Powdery mildew has become a significant disease affecting the yield and quality of rubber trees in recent years. It typically manifests on the leaf surface at an early stage, rapidly infecting and spreading throughout the leaves. Therefore, early detection and intervention are essential to reduce the resulting losses due to this disease. However, the conventional methods of disease detection are both time-consuming and labor-intensive. In this study, we proposed a novel deep-learning-based approach for detecting powdery mildew in rubber trees, even in complex backgrounds. First, to address the lack of existing datasets on rubber tree powdery mildew, we constructed a dataset comprising 6200 images and 38,000 annotations. Second, based on the YOLO framework, we integrated a multi-scale fusion module that combines a Feature Focus and Diffusion Mechanism (FFDM) into the neck of the detection architecture. We designed an overall focus diffusion architecture and introduced a Dimension-Aware Selective Integration (DASI) module to enhance the detection of small powdery mildew targets, naming the model PM-YOLO. Furthermore, we proposed an automatic grading detection algorithm to evaluate the severity of powdery mildew on rubber tree leaves. The experimental results demonstrated that the proposed method achieved 86.9% mean average precision (mAP) and 85.6% recall, which outperformed the standard YOLOv10 by 7.6% mAP and 8.2% recall. This approach offered accurate and real-time detection of powdery mildew rubber trees, providing an effective solution for early diagnosis through automated grading.

The global economy and society are seriously threatened by the pervasive spread of counterfeit goods. Their high level of simulation makes real and fake goods extremely similar in appearance and difficult to distinguish. The existing identification techniques mostly use CNNs and transformer architectures. However, CNNs have limitations in modeling long‐range dependencies, leading to their limited classification performance, while vision transformers (ViTs), although excellent in modeling long‐range dependencies, the quadratic computational complexity of their self‐attention mechanism makes it difficult to be widely used in real‐world scenarios with limited computational resources. According to recent research, long‐range relationships can be accurately modeled using the state space model (SSM), which is represented by Mamba, while preserving linear computational complexity. Motivated by this, we proposed CGMamba, a SSM‐based intelligent recognition model for counterfeit goods. Specifically, we constructed a novel hybrid basic block called global‐local feature aggregation (GLFA). This block greatly enhances the feature extraction capability for counterfeit goods by deeply integrating the local feature extraction capability of the CNN and the global modeling capability of SSM. It is composed of three components: a local feature extractor, a global feature extractor, and an adaptive feature aggregation module (AFAM). In addition, to address the problem of lack of counterfeit goods image data, we constructed a large counterfeit goods dataset containing 101,480 images covering 104 categories for model training and evaluation. The experimental results showed that CGMamba achieved 90.99% Top 1 accuracy on the self‐constructed dataset and 79.5% on the public dataset CNFOOD‐241, which significantly outperforms the existing methods. The source code is available at https://github.com/wth1998/CGMamba.git .