HomeEducationDoctorate (PhD & EngD)For current candidatesPhD infoUpcoming public defencesPARTLY DIGITAL - ONLY FOR INVITEES (1,5 m) : PhD Defence Azar Zafari | Integrating tree-based kernels and support vector machine for remote sensing image classification

PARTLY DIGITAL - ONLY FOR INVITEES (1,5 m) : PhD Defence Azar Zafari | Integrating tree-based kernels and support vector machine for remote sensing image classification

Integrating tree-based kernels and support vector machine for remote sensing image classification

Due to the COVID-19 crisis measures the PhD defence of Azar Zafari will take place (partly) online in the presence of an invited audience. 

The PhD defence can be followed by a live stream.

Azar Zafari is a PhD student in the research group Geo-information Processing (GIP). Her supervisor is prof.dr. R. Zurita Milla from the Faculty of Geo-Information Science and Earth Observation (ITC).

There is an ever-increasing need for land cover information, since the population of the world is dependent on Earth as the source of food production and for various economic developments. Land cover maps are key inputs for policymakers in nurturing sustainable plan- ning and management systems at the local, regional, and national levels.

Owing to advances in remote sensing (RS) technology, abundant sources of timely land cover data at various spectral and spatial res- olutions have become available. Using big geo-data from recent Earth observation sensors providing very high spatial resolution (VHR) satel- lite images makes it possible to obtain land cover maps with higher levels of detail. However, the development of efficient classification methods for the new generations of VHR images has become one   of the most challenging problems addressed by the RS community in recent years. The most important challenge associated with new generations of data is the Hughes phenomenon or curse of dimen- sionality that occurs when the number of features is much larger than the number of training samples. Hyperspectral images, time series of multispectral satellite images, and stacking additional fea- tures on top of the original spectral features are usually associated with the Hughes phenomenon. Tree-based ensemble learners such as the random forest (RF) and extra trees (ET) and kernel-based methods such as the support vector machine (SVM) are well-known classifiers in high-dimensional classification problems. The main objective of this dissertation is to investigate the integration of two of the most well-known and recurrently used classifiers by the geospatial com- munity: tree and kernel-based methods.

The performance of the proposed methods is evaluated for crop classification over small-scale farms. The vast majority of low-income country farming is undertaken by smallholder farmers that often struggle to make ends meet. Currently, little is known in quantitative terms regarding the crop growth processes in smallholder farming. There are barely any systems in place that monitor such information, even though such knowledge is crucial for numerous stakeholders in the food production pyramid. Farmer communities (such as the agribusiness sector that supplies farm inputs and those  marketing farm outputs), the financial sector serving farmers, and the govern- mental agencies that work with farmers could utilize such informa- tion. Eventually, individual farmers could also use such information, of course, if given to them in the form of on-farm advice.

Unlike in high-income country farming (where plots are larger, only a single crop is grown, the farm inputs are well-documented, as are the weather conditions, and farm practices are more standardized), monitoring smallholder farming requires the addressing of a much higher variation in these parameters. Farm plots tend to have more irregular geometries and are often only vaguely delineated. In addi- tion, plots are typically not formally registered in a farm cadastre. Moreover, smallholder plots include multiple crops and numerous crop varieties, there is little information about the soils, and un- known inputs are received and can be subject to variable field man- agement. Therefore, research work in this thesis was focused on em- ploying a number of specific VHR image sources to derive crop maps that can be used to improve the understanding of crop conditions in small-scale farms. Such image sources must be multispectral, of high spatial resolution, and the image series must be sufficiently tempor- ally dense. This results in increasing the dimensionality of the data- set used for this study. Therefore, the research described in this dissertation concentrated on exploring the use of tree-based kernels in an SVM for land cover mapping of small-scale agriculture using VHR satellite images.

First, we studied the synergic use of RF and SVM as two well-known and recurrent classifiers for the production of land cover maps through using an RF-based kernel (RFK) in an SVM (SVM-RFK). The performance of this synergic classifier is evaluated by comparing  it against using a customary radial basis function (RBF) kernel in an SVM (SVM-RBF) and standard RF classifiers. Two datasets were used to illustrate the analyses in this study—a time series of seven multispectral WorldView-2 images acquired over Sukumba (Mali) and a single hyperspectral AVIRIS image acquired over Salinas Valley (CA, USA). The features set for Sukumba was extended by obtaining veget- ation indices (VIs) and grey-level co-occurrence matrices (GLCMs) and stacking them to spectral features. For Sukumba, SVM-RFK, RF, and SVM-RBF were trained and tested over 10 subsets once using only spectral features and once using the extended dataset. As bench- marking, the Salinas dataset with only spectral features was also trained and tested over 10 subsets. The results revealed that the newly proposed SVM-RFK performs at almost same level as that of the SVM-RBF and RF in terms of overall accuracy (OA) for the spec- tral features of both datasets. For the extended Sukumba dataset, the results showed that SVM-RFK yields slightly higher OA than RF and it considerably outperforms the SVM-RBF. Moreover, the SVM-RFK sub- stantially reduced the time and computational cost associated with parametrizing the kernel compared to the SVM-RBF. In addition, RF was also used to derive an RFK based on the most important features, which improved the OA of the previous SVM-RFK by 2%. In summary, the proposed SVM-RFK classier achieved substantial improvements when applied to high-dimensional data and when combined with RF- based feature selection methods; it is at least as good as the SVM-RBF and RF when applied to fewer features.

Second, we explored the connection between random forest and ker- nel methods by using various characteristics of RF to generate an improved design of RFK. The classic design of RFK is obtained based on the end-nodes of trees. Here, we investigated the possibility of de- veloping the classic design of RFK by using tree depths, the number of branches among the leaves of trees, and the class probabilities as- signed to samples with RF. Accordingly, we developed a multi-scale RFK which uses multiple depths of RF to create an RF-based ker- nel. All the obtained RFKs are evaluated by importing them into an SVM classifier (i.e., SVM-RFK) to classify the extended Sukumba data- set. The results showed that investigating the depth improves the OA of RFK, particularly for high-dimensional experiments. Other ex- amined designs of RFKs also outperformed the RBF for the extended Sukumba datasets. Using the spectral features for Sukumba, all sug- gested designs of RFKs performed at almost the same level as that of the RBF kernel when they were used in an SVM.

Third, we introduced the use of ETs to create a kernel (ETK) that can be used in an SVM to overcome the limitations of RFK and RBF ker- nel. The use of these kernels in an SVM is also compared with the ET classifier. Four different sets of features were tested by dividing the extended Sukumba dataset. For datasets with fewer features, SVM- ETK slightly outperforms SVM-RBF and SVM-RFK. Moreover, SVM-ETK almost entirely outperforms ET. Apart from OA, the main advantage of ETK is the lower computational cost associated with parametrizing the kernel compared to the RBF and RFK. Our results showed that tree-based kernels (i.e., RFK and ETK) compete closely and yield higher OA than RBF in high-dimensional and noisy experi- ments. Thus, the proposed SVM-ETK classifier outperforms ET, SVM- RFK, and SVM-RBF in a majority of the cases.

Fourth, with regard to the context of open science, we include an R- function to implement the ideas of different designs of tree-based kernels evaluated in this thesis.

In a nutshell, the main conclusion of this PhD thesis is that the ker- nels obtained on the basis of supervised tree-based ensemble learn- ing methods can be used as efficient alternatives to the conventional kernels in kernel-based classifications methods such as the SVM, in particular, in dealing with high-dimensional noisy problems such as mapping small-scale agriculture.