Pine wood nematode disease is a devastating pine disease that poses a great threat to forest ecosystems. The use of remote sensing methods can achieve macroscopic and dynamic detection of this disease; however, the efficiency and accuracy of traditional remote sensing image recognition methods are not always sufficient for disease detection. Deep convolutional neural networks (D-CNNs), a technology that has emerged in recent years, have an excellent ability to learn massive, high-dimensional image features and have been widely studied and applied in classification, recognition, and detection tasks involving remote sensing images. This paper uses Gaofen-1 (GF-1) and Gaofen-2 (GF-2) remote sensing images of areas with pine wood nematode disease to construct a D-CNN sample dataset, and we train five popular models (AlexNet, GoogLeNet, SqueezeNet, ResNet-18, and VGG16) through transfer learning. Finally, we use the “macroarchitecture combined with micromodules for joint tuning and improvement” strategy to improve the model structure. The results show that the transfer learning effect of SqueezeNet on the sample dataset is better than that of other popular models and that a batch size of 64 and a learning rate of 1 × 10⁻⁴ are suitable for SqueezeNet’s transfer learning on the sample dataset. The improvement of SqueezeNet’s fire module structure by referring to the Slim module structure can effectively improve the recognition efficiency of the model, and the accuracy can reach 94.90%. The final improved model can help users accurately and efficiently conduct remote sensing monitoring of pine wood nematode disease.

Research Highlights: This study looks at poplar canker caused by Cytospora chrysosperma as a geographical phenomenon, and it applies spatial statistics to reveal the pattern and aggregation effects of the disease on a large scale in time and space. The incidence area of poplar canker in Northeast China has spatial (spatiotemporal) aggregation effects, which emphasize the importance of coordinated prevention. The results of spatial and spatiotemporal clusters can guide specific regional prevention and indicate the possible predisposing factors, respectively. Background and Objectives: Poplar canker, a harmful forest biological disease that is widespread throughout Northeast China, brings enormous ecological and economic losses. The limited cognition of its spatiotemporal pattern and aggregation effects restricts the decision-making for regional prevention and the identification of disease-inducing conditions. This study aims to explore the spatiotemporal pattern and to detect the aggregation effects of the disease, trying to provide references for prevention. Materials and Methods: According to the incidence data of poplar canker reported by each county in Northeast China from 2002 to 2015, we mapped the distribution of the incidence rate in ArcGIS and performed retrospective scan statistics in SaTScan to detect the spatial and spatiotemporal aggregation effects of the incidence area. Results: The spatiotemporal pattern of poplar canker’s incidence rate presents the characteristic of “outbreak-aggregation-spread-stability.” The incidence area of the disease when we performed spatial aggregation scan statistics showed the primary cluster covering Liaoning province (LLR = 86469.86, p < 0.001). The annual spatial scan statistics detected a total of 14 primary clusters and 37 secondary clusters, indicating three phases of aggregation. The incidence area of disease also shows spatiotemporal aggregation effects with the primary cluster located around Liaoning province, appearing from 2009 to 2015 (LLR = 64182.00, p < 0.001). Conclusions: The incidence area of poplar canker presents significant characteristics of spatial and spatiotemporal aggregation, and we suggest attaching importance to the clues provided by the aggregation effects in disease prevention and identification of predisposing factors.