation Deformation (CMAD) module that incorporates cycleMLP to expand the receptive field and improve contour accuracy. Experimental results demonstrate the model’s effectiveness, achieving average precision (AP) and average recall (AR) of 75.5% and 80.4%, respectively, for green persimmons and 57.8% and 64.0% for green apples, outperforming previous segmentation methods. These advancements contribute to the development of more robust smart agriculture systems.

Corn leaf pest and disease detection techniques play a vital role in enabling precise pesticide application and optimizing disease management strategies. However, traditional machine learning methods require manual feature design and exhibit limited generalization capabilities. While early deep learning models improved automation, they struggled to handle the complexity of diverse pest and disease features effectively. ​The primary objective of this study is to address these limitations by developing a novel detection model, Gsconv-Adaptive Feature Fusion Network (GAF-Net), which enhances the accuracy and robustness of corn pest and disease detection. Built upon YOLOv8, GAF-Net incorporates the adaptive spatial feature fusion strategy and the Grouped Spatial Convolution (GSConv) module, which significantly improve multiscale spot feature integration and effective processing of damaged leaf regions. ​This innovation enables GAF-Net to achieve superior performance in complex agricultural environments. To overcome the scarcity of high-quality datasets, we constructed an integrated image dataset encompassing four major pest and disease features: Northern corn leaf blight, pest damage, brown spot, and downy mildew. Experimental results demonstrate that GAF-Net achieves outstanding performance, with a mean average precision (mAP)_50 score of 80.8% (a 4.1% improvement over the baseline model) and an mAP (50–95) score of 52.6% (a 3.3% enhancement). These quantified results highlight the efficacy of GAF-Net in addressing the challenges of corn pest and disease detection, making it a valuable tool for precision agriculture.