Lab

Research Lab: Remote Sensing for Ecology and Ecosystem Conservation (RSEEC)

About the lab

The main focus in my lab is the development of application-oriented research on digital photogrammetry- and remote sensing-assisted analysis of ecological indicators by means of terrestrial, airborne and spaceborne data sources. Besides, I put considerable efforts on educational and research projects for efficient combination of field- and remote methods for ecosystem invetory as well as biodiversity and lanudse monitoring. My students are aimed to form the next generation of practice-oriented photogrammetry and remote sensing experts specielized for protecting the natural environment .

Featured projects (1)

Project
The aim of this project is to further develop remote sensing solutions (both large-scale satellite remote sensing and local approaches based on unmanned aerial vehicles) for disturbance monitoring and to link them with methods of dendrochronology in order to contribute to an improved overall understanding of current forest damage. Moreover, this aims to better understand the historical development of forest damage processes. The improved understanding of the development of forest damage should ultimately contribute to the development of strategies to improve the sustainability of affected and unaffected forest ecosystems.

Featured research (27)

Vegetation cover maps across ecologically-fragile and particularly arid and semi-arid forest ecosystems are prerequisites for their monitoring and management. Direct and field-based measurements of vegetation cover pose serious challenges due to high costs and inaccessibility in harsh terrains, whereas multispectral remote sensing offers objective, spatially-explicit and rapid alternatives. One of the most straightforward tools is the use of broadband vegetation indices (VIs), which are mathematical derivations from multispectral bands that are correlated with various vegetation traits. There are a number of broadband VIs that reach their optimum performance by calibrating their regulatory parameters. We improved the performance of selected VIs for both greenness estimation and land-cover classification across semi-arid woodlands by optimizing their regulatory parameters. We showed this across two separate areas in highly-fragile and sparse vegetation of Zagros mountains of Iran. Regulatory parameters were optimized by multi-objective particle swarm optimization (MOPSO) for Enhanced VI (EVI) and two innovative, more complex broadband indices that use red, blue, and near-infrared multispectral bands. Then, they were applied to estimate greenness and classify vegetation, and were validated by subsets of very high-resolution optical imagery. The results suggest high accuracy of these indicators for estimating and classifying vegetation compared with the commonly-used broadband VIs. Amongst the improved VIs, the one with a more complex combination of spectral bands comparatively returned the best performance, that was 1.34× and 1.33× higher in greenness estimation and 1.58× higher in classification compared with the benchmark NDVI. They also described a higher variance across systematic transects in both regions. In conclusion, both greenness estimation and classification of semi-arid, sparse woodlands were more accurate by optimizing their regulatory parameters.
پایش تالاب ها برای حفاظت از آنها ضروری است. در این راستا سنجش از دور به دلیل در دسترس بودن تصاویر آرشیوی و تاریخی مقرون به صرفه در مقیاس های مکانی مختلف راه حل های کارآمدی را ارائه می دهد. با این حال، فقدان نمونه‌های آموزشی کافی در زمان‌های مختلف محدودیت قابل‌توجهی برای پایش چند زمانی زیست بوم های تالابی است. در این مطالعه، یک روش جدید مبتنی بر انتقال نمونه های آموزشی (Training sample migration) برای شناسایی نمونه‌های بدون تغییر جهت استفاده در طبقه‌بندی و پایش تغییرات تالاب‌ بین‌المللی شادگان در استان خوزستان توسعه داده شد. برای این منظور ابتدا نقشه تالاب در سال مرجع با نمونه های موجود آموزشی با ترکیب داده های ماهواره های سنتینل 1 و 2 در سامانه گوگل ارث انجین (Google Earth Engine) تهیه شده و سپس یک روش خودکار تشخیص تغییرات برای انتقال نمونه‌های آموزشی بدون تغییر از سال مرجع به سال‌های هدف توسعه داده شد. صحت کلی (OA) و ضریب کاپا (KC) این نقشه مرجع به ترتیب 97.93% و 0.97 بود. سپس، یک روش تشخیص تغییر خودکار برای انتقال نمونه‌های آموزشی بدون تغییر از سال مرجع به سال‌های هدف 2018، 2019، و 2021 توسعه داده شد. در روش پیشنهادی، سه شاخص NDVI، NDWI، و انحراف میانگین باندهای طیفی، همراه با دو معیار تشابه فاصله اقلیدسی (ED) و فاصله زاویه طیفی (SAD) برای هر جفت سال مرجع - هدف محاسبه شد. آستانه بهینه برای نمونه های بدون تغییر نیز با استفاده از رویکرد آستانه گذاری هیستوگرام به دست آمد که منجر به انتخاب نمونه هایی شد که به احتمال زیاد برای طبقه بندی مجموعه داده آزمایشی بدون تغییر بودند. روش پیشنهادی به ترتیب منجر به صحت کلی89/95، 83/96، و 06/97 درصد و ضرایب کاپای 95/0، 96/0 و 96/0 برای سال های هدف 2018، 2019 و 2021 شد. در نهایت از نمونه های انتقال داده شده برای تهیه نقشه تالاب برای سال های مورد نظر استفاده شد. محصولات توسعه داده شده شامل کد های متن باز، نقشه ها و الگوریتم های موجود از قابلیت تجاری سازی بالایی در آینده نزدیک بهره مند هستند.
جنگل‌ها نقش اساسی در حفاظت از منابع آب و خاک ایفا می کنند و خدمات اکوسیستمی متعددی را ارائه می دهد. به همین دلیل باید پایش مستمر در جنگل‌ها انجام شود. کمی سازی ساختار جنگلی از جمله اقدامات مهم در این راستا است. در این تحقیق، ما چارچوبی مبتنی بر فتوگرامتری پهپاد، جهت اندازه‌گیری متغیرهای ساختاری مهم ارتفاع (H) و سطح تاج (A) درختان در سرتاسر جنگل‌های زاگرس در غرب ایران در هر دو فرم رویشی دانه‌زاد و شاخه‌زاد، ارائه داده‌ایم. سپس، رگرسیون خطی چند متغیره با H و A جهت برآورد قطر برابر سینه (DBH) درختان دانه‌زاد استفاده کردیم. متغیرهای برآورد شده در نهایت برای مدل‌سازی زیست‌توده بالای زمینی درخت (AGB) برای هر دو فرم رویشی، توسط معادلات آلومتریک محلی و مدل‌های رگرسیون جنگل تصادفی استفاده شدند. در هر مرحله، متغیرهای برآورد شده با مقایسه‌ی مقادیر دقیق میدانی، مورد ارزیابی قرار گرفتند، که میانگین RMSE برابر با 0.68 متر و 4.74 سانتی‌متر به ترتیب برای H و DBH و RMSE نسبی زیر 10 درصد برای تخمین AGB بیانگر دقت بالای روش ارائه شده است. نتایج به طور کلی یک چارچوب کارآمد را برای تخمین ویژگی های درختان در نواحی کوهستانی و نیمه خشک پیشنهاد می کند.
Biodiversity assessment and forest management require accurate tree species maps, which can be provided by remote sensing. Whereas the application of high-spatial resolution remote sensing data is constrained by high costs, Sentinel-2 (S2) satellites provide free imagery with appropriate spatial, spectral and temporal resolutions for mapping of various forest traits across larger spatial scales. Here we assessed the potential of multidate S2 as well as a Digital Elevation Model (DEM) in classifying tree species across a highly structured and heterogeneous broadleaf forest ecosystem in the Hyrcanian zone of northern Iran. We applied multidate S2 and DEM data as input to a variable selection using random forests algorithm for feature reduction. Ten forest types were classified using random forest algorithm and to evaluate the results we computed area-adjusted confusion matrices. Classifications based on single-date S2 data reached overall accuracies of 67–74 per cent, whereas results for multidate S2 images increased the accuracy by ~28 per cent. Joint use of DEM data along with multidate S2 images showed improvement of overall accuracy by ~3 per cent. In addition, we studied the effect of topographic correction of S2 data on classification performance. The results imply that applying topographically corrected imagery had no significant effect on the classification accuracy. Our results demonstrate the high potential of freely available multisource remotely sensed data for broadleaf tree species classification across complex broad-leaved forest landscapes.
Management zones (MZ) are defined as sub-units of farm fields with a relatively homogeneous combination of yield-limiting factors. Each zone can be managed with a different but specific single-rate management practice to maximize the efficiency of farm inputs in the context of precision agriculture. The purpose of this work was to generate MZ based on the Sentinel-2 satellite data for variable rate application using a remote sensing (RS)-based model which merely relies on the RS data. Considering agronomy and climate information in MZ delineation is an inevitable matter since currently, the occurrence of drought is frequent all around the world, e.g., the severe spring drought in major parts of Poland in 2020. In terms of performance and cost, RS data such as UAV or satellite data are more suitable for performing variable rate application than soil characteristics’ data, such as electric conductivity and soil texture. To validate and evaluate the final MZ map, a variety of statistical tests were performed using yield data as reference data. The results showed that delineating MZ using RS satellite data within one growing season incorporating agronomy (wheat phenological phases) and climate information (drought) produces a promising outcome. The simplicity of such a model increases the feasibility of implementing it in farm management information systems. Finally, the RS-based model enables delineating MZ for generating variable rate application maps which can be utilized for specific purposes during a plant growing season, e.g., for fertilizer recommendations that are temporally closer to fertilization times.

Lab head

Hooman Latifi
Department
  • Faculty of Geodesy and Geomatics Engineering
About Hooman Latifi
  • I am an Assistant Professor at the Dept of Photogrammetry and Remote Sensing of the K. N. Toosi University of Technology and an Associate Professor at the Dept. of Remote Sensing of the University of Würzburg. My current research interests are applied spatial analysis of forest entities (Structure, biodiversity and health indicators) by means of spaceborne and airborne remote sensing. In particular, I treasure dealing with UAV and LiDAR--based applications for forest structure analysis, as well as multitemporal and time series of optical data for monitoring vegetation health and phenology.

Members (5)

Dan Kanmegne
  • University of Wuerzburg
Marziye Ghasemi Mobaraki
  • Khaje Nasir Toosi University of Technology
Faez Hussein
  • Universiti Putra Malaysia - K. N. Toosi University of Technology.
Melika Bayat
  • Khaje Nasir Toosi University of Technology
Vahideh Bolandi
  • Khaje Nasir Toosi University of Technology
Elham rezagholi
Elham rezagholi
  • Not confirmed yet

Alumni (12)

Siddhartha Khare
  • Indian Institute of Technology Roorkee
Raja Ram Aryal
  • The University of Queensland
Christian Bauer
  • University of Wuerzburg
Steven Hill
  • University of Wuerzburg