Geographic Object-Based Image Analysis – Towards a new paradigm

The amount of scientific literature on (Geographic) Object-based Image Analysis – GEOBIA has been and still is sharply increasing. These approaches to analysing imagery have antecedents in earlier research on image segmentation and use GIS-like spatial analysis within classification and feature extraction approaches. This article investigates these development and its implications and asks whether or not this is a new paradigm in remote sensing and Geographic Information Science (GIScience). We first discuss several limitations of prevailing per-pixel methods when applied to high resolution images. Then we explore the paradigm concept developed by Kuhn (1962) and discuss whether GEOBIA can be regarded as a paradigm according to this definition. We crystallize core concepts of GEOBIA, including the role of objects, of ontologies and the multiplicity of scales and we discuss how these conceptual developments support important methods in remote sensing such as change detection and accuracy assessment. The ramifications of the different theoretical foundations between the ‘per-pixel paradigm’ and GEOBIA are analysed, as are some of the challenges along this path from pixels, to objects, to geo-intelligence. Based on several paradigm indications as defined by Kuhn and based on an analysis of peer-reviewed scientific literature we conclude that GEOBIA is a new and evolving paradigm.     

Validating the use of object-based image analysis to map commonly recognized landform features in the United States

ABSTRACT

The U.S. Geological Survey (USGS) National Geospatial Program (NGP) seeks to i) create semantically accessible terrain features from the pixel-based 3D Elevation Program (3DEP) data, and ii) enhance the usability of the USGS Geographic Names Information System (GNIS) by associating boundaries with GNIS features whose spatial representation is currently limited to 2D point locations. Geographic object-based image analysis (GEOBIA) was determined to be a promising method to approach both goals. An existing GEOBIA workflow was modified and the resulting segmented objects and terrain categories tested for a strategically chosen physiographic province in the mid-western US, the Ozark Plateaus. The chi-squared test of independence confirmed that there is significant overall spatial association between terrain categories of the GEOBIA and GNIS feature classes. Contingency table analysis also suggests strong category-specific associations between select GNIS and GEOBIA classes. However, 3D visual analysis revealed that GEOBIA objects resembled segmented regions more than they did individual landform objects, with their boundaries often failing to correspond to match what people would likely perceive as landforms. Still, objects derived through GEOBIA can provide initial baseline landscape divisions that can improve the efficiency of more specialized feature extraction methods.     

A framework of region-based spatial relations for non-overlapping features and its application in object based image analysis

Object based image analysis (OBIA) is an approach increasingly used in classifying high spatial resolution remote sensing images. Object based image classifiers first segment an image into objects (or image segments), and then classify these objects based on their attributes and spatial relations. Numerous algorithms exist for the first step of the OBIA process, i.e. image segmentation. However, less research has been conducted on the object classification part of OBIA, in particular the spatial relations between objects that are commonly used to construct rules for classifying image objects and refining classification results. In this paper, we establish a context where objects are areal (not points or lines) and non-overlapping (we call this “single-valued” space), and propose a framework of binary spatial relations between segmented objects to aid in object classification. In this framework, scale-dependent “line-like objects” and “point-like objects” are identified from areal objects based on their shapes. Generally, disjoint and meet are the only two possible topological relations between two non-overlapping areal objects. However, a number of quasi- topological relations can be defined when the shapes of the objects involved are considered. Some of these relations are fuzzy and thus quantitatively defined. In addition, we define the concepts of line-like objects (e.g. roads) and point-like objects (e.g. wells), and develop the relations between two line-like objects or two point-like objects. For completeness, cardinal direction relations and distance relations are also introduced in the proposed context. Finally, we implement the framework to extract roads and moving vehicles from an aerial photo. The promising results suggest that our methods can be a valuable tool in defining rules for object based image analysis.     

An intelligent geospatial processing unit for image classification based on geographic vector agents (GVAs)

Spatial modeling methods usually use pixels and image objects as fundamental processing units to address real‐world objects, geo‐objects, in image space. To do this, both pixel‐based and object‐based approaches typically employ a linear two‐staged workflow of segmentation and classification. Pixel‐based methods segment a classified image to address geo‐objects in image space. In contrast, object‐based approaches classify a segmented image to identify geo‐objects from raster datasets. These methods lack the ability to simultaneously integrate the geometry and theme of geo‐objects in image space. This article explores Geographical Vector Agents (GVAs) as an automated and intelligent processing unit to directly address real‐world objects in the process of remote sensing image classification. The GVA is a distinct type of geographic automata characterized by elastic geometry, dynamic internal structure, neighborhoods and their respective rules. We test this concept by modeling a set of objects on a subset IKONOS image and LiDAR DSM datasets without the setting parameters (e.g. scale, shape information), usually applied in conventional Geographic Object‐Based Image Analysis (GEOBIA) approaches. The results show that the GVA approach achieves more than 3.5% improvement for correctness, 2% improvement for quality, although no significant improvement for completeness to GEOBIA, thus demonstrating the competitive performance of GVAs classification.     

高分辨率影像城市绿地快速提取技术与应用

高分辨率遥感影像是城市绿地信息快速提取的主要数据源 ,文中以多尺度影像分割与面向对象影像分析方法为主要技术 ,利用样本多边形对象的成员函数建立训练区 ,自动提取大庆市城市绿地覆盖信息 ,达到清查城市绿地的目的。该方法信息获取周期短、精度高、成本低 ,实现了城市绿地信息精确获取与快速更新。     

Change of spatial information under rescaling: A case study using multi-resolution image series

Spatial structure in imagery depends on a complicated interaction between the observational regime and the types and arrangements of entities within the scene that the image portrays. Although block averaging of pixels has commonly been used to simulate coarser resolution imagery, relatively little attention has been focused on the effects of simple rescaling on spatial structure and the explanation and a possible solution to the problem. Yet, if there are significant differences in spatial variance between rescaled and observed images, it may affect the reliability of retrieved biogeophysical quantities. To investigate these issues, a nested series of high spatial resolution digital imagery was collected at a research site in eastern Nebraska in 2001. An airborne Kodak DCS420IR camera acquired imagery at three altitudes, yielding nominal spatial resolutions ranging from 0.187 m to 1 m. The red and near infrared (NIR) bands of the co-registered image series were normalized using pseudo-invariant features, and the normalized difference vegetation index (NDVI) was calculated. Plots of grain sorghum planted in orthogonal crop row orientations were extracted from the image series. The finest spatial resolution data were then rescaled by averaging blocks of pixels to produce a rescaled image series that closely matched the spatial resolution of the observed image series. Spatial structures of the observed and rescaled image series were characterized using semivariogram analysis. Results for NDVI and its component bands show, as expected, that decreasing spatial resolution leads to decreasing spatial variability and increasing spatial dependence. However, compared to the observed data, the rescaled images contain more persistent spatial structure that exhibits limited variation in both spatial dependence and spatial heterogeneity. Rescaling via simple block averaging fails to consider the effect of scene object shape and extent on spatial information. As the features portrayed by pixels are equally weighted regardless of the shape and extent of the underlying scene objects, the rescaled image retains more of the original spatial information than would occur through direct observation at a coarser sensor spatial resolution. In contrast, for the observed images, due to the effect of the modulation transfer function (MTF) of the imaging system, high frequency features like edges are blurred or lost as the pixel size increases, resulting in greater variation in spatial structure. Successive applications of a low-pass spatial convolution filter are shown to mimic a MTF. Accordingly, it is recommended that such a procedure be applied prior to rescaling by simple block averaging, if insufficient image metadata exist to replicate the net MTF of the imaging system, as might be expected in land cover change analysis studies using historical imagery.     

A GEOBIA framework for the implementation of national and international forest definitions using very high spatial resolution optical satellite data

In recent decades, there is an increasing need for harmonised and accurate information on the status and extent of forests. However, delineating the extent of forest areas is a complex task, since the existence of more than 100 definitions of forest worldwide causes considerable discrepancies in forested area estimates. The aim of this work was to examine the potential of geographic object based image analysis (GEOBIA) and very high spatial resolution imagery to discriminate forest areas following two different definitions of forest in northern Greece. In particular, we examined the definition of forest under the Greek law as well as the United Nations Food and Agricultural Organisation definition. Our findings suggest that the developed GEOBIA approach not only performed remarkably well for the discrimination of forest areas but also allowed to estimate rapidly and reliably forest extents when the two aforementioned forest definitions were employed.     

融合像素—多尺度区域特征的高分辨率遥感影像分类算法

针对基于像素多特征的高分辨率遥感影像分类算法的"胡椒盐"现象和面向对象影像分析方法的"平滑地物细节"现象,提出了一种融合像素特征和多尺度区域特征的高分辨率遥感影像分类算法。(1)首先采用均值漂移算法对原始影像进行初始过分割,然后对初始过分割结果进行多尺度的区域合并,形成多尺度分割结果。根据多尺度区域合并RMI指数变化和分割尺度对分类精度的影响,确定最优分割尺度。(2)融合光谱特征、像元形状指数PSI(Pixel Shape Index)、初始尺度和最优尺度区域特征,并对多类型特征进行归一化,最后结合支持向量机(SVM)进行分类。实验结果表明该算法既能有效减少基于像素多特征的高分辨率遥感影像分类算法的"胡椒盐"现象,又能保持地物对象的完整性和地物细节信息,提高易混淆类别(如阴影和街道,裸地和草地)的分类精度。     

Selection of Optimal Object Features in Object-Based Image Analysis Using Filter-Based Algorithms

With the increase in spatial resolution of recent sensors, object-based image analysis (OBIA) has gained importance for producing detailed land use maps. One of the main advantages of OBIA is that a variety of spectral, spatial and textural features can be extracted for the segmented image objects that are later utilized in classification. However, using a large number of features not only increases the required computational time, but also requires a large number of ground samples, which is unavailable in most cases. For these reasons, feature selection (FS) has become an important research topic for OBIA based classification studies. In this study, three filter-based FS algorithms namely, Chi square, information gain and ReliefF were applied to determine the most effective object features that ensure high separability among landscape features. For this purpose, importance degree (i.e. ranks) of 110 input object features were firstly estimated by the algorithms, and correlation-based merit function was then applied to determine optimum feature subset size. Multi-resolution segmentation algorithm was applied for segmenting a WorldView-2 image. Support vector machine, random forest and nearest neighbour classifiers were all utilized to classify segmented image objects using the selected object features. Results revealed that the FS algorithms were effective for selecting the most relevant features. Also, the classifiers produced the highest performances with 24 out of 110 features selected by the information gain (IG) algorithm. Particularly, the support vector machine classifier produced the highest overall accuracy (92.00%) with 24 selected features determined by the IG algorithm. A significant improvement of about 4% was achieved by applying FS procedures that was found statistically significant in terms of Wilcoxon signed-ranks test.     

Evaluation of data fusion and image segmentation in earth observation based rapid mapping workflows

This paper is an exploratory study, which aimed to discover the synergies of data fusion and image segmentation in the context of EO-based rapid mapping workflows. Our approach pillared on the geographic object-based image analysis (GEOBIA) focusing on multiscale, internally-displaced persons’ (IDP) camp information extraction from very high spatial resolution (VHSR) images. We applied twelve pansharpening algorithms to two subsets of a GeoEye-1 image scene that was taken over a former war-induced ephemeral settlement in Sri Lanka. A multidimensional assessment was employed to benchmark pansharpening algorithms with respect to their spectral and spatial fidelity. The multiresolution segmentation (MRS) algorithm of the eCognition Developer software served as the key algorithm in the segmentation process. The first study site was used for comparing segmentation results produced from the twelve fused products at a series of scale, shape, and compactness settings of the MRS algorithm. The segmentation quality and optimum parameter settings of the MRS algorithm were estimated by using empirical discrepancy measures. Non-parametric statistical tests were used to compare the quality of image object candidates, which were derived from the twelve pansharpened products. A wall-to-wall classification was performed based on a support vector machine (SVM) classifier to classify image objects candidates of the fused images. The second site simulated a more realistic crisis information extraction scenario where the domain expertise is crucial in segmentation and classification. We compared segmentation and classification results of the original images (non-fused) and twelve fused images to understand the efficacy of data fusion. We have shown that the GEOBIA has the ability to create meaningful image objects during the segmentation process by compensating the fused image’s spectral distortions with the high-frequency information content that has been injected during fusion. Our findings further questioned the necessity of the data fusion step in rapid mapping context. Bypassing time-intensive data fusion helps to actuate EO-based rapid mapping workflows. We, however, emphasize the fact that data fusion is not limited to VHSR image data but expands over many different combinations of multi-date, multi-sensor EO-data. Thus, further research is needed to understand the synergies of data fusion and image segmentation with respect to multi-date, multi-sensor fusion scenarios and extrapolate our findings to other remote sensing application domains beyond EO-based crisis information retrieval.     

Area-based and location-based validation of classified image objects

Geographic object-based image analysis (GEOBIA) produces results that have both thematic and geometric properties. Classified objects not only belong to particular classes but also have spatial properties such as location and shape. Therefore, any accuracy assessment where quantification of area is required must (but often does not) take into account both thematic and geometric properties of the classified objects. By using location-based and area-based measures to compare classified objects to corresponding reference objects, accuracy information for both thematic and geometric assessment is available. Our methods provide location-based and area-based measures with application to both a single-class feature detection and a multi-class object-based land cover analysis. In each case the classification was compared to a GIS layer of associated reference data using randomly selected sample areas. Error is able to be pin-pointed spatially on per-object, per class and per-sample area bases although there is no indication whether the errors exist in the classification product or the reference data. This work showcases the utility of the methods for assessing the accuracy of GEOBIA derived classifications provided the reference data is accurate and of comparable scale.     

基于相似度验证的自动变化探测研究

变化检测技术越来越多地应用于城市遥感分析和应用领域,但目前城市变化检测的研究主要基于中低空间分辨率的遥感数据,使用的方法也主要是像元直接比较法或者是分类后比较法。提出一种基于变化向量分析和相似度验证相结合的变化检测方法,应用高空间分辨率影像来快速实现城市建筑物、街道等目标的自动变化检测。并详细阐述了变化目标的提取以及验证的方法和过程,其结果真实地反映了地面目标的实际变化程度和类型。     

Automatic fuzzy object-based analysis of VHSR images for urban objects extraction

We present an automatic approach for object extraction from very high spatial resolution (VHSR) satellite images based on Object-Based Image Analysis (OBIA). The proposed solution requires no input data other than the studied image. Not input parameters are required. First, an automatic non-parametric cooperative segmentation technique is applied to create object primitives. A fuzzy rule base is developed based on the human knowledge used for image interpretation. The rules integrate spectral, textural, geometric and contextual object proprieties. The classes of interest are: tree, lawn, bare soil and water for natural classes; building, road, parking lot for man made classes. The fuzzy logic is integrated in our approach in order to manage the complexity of the studied subject, to reason with imprecise knowledge and to give information on the precision and certainty of the extracted objects. The proposed approach was applied to extracts of Ikonos images of Sherbrooke city (Canada). An overall total extraction accuracy of 80% was observed. The correctness rates obtained for building, road and parking lot classes are of 81%, 75% and 60%, respectively.     

RS与GIS集成环境下的遥感图像分析技术

通过多源数据融合，将目前GIS中包含的地理信息作为辅助数据，引入到RS图像分析过程中。GIS数据在空间信息、目标属性和目标范围等方面对RS图像分析具有重要的引导作用。     

The 13th Asian conference on remote sensing,Mongolia

Abstract

Three spatial resolutions of airborne remote sensing imagery (60 cm, 1 m, and 2 m) collected over multi‐layer aspen, pine, spruce, and mixedwood forest stands in Alberta on July 18^th, 1998 were tested for their ability to provide a statistical stand discrimination based on spatial co‐occurrence texture analysis. As spatial resolution increased, classification accuracies increased. The highest classification accuracy of 86.7% was obtained using the highest image spatial resolution data (60 cm), with spatial co‐occurrence texture and spectral signatures combined, and a thirteen‐class multi‐layer stand stratification. The texture of the highest spatial resolution imagery (60 cm pixel resolution) was interpreted to contain information on the crown architecture of individual trees. In larger windows, the texture was interpreted to contain information on stand structure. Texture of lower spatial resolution imagery (1 m and 2 m pixel resolution) could not detect individual tree crown architecture and was determined to be related primarily to stand structure characteristics. The use of texture channels improved the per‐plot classification accuracies by 15.7%, compared to the use of the spectral data alone.     

Database structure for the integration of RS with GIS based on semantic network

The integration of remote sensing (RS) with geographical information system (GIS) is a hotspot in geographical information science. A good database structure is important to the integration of RS with GIS, which should be beneficial to the complete integration of RS with GIS, able to deal with the disagreement between the resolution of remote sensing images and the precision of GIS data, and also helpful to the knowledge discovery and exploitation. In this paper, the database structure storing the spatial data based on semantic network is presented. This database structure has several advantages. Firstly, the spatial data is stored as raster data with space index, so the image processing can be done directly on the GIS data that is stored hierarchically according to the distinguishing precision. Secondly, the simple objects are aggregated into complex ones. Thirdly, because we use the indexing tree to depict the relationship of aggregation and the indexing pictures expressed by 2-D strings to describe the topology structure of the objects, the concepts of surrounding and region are expressed clearly and the semantic content of the landscape can be illustrated well. All the factors that affect the recognition of the objects are depicted in the factor space, which provides a uniform mathematical frame for the fusion of the semantic and non-semantic information. Lastly, the object node, knowledge node and the indexing node are integrated into one node. This feature enhances the ability of system in knowledge expressing, intelligent inference and association. The application shows that this database structure can benefit the interpretation of remote sensing image with the information of GIS.     

Utilizing the Potential of World View −2 for Discriminating Urban and Vegetation Features Using Object Based Classification Techniques

With the availability of very high resolution multispectral imagery, it is possible to identify small features in urban environment. Because of the multiscale feature and diverse composition of land cover types found within the urban environment, the production of accurate urban land cover maps from high resolution satellite imagery is a difficult task. This paper demonstrates the potential of 8 bands capability of World View 2 satellite for better automated feature extraction and discrimination studies. Multiresolution segmentation and object based classification techniques were then applied for discrimination of urban and vegetation features in a part of Dehradun, Uttarakhand, India. The study demonstrates that scale, colour, shape, compactness and smoothness have a significant influence on the quality of image objects achieved, which in turn governs the classified result. The object oriented analysis is a valid approach for analyzing high spatial and spectral resolution images. World View 2 imagery with its rich spatial and spectral information content has very high potential for discrimination of the less varied varieties of vegetation.     

Image Analysis for GIS Data Acquisition

The automatic interpretation of aerial and satellite imagery by means of image analysis is currently one of the major research tasks in photogrammetry and related disciplines. The primary goal is the automatic extraction of visible, spatial topographic objects from imagery. The extracted objects represent one possible form of input for creating geographic databases. In this paper, different aspects of image analysis are discussed and a framework is provided for scene interpretation, which is based on the integration of image analysis and a GIS data model. Two examples concerned with the combined extraction of roads and trees, and with the multitemporal interpretation and monitoring of moorland, are given to illustrate the research.     

图像匹配的物体空间关系推理表达

物体空间关系指的是物体在欧氏空间中的邻近关系,根据图像中包含物体的邻近关系解决图像匹配的问题。本文首先基于对比机制训练物体块特征提取网络,构建物体块深度特征,该特征可以有效匹配不同图像中的相同物体块;其次,基于已有的先验图像数据推理表达图像中物体的空间邻近关系,构建场景物体空间邻近图;进而基于该空间邻近图计算场景图像对的空间邻近度,完成图像空间关系匹配。试验表明不匹配图像间的空间邻近度一般为0,而匹配图像间的空间邻近度一般大于0,本文空间关系匹配涉及多个物体间的相互关系,具有更强的稳健性,其匹配效果明显优于对比试验中的其他方法,可以高效稳定地完成图像匹配任务。