Bang Nguyen Tran

Vietnam National University of Agriculture · Department of Agro-Ecology

Hi folks,

I wonder if anyone update and can share with me the most effective method for mapping agricultural residue open burning from remote sensing data pls?

- Potential case study: Asia with typical rice cultivation as major food crop

- Scale: Region

- Data: Should active data overcome passive ones or not?

- Method: Spectral shareholding or machine learning???

Thank you very much

Bang

Hi folks,

I wonder if anyone know how to calculate the fractional vegetation cover (FVC) from Landsat imagery? As far as I have read about that NDVI, NDVIs, and NDVIv are three required input for calculation FVC as follow equation:

FVC = ( NDVI − NDVIs )/( NDVIv − NDVIs ) (1)

where:

- NDVI is given by: NDVI = (nir − red)/(nir − red); nir and red are corrected reflectance obtained from the sensor bands located in the near infrared (NIR) and the red spectral regions for each pixel within an image

- The NDVIs and NDVIv are values of the NDVI for bare soil (FVC = 0) and pure green vegetation (FVC = 1) within an image, respectively

However, the major problem when applying Eq.(1) is how to accurately estimate the NDVIs and NDVIv values on the Landsat imagery. Therefore, I need you support on how to retrieve the values for NDVIs and NDVIv and the way we can calculate them from Landsat or other RS images pls?

Thank you very much with kind regards

Bang

Hi mates,

I wonder on current methods for assessing forest recovery from remote sensing perspectives, anyone can recommend me certain approaches pls? The data will be expected from both optical and LIDAR data. My expected analysis is around 11 years in my study site after a major disturbance. I want to analyze trend in forest recovery rates measured by different indices obtained or generated from remote sensing data.

Kind regards

Bang Tran

Hi mate,

I wonder if there is common ways to perform back-transform predicted values from random forest model results pls?

I have run different random forest models for my data based on the transformed observed and predicted variables. Now I need to perform back-transform to the untransformed observed and predicted variables before re-calculating the error. I am not sure if back-transforming the error directly will work given the scaling issue of the transformations pls?

Thank you very much with kind regards

Bang Tran   

I wonder if anyone know how to calculate the Potential solar Radiation (PSR) & Topographic Wetness Index (TWI). in R?

I had DEM imported in R already also with slope, aspect and Topographic position index (TPI). Now I am stuck with how to calculate the Potential solar Radiation (PSR) and also Topographic Wetness Index (TWI).

Thank you.

Hi all,

I ran Random Forest Model in R then now i call it to predict my dataset

predict.rf<-predict(layers.stack,random.forest, na.rm=T, type='response')

However it comes with one error as follow:

Error in UseMethod("predict") :

no applicable method for 'predict' applied to an object of class "c('randomForest.formula', 'randomForest')"

Does anyone know how to fix this error pls?

Regards

Hi everyone,

I am wonder if anyone know codes or the ways to down download Landsat Surface Reflectance dataset (4/5; 7 and 8) automatically in Python? or R? or Matlab?

I need to download datasets for around 10 Landsat scenes in Victoria, Australia from 1983 to 2017 so it would be very useful if anyone can help to guide me the ways for downloading them automatically

I have made alot of attempts with R codes (getSpatialData-master; nasadata-master; getlandsat-master) Python (espa-bulk-downloader-master; theia_download-master; landsat-util-develop); mut they did not work for me

Thank you very much

Bang Tran