New publication on classification uncertainty

New publication on classification uncertainty

April 15, 2015

Uncertainty_Fabian_LoewA new paper published recently in ISPRS presents our ongoing research on classification uncertainty in the context of multi-temporal object-based classification, based on support vector machines (SVM). It adds a contribution to the question of which factors influence the spatial distribution of classification uncertainty in land use maps.

Agricultural management increasingly uses crop maps based on classification of remotely sensed data. However, classification errors can translate to errors in model outputs, for instance agricultural production monitoring (yield, water demand) or crop acreage calculation. Hence, knowledge on the spatial variability of the classier performance is important information for the user. But this is not provided by traditional assessments of accuracy, which are based on the confusion matrix. In this study, classification uncertainty was analyzed, based on the support vector machines (SVM) algorithm.

SVM was applied to multi-spectral time series data of RapidEye from different agricultural landscapes and years. Entropy was calculated as a measure of classification uncertainty, based on the per-object class membership estimations from the SVM algorithm. Permuting all possible combinations of available images allowed investigating the impact of the image acquisition frequency and timing, respectively, on the classification uncertainty. Results show that multi-temporal datasets decrease classification uncertainty for different crops compared to single data sets, but there was no “one-image-combination-fits-all” solution. The number and acquisition timing of the images, for which a decrease in uncertainty could be realized, proved to be specific to a given landscape, and for each crop they differed across different landscapes. For some crops, an increase of uncertainty was observed when increasing the quantity of images, even if classification accuracy was improved. Random forest regression was employed to investigate the impact of different explanatory variables on the observed spatial pattern of classification uncertainty. It was strongly influenced by factors related with the agricultural management and training sample density. Lower uncertainties were revealed for fields close to rivers or irrigation canals.

This study demonstrates that classification uncertainty estimates by the SVM algorithm provide a valuable addition to traditional accuracy assessments. This allows analyzing spatial variations of the classifier performance in maps and also differences in classification uncertainty within the growing season and between crop types, respectively.

A full text of this paper can be found at:

http://www.sciencedirect.com/science/article/pii/S0924271615000635

 

you may also like:

EORC and EAGLE summer BBQ

EORC and EAGLE summer BBQ

We’re happy to announce that our summer BBQ is happening again on Thursday, July 24th at 4pm! Alongside good food and a relaxed atmosphere, we’re also hosting a series of short talks highlighting exciting topics in Earth Observation and environmental science: “The...

🗺 Exploring Map Visualizations

🗺 Exploring Map Visualizations

Within our EAGLE courses our students have to learn a wide variety of skills - beside the fundamental earth observation theory and practice also skills like map creation is part of the curriculum. One of our students Ronja Seitz has created three visualizations guides...

Successful Completion of UNIversInternational Certificate

Successful Completion of UNIversInternational Certificate

In line with its internationalization strategy, the University of Würzburg supports administrative staff in their task of advising and supporting international students, guests, and academics. To this end, it has launched the "UNIversInternational" certificate...

The “Geolingual Studies” team visited the DLR EOC

The “Geolingual Studies” team visited the DLR EOC

The "Geolingual Studies" team of the University Würzburg visited the DLR-EOC on 3 and 4 July 2025. Geolingual Studies is an innovative area of research and teaching which takes a decisively applied linguistic approach and combines methodologies from linguistics,...

Course on Object-based image analysis

Course on Object-based image analysis

Dr. Michael Wurm from the German Aerospace Center (DLR) gave a class about Object-based image analysis (OBIA) using the eCognition Software for the EAGLE students. The course gives an insight into the theoretical basis of OBIA and using different datasets and tasks...