Detection of mistletoe in digital colour infrared images of infested fir trees

VNIR公司 高光谱成像 内脏相册 遥感 白冷杉 计算机科学 人工智能 数码相机 模式识别(心理学) 计算机视觉 植物 地理 生物 生态学 冷杉云杉
作者
Renata Fintić Pernar,Milan Bajić,Mario Ančić,Ante Seletković,Marilena Idžojtić
出处
期刊:Periodicum Biologorum [Croatian Natural Society]
卷期号:109 (1): 67-75 被引量:8
链接
摘要

Silver fir (Abies alba Mill.) is the most widely distributed and the most important commercial conifer species in Croatia. However, silver fir is also the most endangered tree species in these regions. The results of field research show a significant presence of mistletoe (Viscum album L. ssp. abietis (Wiesb.) Abrom.) on silver firs in Croatia. The current work deals with the issue of mistletoe detection, while our previous research was aimed at developing an efficient method of mistletoe imagery acquisition. Ground-based tests, were aimed at formulating a method that will be implemented in the aerial acquisition system. A digital camera MS-3100 and a hyperspectral line scanner V9 covering visible and near infrared bands were used with the acquisition systems. In this article we analyse the suitability of the supervised classification, unsupervised classification combined by interactive manual merging of classes for the detection of mistletoe on fir. While the spectral resolution of available sensors is limited to three channels in visible wavelengths (blue, green and red) and one in near infrared (NIR) wavelengths, several enhancement methods were used with the aim of increasing the probability of the detection of mistletoe. The achieved overall classification accuracy was in the range of 62.50 to 70.56%. The results confirm the expectation that mistletoe can be detected using a digital visible near infrared (VNIR) camera and hybrid interpretation that consists of unsupervised classification and the manual merging of clusters. The hyperspectral mistletoe analysis provided the first images of the mistletoe reflection spectrum.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
JinghaoLi发布了新的文献求助10
刚刚
安静的雨兰的应助被榛子松鼠采纳,获得10
刚刚
打打的应助被天涯赤子采纳,获得10
刚刚
Yolo发布了新的文献求助10
刚刚
1秒前
1秒前
wgeg完成签到,获得积分20
1秒前
2秒前
di完成签到,获得积分20
3秒前
3秒前
3秒前
宠爱完成签到,获得积分20
3秒前
red完成签到 ,获得积分20
4秒前
hooooo发布了新的文献求助10
4秒前
ATOM完成签到,获得积分20
4秒前
4秒前
途中发布了新的文献求助20
5秒前
5秒前
和平发展发布了新的文献求助10
6秒前
6秒前
bluee发布了新的文献求助10
6秒前
6秒前
默默航空的应助被ppp采纳,获得10
6秒前
洋芋发布了新的文献求助10
6秒前
6秒前
默默航空的应助被ppp采纳,获得10
6秒前
李健的应助被fzw采纳,获得10
7秒前
王开轩发布了新的文献求助10
7秒前
瓜洲发布了新的文献求助50
7秒前
段存煜完成签到,获得积分10
7秒前
ATOM发布了新的文献求助10
8秒前
JYM完成签到,获得积分10
8秒前
小吉发布了新的文献求助10
8秒前
nmn发布了新的文献求助10
8秒前
9秒前
10秒前
10秒前
10秒前
要吃虾饺发布了新的文献求助10
10秒前
爱学术的小宋完成签到,获得积分10
11秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
Markov Decision Processes and Reinforcement Learning 700
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8171039
求助须知:如何正确求助?哪些是违规求助? 9563751
关于积分的说明 21772122
捐赠科研通 7762548
什么是DOI,文献DOI怎么找? 3430786
关于科研通互助平台的介绍 2554676
邀请新用户注册赠送积分活动 2475657