Maize tassels play a critical role in plant growth and yield. Extensive RGB images obtained using unmanned aerial vehicle (UAV) and the prevalence of deep learning provide a chance to improve the accuracy of detecting maize tassels. We used images from UAV, a mobile phone, and the Maize Tassel Counting dataset (MTC) to test the performance of faster region-based convolutional neural network (Faster R-CNN) with residual neural network (ResNet) and a visual geometry group neural network (VGGNet). The results showed that the ResNet, as the feature extraction network, was better than the VGGNet for detecting maize tassels from UAV images with 600 × 600 resolution. The prediction accuracy ranged from 87.94% to 94.99%. However, the prediction accuracy was less than 87.27% from the UAV images with 5280 × 2970 resolution. We modified the anchor size to [85², 128², 256²] in the region proposal network according to the width and height of pixel distribution to improve detection accuracy up to 89.96%. The accuracy reached up to 95.95% for mobile phone images. Then, we compared our trained model with TasselNet without training their datasets. The average difference of tassel number was 1.4 between the calculations with 40 images for the two methods. In the future, we could further improve the performance of the models by enlarging datasets and calculating other tassel traits such as the length, width, diameter, perimeter, and the branch number of the maize tassels.

Leaf age is an important trait in the process of maize (Zea mays L.) growth. It is significant to estimate the seed activity and yield of maize by counting leaves. Detection and counting of the maize leaves in the field are very difficult due to the complexity of the field scenes and the cross-covering of adjacent seedling leaves. A method was proposed in this study for detecting and counting maize leaves based on deep learning with RGB images collected by unmanned aerial vehicles (UAVs). The Mask R-CNN was used to separate the complete maize seedlings from the complex background to reduce the impact of weeds on leaf counting. We proposed a new loss function SmoothLR for Mask R-CNN to improve the segmentation performance of the model. Then, YOLOv5 was used to detect and count the individual leaves of maize seedlings after segmentation. The 1005 field seedlings images were randomly divided into the training, validation, and test set with the ratio of 7:2:1. The results showed that the segmentation performance of Mask R-CNN with Resnet50 and SmoothLR was better than that with LI Loss. The average precision of the bounding box (Bbox) and mask (Mask) was 96.9% and 95.2%, respectively. The inference time of single image detection and segmentation was 0.05 s and 0.07 s, respectively. YOLOv5 performed better in leaf detection compared with Faster R-CNN and SSD. YOLOv5x with the largest parameter had the best detection performance. The detection precision of fully unfolded leaves and newly appeared leaves was 92.0% and 68.8%, and the recall rates were 84.4% and 50.0%, respectively. The average precision (AP) was 89.6% and 54.0%, respectively. The rates of counting accuracy for newly appeared leaves and fully unfolded leaves were 75.3% and 72.9%, respectively. The experimental results showed the possibility of current research on exploring leaf counting for field-grown crops based on UAV images.

Apple flower detection is an important project in the apple planting stage. This paper proposes an optimized detection network model based on a generative module and pruning inference. Due to the problems of instability, non-convergence, and overfitting of convolutional neural networks in the case of insufficient samples, this paper uses a generative module and various image pre-processing methods including Cutout, CutMix, Mixup, SnapMix, and Mosaic algorithms for data augmentation. In order to solve the problem of slowing down the training and inference due to the increasing complexity of detection networks, the pruning inference proposed in this paper can automatically deactivate part of the network structure according to the different conditions, reduce the network parameters and operations, and significantly improve the network speed. The proposed model can achieve 90.01%, 98.79%, and 97.43% in precision, recall, and mAP, respectively, in detecting the apple flowers, and the inference speed can reach 29 FPS. On the YOLO-v5 model with slightly lower performance, the inference speed can reach 71 FPS by the pruning inference. These experimental results demonstrate that the model proposed in this paper can meet the needs of agricultural production.

Crop seedling detection is an important task in the seedling stage of crops in fine agriculture. In this paper, we propose a high-precision lightweight object detection network model based on a multi-activation layer and depth-separable convolution module to detect crop seedlings, aiming to improve the accuracy of traditional artificial intelligence methods. Due to the insufficient dataset, various image enhancement methods are used in this paper. The dataset in this paper was collected from Shahe Town, Laizhou City, Yantai City, Shandong Province, China. Experimental results on this dataset show that the proposed method can effectively improve the seedling detection accuracy, with the F1 score and mAP reaching 0.95 and 0.89, respectively, which are the best values among the compared models. In order to verify the generalization performance of the model, we also conducted a validation on the maize seedling dataset, and experimental results verified the generalization performance of the model. In order to apply the proposed method to real agricultural scenarios, we encapsulated the proposed model in a Jetson logic board and built a smart hardware that can quickly detect seedlings.

Accurate segmentation of individual leaves of sugar beet plants is of great significance for obtaining the leaf-related phenotypic data. This paper developed a method to segment the point clouds of sugar beet plants to obtain high-quality segmentation results of individual leaves. Firstly, we used the SFM algorithm to reconstruct the 3D point clouds from multi-view 2D images and obtained the sugar beet plant point clouds after preprocessing. We then segmented them using the multiscale tensor voting method (MSTVM)-based region-growing algorithm, resulting in independent leaves and overlapping leaves. Finally, we used the surface boundary filter (SBF) method to segment overlapping leaves and obtained all leaves of the whole plant. Segmentation results of plants with different complexities of leaf arrangement were evaluated using the manually segmented leaf point clouds as benchmarks. Our results suggested that the proposed method can effectively segment the 3D point cloud of individual leaves for field grown sugar beet plants. The leaf length and leaf area of the segmented leaf point clouds were calculated and compared with observations. The calculated leaf length and leaf area were highly correlated with the observations with R² (0.80–0.82). It was concluded that the MSTVM-based region-growing algorithm combined with SBF can be used as a basic segmentation step for high-throughput plant phenotypic data extraction of field sugar beet plants.

Extracting robust features in complex scenes remains challenging in person re-identification (ReID). Convolutional neural networks (CNNs) have achieved great success but struggle to capture global pedestrian information. While Transformers excel at modeling global dependencies, they often miss local details. Although CNN-Transformer fusion has been explored, existing methods often adopt cascaded architectures or simple concatenation, which lack continuous and deep interaction between the two branches, thereby limiting their ability to exploit complementary visual information. To address these limitations, we propose a hierarchical deep synergistic fusion (HDSF) framework of CNN and Transformer for person ReID. HDSF facilitates bidirectional complementary interaction and feature enhancement between the two branches across multiple hierarchical levels, allowing them to fully complement and benefit each other. Specifically, we design a Deep Synergistic Fusion Mechanism (DSFM), comprising Bidirectional Feature Calibration (BFC) and Bidirectional Synergistic Enhancement (BSE) modules. The BFC module performs structural alignment and refinement of features between the heterogeneous branches. The BSE module facilitates both cross-branch interaction and self-branch optimization. Its key submodules include Cross-branch Complementary Enhancement (CCE), which enables complementary feature learning between branches, and Self-branch Salience Enhancement (SSE), which emphasizes salient identity-related cues within branches to reduce the influence of background clutter. Furthermore, we propose a multi-granularity supervision strategy to provide effective supervisory signals across multiple scales and semantic levels. Extensive experiments on four large-scale benchmarks demonstrate that HDSF outperforms most state-of-the-art approaches. Compared to our baseline, HDSF improves mAP and Rank-1 by 11.1% and 5.0% on Market1501, and by 12.2% and 8.8% on DukeMTMC.