Yan Xiaoyangdi, Zhou Honglin, Zhu Jiangzhang, He Mingfang, Zhao Tianrui, Tan Xiaobo, Zeng Jiangquan
College of Electronic Information & Physics, Central South University of Forestry and Technology, Changsha 410004, China.
College of Computer & Mathematics, Central South University of Forestry and Technology, Changsha 410004, China.
Plants (Basel). 2025 Aug 8;14(16):2465. doi: 10.3390/plants14162465.
High-quality disease segmentation plays a crucial role in the precise identification of rice diseases. Although the existing deep learning methods can identify the disease on rice leaves to a certain extent, these methods often face challenges in dealing with multi-scale disease spots and irregularly growing disease spots. In order to solve the challenges of rice leaf disease segmentation, we propose KBNet, a novel multi-modal framework integrating language and visual features for rice disease segmentation, leveraging the complementary strengths of CNN and Transformer architectures. Firstly, we propose the Kalman Filter Enhanced Kolmogorov-Arnold Networks (KF-KAN) module, which combines the modeling ability of KANs for nonlinear features and the dynamic update mechanism of the Kalman filter to achieve accurate extraction and fusion of multi-scale lesion information. Secondly, we introduce the Boundary-Constrained Physical-Information Neural Network (BC-PINN) module, which embeds the physical priors, such as the growth law of the lesion, into the loss function to strengthen the modeling of irregular lesions. At the same time, through the boundary punishment mechanism, the accuracy of edge segmentation is further improved and the overall segmentation effect is optimized. The experimental results show that the KBNet framework demonstrates solid performance in handling complex and diverse rice disease segmentation tasks and provides key technical support for disease identification, prevention, and control in intelligent agriculture. This method has good popularization value and broad application potential in agricultural intelligent monitoring and management.
高质量的病害分割在水稻病害的精准识别中起着至关重要的作用。尽管现有的深度学习方法能够在一定程度上识别水稻叶片上的病害，但这些方法在处理多尺度病斑和不规则生长的病斑时往往面临挑战。为了解决水稻叶片病害分割的挑战，我们提出了KBNet，这是一种新颖的多模态框架，它整合了语言和视觉特征用于水稻病害分割，利用了卷积神经网络（CNN）和Transformer架构的互补优势。首先，我们提出了卡尔曼滤波器增强的柯尔莫哥洛夫 - 阿诺德网络（KF - KAN）模块，该模块结合了KANs对非线性特征的建模能力和卡尔曼滤波器的动态更新机制，以实现多尺度病斑信息的准确提取和融合。其次，我们引入了边界约束物理信息神经网络（BC - PINN）模块，该模块将诸如病斑生长规律等物理先验知识嵌入到损失函数中，以加强对不规则病斑的建模。同时，通过边界惩罚机制，进一步提高了边缘分割的准确性并优化了整体分割效果。实验结果表明，KBNet框架在处理复杂多样的水稻病害分割任务方面表现出色，为智能农业中的病害识别、预防和控制提供了关键技术支持。该方法在农业智能监测与管理中具有良好的推广价值和广阔的应用潜力。
