Application Evaluation of BJ3 Satellite Data in Remote Sensing Monitoring of Pine Wilt Disease

doi:10.13466/j.cnki.lyzygl.2022.04.016

Abstract

Abstract:

BJ3N is an international leading high-resolution optical remote sensing satellite with independent intellectual property rights of China.The first two satellites were successfully launched in 2021.In order to better understand the application performance of BJ3N,this paper used the relevant image data of BJ3N in Shaoguan City,Guangdong Province to carry out remote sensing monitoring of pine wilt disease,and the application evaluation in image fusion,index calculation,information extraction and other aspects.The results showed that BJ3N data could greatly improve the difference between healthy pines and discolored pines through PANSHARP fusion method and NGRDI index calculation,and effectively enhance the identification effect of discolored pine;the precision rate of intelligently extracting discolored pine through deep learning was 95.8%.,and the recall rate was 88.3%,which met the work needs of remote sensing monitoring of pine wilt disease,and was conducive to the accurate monitoring and prevention of pine wilt disease.

Key words: BJ3N, pine wilt disease, remote sensing, evaluation

CLC Number:

S763
S771.8

QIN Lin, MENG Xianjin, ZHANG Shuihua, XUE Yadong, LIU Xinke, Xing Peng. Application Evaluation of BJ3 Satellite Data in Remote Sensing Monitoring of Pine Wilt Disease[J]. FOREST RESOURCES WANAGEMENT, 2022, 0(4): 126-133.

Figures/Tables 14

Fig.1

Tab.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Tab.2

Fig.7

Tab.3

Tab.4

Fig.8

Tab.5

Fig.9

References 18

[1]	国家林业和草原局政府网. 国家林业和草原局公告(2021年松材线虫病疫区)[EB/OL].(2021-03-29)[2022-04-10]. http://www.forestry.gov.cn/main/3457/20210329/151957233445926.html.
[2]	国家林业和草原局政府网. 国家林业和草原局公告(2021年全国松材线虫病新发县级疫区公告)[EB/OL].(2021-08-09)[2022-04-10]. http://www.forestry.gov.cn/main/6206/20220406/151529632107688.html.
[3]	陶欢, 李存军, 程成, 等. 松材线虫病变色松树遥感监测研究进展[J]. 林业科学研究, 2020, 33(3):172-183.
[4]	亓兴兰, 曹祖宁, 刘健, 等. 基于卫星遥感影像的森林病虫害监测研究进展[J]. 林业资源管理, 2020(2):181-186.
[5]	汪飞剑. 松材线虫病遥感监测应用潜力及难点分析[J]. 安徽林业科技, 2021, 47(4):37-41.
[6]	武红敢, 王晗, 常原飞, 等. 刍议枯死松树的天空地协同监测技术体系建设[J]. 中国森林病虫, 2020, 39(3):35-39.
[7]	武红敢, 苗振旺, 王晓俪, 等. 森林病虫灾害的天空地协同监测技术体系示范[J]. 山西林业科技, 2020, 49(3):9-12.
[8]	戴进, 罗伦, 张华. 基于ENVI的北京2号遥感影像融合方法对比研究[J]. 江西测绘, 2020(3):40-43.
[9]	邵亚奎, 朱长明, 张新, 等. 国产高分卫星遥感影像融合方法比较与评价[J]. 测绘通报, 2019(6):5-10.
[10]	孙攀, 董玉森, 陈伟涛, 等. 高分二号卫星影像融合及质量评价[J]. 国土资源遥感, 2016, 28(4):108-113.
[11]	何静. WorldViewⅡ全色与多光谱影像融合算法的比较研究[J]. 测绘技术装备, 2016, 18(1):33-36.
[12]	孙勇. 应用遥感数据对植被指数提取分析[D]. 哈尔滨: 东北林业大学, 2017.
[13]	苗静, 赵梓淇, 刘焕莉, 等. 应用MODIS数据监测大范围病虫害植被指数变化[J]. 中国农学通报, 2015, 31(14):148-155.
[14]	麻坤, 赵鹏祥, 季菲菲, 等. 基于植被指数的秦岭森林健康评价方法研究[J]. 西北林学院学报, 2013, 28(6):145-150.
[15]	方灿莹, 王琳, 徐涵秋. 不同植被红边指数在城市草地健康判别中的对比研究[J]. 地球信息科学学报, 2017, 19(10):1382-1392. doi: 10.3724/SP.J.1047.2017.01382
[16]	张日升, 张燕琴. 基于深度学习的高分辨率遥感图像识别与分类研究[J]. 信息通信, 2017(1):110-111.
[17]	李浩, 方伟泉, 李浪浪, 等. 基于深度学习的松材线虫病害松木识别[J]. 林业工程学报, 2021, 6(6):142-147.
[18]	汪晨, 张辉辉, 乐继旺, 等. 基于深度学习和遥感影像的松材线虫病疫松树目标检测[J]. 南京师大学报:自然科学版, 2021, 44(3):84-89.

影像	均值	标准差	信息熵	平均梯度	空间频率
PCA融合影像	102.7651	89.7196	5.3599	6.0688	15.3446
Pansharp融合影像	110.1021	90.9114	5.3631	7.8772	19.7200
HIS融合影像	123.0425	100.9785	5.0588	7.1121	18.9180

植被指数	均值	标准差	变异系数/%	相对极差
归一化植被指数(NDVI)	0.4094	0.2243	54.7875	4.8852
归一化绿红差异指数(NGRDI)	0.1430	0.0865	60.4895	13.9860
归一化差异红色边缘指数 (NDRE)	0.1531	0.1002	65.4474	13.0633

序号	TP/株	FP/株	FN/株	精确率/%	召回率/%
1	262	12	54	95.6	82.9
2	436	15	27	96.7	94.2
3	181	11	37	94.3	83.0
4	106	5	12	95.5	89.8
总计	985	43	130	95.8	88.3