Geographic Object Based Image Analysis

The primary goal of collecting Earth observation (EO) imagery is to map, analyze, and contribute to an understanding of the status and dynamics of geographic phenomena. In geographic information science (GIScience), the term object-based image analysis (OBIA) was tentatively introduced in 2006. When it was re-formulated in 2008 as geographic object-based image analysis (GEOBIA), the primary focus was on integrating multiscale EO data with GIScience and computer vision (CV) solutions to cope with the increasing spatial and temporal resolution of EO imagery. Building on recent trends in the context of big EO data analytics as well as major achievements in CV, the objective of this article is to review the role of spatial concepts in the understanding of image objects as the primary analytical units in semantic EO image analysis, and to identify opportunities where GEOBIA may support multi-source remote sensing analysis in the era of big EO data analytics. We (re-)emphasize the spatial paradigm as a key requisite for an image understanding system capable to deal with and exploit the massive data streams we are currently facing; a system which encompasses a combined physical and statistical model-based inference engine, a well-structured CV system design based on a convergence of spatial and colour evidence, semantic content-based image retrieval capacities, and the full integration of spatio-temporal aspects of the studied geographical phenomena.

Aerial photography is an important source of high resolution remotely sensed data. Before 1970, aerial photographs were the only remote sensing data source for land use and land cover classification. Using these old aerial photographs improve the final output of land use and land cover change detection. However, classic techniques of aerial photographs classification like manual interpretation or screen digitization require great experience, long processing time and vast effort. A new technique needs to be developed in order to reduce processing time and effort and to give better results. Geographic object based image analysis (GEOBIA) is a newly developed area of Geographic In-formation Science and remote sensing in which automatic segmentation of images into objects of similar spectral, temporal and spatial characteristics is undertaken. Unlike pixel-based technique, GEOBIA deals with the object properties such as texture, square fit, roundness and many other properties that can improve classification results. GEOBIA technique can be divided into two main steps; segmentation and classification. Segmentation process is grouping adjacent pixels into objects of similar spectral and spatial characteristics. Classification process is assigning classes to the generated objects based on the characteristics of the individual objects. This study aimed to use GEOBIA technique to develop a novel approach for land use and land cover classification of aerial photographs that saves time and effort and gives improved results. Aerial photographs from 1954 of Valle Telesina in Italy were used in this study. Images were rectified and georeferenced in Arcmap using topographic maps. Images were then processed in eCognition software to generate land use and land cover map of 1954. A decision tree rule set was developed in eCognition to classify images and finally nine classes of general land use and land cover in the study area were recognized (forest, trees stripes, agricultural crops fields, olive tree fields, vineyards, mixed vineyards and olive trees, roads, rivers and urban areas). Results showed that in the segmentation process, grater values of scale, shape and compactness parameters are more suitable for classifying forest and river objects. While lower values of scale, shape and compactness parameters are more suitable for classifying objects of agricultural crops fields, olive tree fields, vineyards, mixed vineyards and olive trees, roads and urban areas. Applying map accuracy assessment test revealed high accuracy of the generated map of approximately 87%. Forest and river objects were recognized through the main spectral signal value of the object. Tree stripes were recognized through the object’s main spectral signal value, shape and size. Many different features were used for classifying the other classes’ objects. Among others, features like object’s density, texture, relative border to other classes, width, contrast and standard deviation were used in the classification process. The results showed that after developing the rule set process tree, classifying an aerial photograph can be faster by more than thirty times the classical techniques based on the computer machine speed. Generally, GEOBIA is a promising technique for classifying and interpreting gray scale aerial photographs.

Within the 2015 IEEE GRSS Data Fusion Contest extremely high resolution LiDAR point cloud must be fused with multi-spectral aerial imagery. We propose an innovative geospatial 2D and 3D object-based classification system capable of counting two populations of ISO-containers (estimated based on their standard size) located in the harbor area depicted by the two test datasets. The degree of novelty of the proposed classification system is twofold. First, it combines inductive (bottom-up, data-driven) and deductive (top-down, prior rule-based) inference mechanisms, where the latter initializes the former in a hybrid inference framework. Second, it is provided with feedback loops, which increase its robustness to changes in input data and augment its degree of automation. The geospatial outcome are tangible vector objects, which allow estimation of statistics per container together with a detailed reconstruction of the 3D scene in a GIS system.

Within the 2015 IEEE GRSS Data Fusion Contest, an extremely high-resolution 3D LiDAR point cloud of a harbor test site must be “fused” with a 2D multi-spectral aerial image, featuring no radiometric calibration metadata file, of the same surface area. In this scenario we propose an innovative geospatial 2D and 3D object-based classification system, capable of counting instances of two populations of ISO-containers, whose standard dimensions are known a priori based on the ISO 668 - Series 1 freight containers documentation, detected in the 2D and 3D datasets at hand. The degree of novelty of the proposed classification system is twofold. First, it combines inductive (bottom-up, data-driven) and deductive (top-down, prior knowledge-based) inference mechanisms, where the latter initializes the former in a hybrid inference framework. Second, it is provided with feedback loops, which increase its robustness to changes in input data and augment its degree of automation. The geospatial outcome consists of tangible vector objects, which allow estimation of statistics per container together with a detailed reconstruction of the 3D scene in a geographic information system.