Effective identification of apple leaf diseases can reduce pesticide spraying and improve apple fruit yield, which is significant to agriculture. However, the existing apple leaf disease detection models lack consideration of disease diversity and accuracy, which hinders the application of intelligent agriculture in the apple industry. In this paper, we explore an accurate and robust detection model for apple leaf disease called Apple-Net, improving the conventional YOLOv5 network by adding the Feature Enhancement Module (FEM) and Coordinate Attention (CA) methods. The combination of the feature pyramid and pan in YOLOv5 can obtain richer semantic information and enhance the semantic information of low-level feature maps but lacks the output of multi-scale information. Thus, the FEM was adopted to improve the output of multi-scale information, and the CA was used to improve the detection efficiency. The experimental results show that Apple-Net achieves a higher mAP@0.5 (95.9%) and precision (93.1%) than four classic target detection models, thus proving that Apple-Net achieves more competitive results on apple leaf disease identification.

Early detection of apple leaf diseases is essential for enhancing orchard management efficiency and crop yield. This study introduces LightYOLO-AppleLeafDx, a lightweight detection framework based on an improved YOLOv8 model. Key enhancements include the incorporation of Slim-Neck, SPD-Conv, and SAHead modules, which optimize the model’s structure to improve detection accuracy and recall while significantly reducing the number of parameters and computational complexity. Ablation studies validate the positive impact of these modules on model performance. The final LightYOLO-AppleLeafDx achieves a precision of 0.930, mAP@0.5 of 0.965, and mAP@0.5:0.95 of 0.587, surpassing the original YOLOv8n and other benchmark models. The model is highly lightweight, with a size of only 5.2 MB, and supports real-time detection at 107.2 frames per second. When deployed on an RV1103 hardware platform via an NPU-compatible framework, it maintains a detection speed of 14.8 frames per second, demonstrating practical applicability. These results highlight the potential of LightYOLO-AppleLeafDx as an efficient and lightweight solution for precision agriculture, addressing the need for accurate and real-time apple leaf disease detection.

Multispectral remote sensing data with abundant spectral information can be used to compute vegetation indices to improve the accuracy of Ginkgo biloba yield prediction. The limited spatial resolution of multispectral cameras restricts the detail capture over wide farmland, but super-resolution (SR) reconstruction methods can enhance image quality. However, most existing SR models have been trained on images processed from downsampled high-resolution (HR) images, making them less effective in reconstructing real low-resolution (LR) images. This study proposes a GAN-based super-resolution reconstruction method (RMSRGAN) for multispectral remote sensing images of Ginkgo biloba trees in real scenes. A U-Net-based network is employed instead of the traditional discriminator. Convolutional block attention modules (CBAMs) are incorporated into the Residual-in-Residual Dense Blocks (RRDBs) of the generator and the U-Net of the discriminator to preserve image details and texture features. An unmanned aerial vehicle (UAV) equipped with a multispectral camera was employed to capture field multispectral remote sensing images of Ginkgo biloba trees at different spatial resolutions. Four matching HR and LR datasets were created from these images to train RMSRGAN. The proposed model outperforms the traditional models by achieving superior results in both quantitative evaluation metrics (peak signal-to-noise ratio (PSNR) is 32.490, 31.085, 27.084, 26.819, and structural similarity index (SSIM) is 0.894, 0.881, 0.832, 0.818, respectively) and qualitative evaluation visualization. Furthermore, the efficiency of our proposed method was tested by generating individual vegetation indices (VIs) from images taken before and after reconstruction to predict the yield of Ginkgo biloba. The results show that the SR images exhibit better $R^2$ and $RMSE$ values than LR images. These findings show that RMSRGAN can improve the spatial resolution of real multispectral images, increasing the accuracy of Ginkgo biloba yield prediction and providing more effective and accurate data support for crop management.

Face recognition is one of the essential applications in computer vision, while current face recognition technology is mainly based on 2D images without depth information, which are easily affected by illumination and facial expressions. This paper presents a fast face recognition algorithm combining 3D point cloud face data with deep learning, focusing on key part of face for recognition with an attention mechanism, and reducing the coding space by the sparse loss function. First, an attention mechanism-based convolutional neural network was constructed to extract facial features to avoid expressions and illumination interference. Second, a Siamese network was trained with a sparse loss function to minimize the face coding space and enhance the separability of the face features. With the FRGC face dataset, the experimental results show that the proposed method could achieve the recognition accuracy of 95.33%.

Three-dimensional (3D) face recognition has become a trending research direction in both industry and academia. However, traditional facial recognition methods carry high computational costs and face data storage costs. Deep learning has led to a significant improvement in the recognition rate, but small sample sizes represent a new problem. In this paper, we present an expression-invariant 3D face recognition method based on transfer learning and Siamese networks that can resolve the small sample size issue. First, a landmark detection method utilizing the shape index was employed for facial alignment. Then, a convolutional network (CNN) was constructed with transfer learning and trained with the aligned 3D facial data for face recognition, enabling the CNN to recognize faces regardless of facial expressions. Following that, the weighted trained CNN was shared by a Siamese network to build a 3D facial recognition model that can identify faces even with a small sample size. Our experimental results showed that the proposed method reached a recognition rate of 0.977 on the FRGC database, and the network can be used for facial recognition with a single sample.

This study aimed at the problems of dense distribution and different sizes of pests in cantaloupe fields, as well as the complex and variable characteristics of some pests. To improve the detection accuracy of pests in cantaloupe leaves, a multi-strategy dynamic feature fusion detection algorithm is proposed, which provides a reference for technological paths for pest control in cantaloupe fields. First, the Melon Cantaloupe Pest dataset, containing five types of cantaloupe pests from the Robotflow website, was used and processed for image culling and expansion. Second, the YOLOv12 model is improved by (1) adding the EMA attention mechanism at the end of its backbone network, (2) changing the fusion strategy of the Concat layer in the Neck network as well as the Detect layer in the Head network, (3) introducing the WIoU v3 loss function, and (4) improving the C3k2 module. The experiments showed that the model proposed in this study achieved 85.06%, 86.76%, 79.94%, 54.15% and 83.08% for mAP50, P, R, mAP50-95, and F₁, respectively. It outperformed all other comparative models in four metrics except the R. Additionally, compared to the original YOLOv12 model, the four metrics of the improved model were improved by 3.48%, 3.17%, 2.5%, 3.45% and 3.04%, while the number of parameters was reduced. In addition, ablation experiments as well as generalization experiments on rice pest and corn pest datasets were conducted in this study. The improved model proposed in this paper effectively enhances the performance of pest detection in different scenarios.