Remote sensing image classification for forestry using MRF models and VQ method (original) (raw)
Related papers
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019
Markov random field (MRF) based methods have been widely used in high spatial resolution (HSR) image classification. However, many existing MRF-based methods put more emphasis on pixel level contexts while less on superpixel level contextual information. To cope with this issue, this article presents a novel bilevel contextual MRF framework, named BLC-MRF, for HSR imagery classification. Specifically, pixel and superpixel level dependence are incorporated into the proposed MRF model to fully exploit spectral-spatial contextual information and preserve object boundaries in HSR images. In BLC-MRF, a pixel level MRF model is first performed and then cascaded as an input of a superpixel level MRF. In superpixel level, unary and pairwise potential terms are constructed by using the superpixel probability estimation method and spectral histogram distance, respectively. At last, a contextual MRF model is conducted and the final classification map can be computed by using α-expansion algorithm. The benefits of BLC-MRF are twofold: first, the pixel and superpixel level contextual information can be exploited under MRF framework to preserve object boundaries for improving the classification performance, and, second, the algorithm can provide promising results with a small number of training samples. Experimental results on three HSR datasets demonstrate that the proposed approach outperforms several state-of-the-art methods in terms of the classification performance.
IEEE Geoscience and Remote Sensing Letters
This letter proposes two methods for the supervised classification of multisensor optical and SAR images with possibly different spatial resolutions. Both methods are formulated within a unique framework based on hierarchical Markov random fields. Distinct quad-trees associated with the individual information sources are defined to jointly address multisensor, multiresolution, and possibly multifrequency fusion, and are integrated with finite mixture models and the marginal posterior mode criterion. Experimental validation is conducted with Pléiades, COSMO-SkyMed, RADARSAT-2, and GeoEye-1 data.
Context models for CRF-based classification of multitemporal remote sensing data
The increasing availability of multitemporal satellite remote sensing data offers new potential for land cover analysis. By combining data acquired at different epochs it is possible both to improve the classification accuracy and to analyse land cover changes at a high frequency. A simultaneous classification of images from different epochs that is also capable of detecting changes is achieved by a new classification technique based on Conditional Random Fields (CRF). CRF provide a probabilistic classification framework including local spatial and temporal context. Although context is known to improve image analysis results, so far only little research was carried out on how to model it. Taking into account context is the main benefit of CRF in comparison to many other classification methods. Context can be already considered by the choice of features and in the design of the interaction potentials that model the dependencies of interacting sites in the CRF. In this paper, these aspects are more thoroughly investigated. The impact of the applied features on the classification result as well as different models for the spatial interaction potentials are evaluated and compared to the purely label-based Markov Random Field model.
IEEE Access
The advent of modern remote sensors alongside the development of advanced parallel computing has significantly transformed both the theoretical and real implementation aspects of remote sensing. Several algorithms for detecting objects of interest in remote sensing images and subsequent classification have been devised, and these include template matching based methods, machine learning and knowledge-based methods. Knowledge-driven approaches have received much attention from the remote sensing fraternity. They do, however, face challenges in terms of sensory gap, duality of expression, vagueness and ambiguity, geographic concepts expressed in multiple modes, and semantic gap. This paper aims to review and provide an up-to-date survey on machine learning and knowledge driven approaches towards remote sensing forest image analysis. It is envisaged that this work will assist researchers in coming up with efficient models that accurately detect and classify forest images. There is a mismatch between what domain experts expect from remote sensing data and what remote sensing science produces. Such a mismatch or disparity can be reduced or alleviated by adopting an ontology paradigm methodology. Ontologies should be used to support the future of remote sensing in forest object classification. The paper is presented in five parts: (1) a review of methods used for forest image detection and classification; (2) challenges faced by object detection methods; (3) analysis of segmentation techniques employed; (4) feature extraction and classification; and (5) performance of the state-of-the-art methods employed in forest image detection and classification.
Image classification using spectral and spatial information based on MRF models
IEEE Transactions on Image Processing, 1995
A new criterion for classifying multispectral remote sensing images or textured images by using spectral and spatial information is proposed. The images are modeled with a hierarchical Markov Random Field (MRF) model that consists of the observed intensity process and the hidden class label process. The class labels are estimated according to the maximum a posterWri (MAP) criterion, but some reasonable approximations are used to reduce the computational load. A stepwise classification algorithm is derived and is confirmed by simulation and experimental results.
Improving land cover classification through contextual-based optimum-path forest
Information Sciences, 2015
Traditional machine learning algorithms very often assume statistically independent data samples. However, this is clearly not the case in remote sensing image applications, in which pixels present spatial and/or temporal dependencies. In this work, it has been presented an approach to improve land cover image classification using a contextual approach based on optimum-path forest (OPF) and the well-known Markov random fields (MRFs), hereinafter called OPF-MRF. In addition, it is also introduced a framework to the optimization of the amount of contextual information used by OPF-MRF. Experiments over high-and mediumresolution satellite (CBERS-2B, Landsat 5 TM, Ikonos-2 MS and Geoeye) and radar (ALOS-PALSAR) images covering the area of two Brazilian cities have shown the proposed approach can overcome several shortcomings related to standard OPF classification. In some cases, the proposed approach outperformed traditional OPF in about 9% of recognition rate, which is crucial for land cover classification.
On the Influence of Markovian Models for Contextual-Based Optimum-Path Forest Classification
Lecture Notes in Computer Science, 2014
Contextual classification considers the information about a sample's neighborhood to improve standard pixel-based classification approaches. In this work, we evaluated four different Markovian models for Optimum-Path Forest contextual classification considering land use recognition in remote sensing data. Some insights about the situations in which each of them should be applied are stated, as well as the idea behind them is explained.