Geri Dön

Leaf area index and above ground biomass estimation fromunmanned aerial vehicle and terrestrial lidar data usingmachine learning approaches

Başlık çevirisi mevcut değil.

  1. Tez No: 771875
  2. Yazar: GÖZDE COŞKUN
  3. Danışmanlar: DR. BENJAMİN BREDE, DR. HARM BARTHOLOMEUS
  4. Tez Türü: Yüksek Lisans
  5. Konular: Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Bilim ve Teknoloji, Havacılık ve Uzay Mühendisliği, Computer Engineering and Computer Science and Control, Science and Technology, Aeronautical Engineering
  6. Anahtar Kelimeler: Terrestrial LiDAR, UAV LiDAR, Above Ground Biomass, Leaf Area Index, Machine learning, Precision Agriculture
  7. Yıl: 2021
  8. Dil: İngilizce
  9. Üniversite: Wageningen Universiteit
  10. Enstitü: Yurtdışı Enstitü
  11. Ana Bilim Dalı: Belirtilmemiş.
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 67

Özet

Above-ground biomass (AGB) and leaf area index (LAI) are two critical crop parameters for evaluating crop growth, health, and productivity. Various techniques have been used in the past for the estimation of crop parameters. Development and accessibility of remote sensing technologies by airborne, ground devices, satellites opened a new era for estimating the crop parameter. In this paper, the crop parameters LAI and AGB were estimated using Unmanned Aerial Vehicle LiDAR and ground-based Terrestrial Laser Scanning (TLS) data applying three machine learning methods over four crop types with different canopy properties; wheat, sugarbeet, soybean and maize. The study compared how the predictive performance of three modelling techniques multiple linear regression (MLR), random forest regression (RFR), support vector machine regression (SVR) differ, how the predictability performance differ between crop parameters, crop types and LiDAR sensors. The analysis was carried out using all the crop samples obtained and the crop-specific. Two statistical criteria were used as evaluation metrics, the coefficient of determination (R2 ) and Root Mean Square Error (RMSE). The results obtained from the models applied with all crop samples indicated that the predictability performance of AGB was over 0.80 𝑅 2 and the LAI was predicted with the 𝑅 2 of 0.76. In crop-specific modelling, the estimation of soybean and maize AGB and LAI parameters with models applied with features obtained from both UAV-LS and TLS data was satisfactory. The prediction was failed on sugar beet AGB estimation (𝑅 2 in 0.54–0.76 range) and wheat LAI prediction (𝑅 2 in 0.46–0.86 range). Based on the model performances, the RFR and SVR models performed better than the linear model MLR in most of the analyses. The models compared to the data of the sensors used in this study, the models applied with the features obtained from UAV-LS and TLS gave similar 𝑅 2 and 𝑅𝑀𝑆𝐸 when made with all crop samples. However, in crop-specific analyses, the models applied with the features obtained from UAV-LS showed a powerful performance in all LAI estimation models. Although the results obtained from models applied using TLS and UAV-LS features to estimate AGB were very close, only sugarbeet AGB estimation gave better results with TLS data than UAV-LS. Given these results, the lacking application in crop studied with LiDAR remote sensing will be a step to close to fill the gap using both TLS and UAV-LS derived data for several crop types.

Özet (Çeviri)

Özet çevirisi mevcut değil.

Benzer Tezler

  1. Aiding agricultural practices with the exploration of earth observation data via machine learning

    Yer gözlem uydu verilerinin tarımsal uygulamalara yardımcı olmak amacıya makine öğrenme algoritmaları ile incelenmesi

    MEHMET FURKAN ÇELİK

    Doktora

    İngilizce

    İngilizce

    2023

    Jeodezi ve Fotogrametriİstanbul Teknik Üniversitesi

    Geomatik Mühendisliği Ana Bilim Dalı

    PROF. DR. ESRA ERTEN

  2. DSSAT bitki simülasyon modeli ve uzaktan algılama verilerinden elde edilen indisler kullanılarak buğday verim tahmini: İslahiye ve Nurdağı örneği

    Wheat yield estimation using DSSAT crop simulation model and indices obtained from remote sensing: Islahiye and Nurdagi case

    ÖMER VANLI

    Doktora

    Türkçe

    Türkçe

    2019

    Jeodezi ve Fotogrametriİstanbul Teknik Üniversitesi

    Bilişim Uygulamaları Ana Bilim Dalı

    DOÇ. DR. BURAK BERK ÜSTÜNDAĞ

  3. Spektral vejetasyon indeksleri ile bitkilerin biyofiziksel özelliklerinin tespiti ve değerlendirilmesi

    Determination and assessment of the plants' biophysical characteristics through spectral vegetation indices

    SEZEL KARAYUSUFOĞLU UYSAL

    Doktora

    Türkçe

    Türkçe

    2021

    Meteorolojiİstanbul Teknik Üniversitesi

    Meteoroloji Mühendisliği Ana Bilim Dalı

    PROF. DR. LEVENT ŞAYLAN

  4. Vejetasyon yüzeyindeki sera gazı (CO2) değişiminin mikrometeorolojik yöntem ile belirlenmesi

    Determination of the greenhouse gas (CO2) exchange on the vegetation surface by micrometeorological method

    MERVE KIZIL AYDIN

    Yüksek Lisans

    Türkçe

    Türkçe

    2022

    Meteorolojiİstanbul Teknik Üniversitesi

    Meteoroloji Mühendisliği Ana Bilim Dalı

    PROF. DR. LEVENT ŞAYLAN

  5. Kısa rotasyonlu bitkilerle ağaçlandırma ile farklı kavak (Populus sp.) genotiplerinden toprak-üstü odunsu biyokütle üretimi

    Above-ground woody biomass production and below-ground biomass production of different populus genotypes in a short rotation coppice plantation

    KÜBRA SÖNMEZOĞLU

    Yüksek Lisans

    Türkçe

    Türkçe

    2014

    EnerjiMarmara Üniversitesi

    Biyoloji Ana Bilim Dalı

    DOÇ. DR. İBRAHİM İLKER ÖZYİĞİT

    PROF. DR. REINHART CEULEMANS