Year
2006
Abstract
Since the availability of high-resolution satellite imagery, the use of remote sensing data has become very important for nuclear verification and safeguards purposes. Due to technical improvements regarding the spatial and spectral resolution, satellite imagery can now build the basis of complex systems for recognizing and monitoring even small-scale and short-term structural features of interest within nuclear facilities. The thereby increasing amount of high-resolution data requires new methodologies and a higher effort regarding image (pre-) processing, analysis and interpretation. Though a software system will not be able to replace an image analyst completely, there exist many possibilities to obtain faster and more precise image analysis results. These include, e.g., a pre-selection of targets or the automatic detection and classification of changes. For the detection of small structural objects in high-resolution imagery recent object-oriented procedures are more precise and meaningful than the traditional pixel-based ones. In comparison to the purely spectral features analyzed within the pixel-based approaches, the utilization of object features, e.g., the size or orientation of the objects, their shape or texture and the relations between the objects in different scales, extends the image analysis possibilities. Analyzing satellite image data in an object-oriented way also offers the possibility to involve specific knowledge in the classification or recognition process. This paper proposes an automated object-oriented analysis approach for high resolution remote sensing data. Regarding the increased demands new methodologies will be presented for (pre-) processing, analysis, change detection and interpretation as well as their application to case studies on Iranian nuclear facilities. Against the background of nuclear safeguards-related monitoring purposes aspects of automation, standardisation and transferability will be discussed.