A remote sensing identification method for Dendroctonus armandi based on multi-scale spatial attention mechanism and Gaussian kernel soft labeling

doi:10.14088/j.cnki.issn0439-8114.2026.01.026

Abstract

Abstract: To address the time-consuming and labor-intensive nature of traditional canopy boundary annotation, and the issue of decreased detection accuracy in existing deep learning models due to the loss of spatial details from downsampling in complex forest environments, a single-tree positioning method that integrated a multi-scale spatial attention mechanism convolutional network (MSSCN) and Gaussian kernel function soft labeling was proposed. Using high-resolution aerial remote sensing images from three altitude gradients (2 000 m, 2 200 m, and 2 400 m) in the Shennongjia Forestry District as the data source, only the canopy center points of Dendroctonus armandi were annotated. A two-dimensional Gaussian kernel function was employed to generate confidence maps for labeling and creating the training dataset, thereby transforming the regional segmentation task into a single-tree positioning problem. By adjusting the position of the multi-scale feature convolution module, MSSCN1, MSSCN2, and MSSCN3 models were constructed and compared with the U-Net, FCN, and DeepLabV3+ models. The results showed that the Gaussian kernel function soft labeling method reduced manual annotation costs while supporting the precise localization of infested trees. The MSSCN3 model achieved optimal performance after 100 training epochs, with average precision, recall, and F1-score values of 91.97%, 93.68%, and 0.93 in the test area, respectively, outperforming the other comparative models. The MSSCN3 model generally demonstrated superior detection performance in high-altitude areas of the Shennongjia Forestry District, and its detection accuracy was generally higher in high-outbreak-density areas than in low-outbreak-density areas. However, a slight decrease in model accuracy was observed in the high-outbreak-density area at 2 400 m altitude, indicating that topographic and ecological factors might have interactive effects on detection stability.The MSSCN3 model could accurately identify Pinus armandii infested trees in the Shennongjia Forestry District, providing an efficient and robust technical pathway for pest control.

Key words: multi-scale spatial attention mechanism, Gaussian kernel soft labeling, Dendroctonus armandi, infested trees, remote sensing identification

CLC Number:

S771.8

HUANG Guang-ti, LIN Hao-ran, DIAN Yuan-yong, HAN Ze-min, PENG Shou-lian, LIU Xiao-yang, XIAO Xiao. A remote sensing identification method for Dendroctonus armandi based on multi-scale spatial attention mechanism and Gaussian kernel soft labeling[J]. HUBEI AGRICULTURAL SCIENCES, 2026, 65(1): 159-165.

References

[1] 陈辉,刘琳,赵平娟.秦岭华山松小蠹消化道结构与食性分化的研究[J].西北林学院学报,2004,19(2):89-92.
[2] 洪承昊,柏冰洋,易家喜,等.神农架林区华山松大小蠹危害风险分析[J].中国森林病虫,2021,40(2):17-22.
[3] 白雪琪,张晓丽,张凝,等.基于高光谱遥感的油松毛虫危害程度监测模型[J].北京林业大学学报,2016,38(11):16-22.
[4] 吕晓君,王君,喻卫国,等.无人机监测林业有害生物初探[J].湖北林业科技,2016,45(4):30-33.
[5] MEDDENS A J H, HICKE J A, VIERLING L A, et.al. Evaluating methods to detect bark beetle-caused tree mortality using single-date and multi-date Landsat imagery[J]. Remote sensing of environment, 2013, 132: 49-58.
[6] DENNISON P E, BRUNELLE A R, CARTER V A.Assessing canopy mortality during a mountain pine beetle outbreak using GeoEye-1 high spatial resolution satellite data[J]. Remote sensing of environment, 2010, 114(11): 2431-2435.
[7] WHITE J C, WULDER M A, BROOKS D, et al.Detection of red attack stage mountain pine beetle infestation with high spatial resolution satellite imagery[J]. Remote sensing of environment, 2005, 96(3-4): 340-351.
[8] THANH NOI P, KAPPAS M.Comparison of random forest, k-nearest neighbor, and support vector machine classifiers for land cover classification using sentinel-2 imagery[J]. Sensors, 2018, 18(1): 18.
[9] PEÑA M A, CRUZ P, ROIG M. The effect of spectral and spatial degradation of hyperspectral imagery for the Sclerophyll tree species classification[J]. International journal of remote sensing, 2013, 34(20): 7113-7130.
[10] LIU X, FREY J, DENTER M, et al.Mapping standing dead trees in temperate montane forests using a pixel and object-based image fusion method and stereo WorldView-3 imagery[J]. Ecological indicators, 2021, 133: 108438.
[11] 尹宝才,王文通,王立春,等.深度学习研究综述[J].北京工业大学学报,2015,41(1):48-59.
[12] HAN Z M, HU W J, PENG S L, et al.Detection of standing dead trees after pine wilt disease outbreak with airborne remote sensing imagery by multi-scale spatial attention deep learning and Gaussian kernel approach[J]. Remote sensing, 2022, 14(13): 3075.
[13] 徐信罗. 基于卷积神经网络和无人机遥感的松材线虫病受害木监测研究[D].南昌:南昌大学,2021.
[14] FASSNACHT F E, LATIFI H, GHOSH A, et al.Assessing the potential of hyperspectral imagery to map bark beetle-induced tree mortality[J]. Remote sensing of environment, 2014, 140: 533-548.
[15] HALL R J, CASTILLA G, WHITE J C, et al.Remote sensing of forest pest damage: A review and lessons learned from a Canadian perspective[J]. The Canadian entomologist,2016,148(S1):296-356.
[16] LIN H R, LIU X Y, HAN Z M, et al.Identification of tree species in forest communities at different altitudes based on multi-source aerial remote sensing data[J]. Applied sciences, 2023, 13(8): 4911.
[17] 文仕知,宋良友,喻勋林.神农架华山松群落特征研究[J].中南林业科技大学学报,2010,30(4):91-96.
[18] 陈辉. 小蠹类森林害虫可持续控制的策略和方法[J].西北林学院学报,2002,17(4):62-65.
[19] WAN Q, HUANG Z L, LU J C, et al. SeaFormer++: Squeeze-enhanced axial transformer for mobile visual recognition[EB/OL]. (2023-01-30). https://arxiv.org/abs/2301.13156.
[20] HARALICK R M, SHANMUGAM K, DINSTEIN I.Textural features for image classification[J]. IEEE transactions on systems, man, and cybernetics, 1973, S3(6): 610-621.
[21] 张亚昊,佃袁勇,黄光体,等.不同演替阶段马尾松林林分空间结构对物种多样性的影响[J].生态学杂志,2021,40(8):2357-2365.
[22] HAN Z M, DIAN Y Y, XIA H, et al.Comparing fully deep convolutional neural networks for land cover classification with high-spatial-resolution Gaofen-2 images[J]. ISPRS international journal of geo-information, 2020, 9(8): 478.
[23] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2014-09-04). https://arxiv.org/abs/1409.1556.
[24] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking atrous convolution for semantic image segmentation[EB/OL]. (2017-06-17). https://arxiv.org/abs/1706.05587.