Accurate recognition of the canopy is a prerequisite for precision orchard yield estimation. This paper proposed an enhanced LA-dpv3+ approach for the recognition of cherry canopies based on UAV image data, with a focus on enhancing feature representation through the implementation of an attention mechanism. The attention mechanism module was introduced to the encoder stage of the DeepLabV3+ architecture, which improved the network’s detection accuracy and robustness. Specifically, we developed a diagonal discrete cosine transform feature strategy within the attention convolution module to extract finer details of canopy information from multiple frequency components. The proposed model was constructed based on a lightweight DeepLabv3+ network architecture that incorporates a MobileNetv2 backbone, effectively reducing computational costs. The results demonstrate that our proposed method achieved a balance between computational cost and the quality of results when compared to competing approaches. Our model’s accuracy exceeded 89% while maintaining a modest model size of only 46.8 MB. The overall performance indicated that with the help of a neural network, segmentation failures were notably reduced, particularly in high-density weed conditions, resulting in significant increases in accuracy (ACC), F1-score, and intersection over union (IOU), which were increased by 5.44, 3.39, and 8.62%, respectively. The method proposed in this paper may be applied to future image-based applications and contribute to automated orchard management.

Growing consumer demand for high-quality strawberries has highlighted the need for accurate, efficient, and non-destructive methods to assess key postharvest quality traits, such as weight, size uniformity, and quantity. This study proposes a multi-objective learning algorithm that leverages RGB-D multimodal information to estimate these quality metrics. The algorithm develops a fusion expert network architecture that maximizes the use of multimodal features while preserving the distinct details of each modality. Additionally, a novel Heritable Loss function is implemented to reduce redundancy and enhance model performance. Experimental results show that the coefficient of determination (R²) values ​​for weight, size uniformity and number are 0.94, 0.90 and 0.95 respectively. Ablation studies demonstrate the advantage of the architecture in multimodal, multi-task prediction accuracy. Compared to single-modality models, non-fusion branch networks, and attention-enhanced fusion models, our approach achieves enhanced performance across multi-task learning scenarios, providing more precise data for trait assessment and precision strawberry applications.