The images range in scale from hundreds of square miles for monitoring large areas of land and sea to high resolution images of individual objects.
Standardized image analysis and information extraction are essential to compare results across geographic areas and support political directives like the Kyoto Protocol, sustainable management of natural resources and future-proof infrastructure planning. Similarly, in natural disasters and other emergencies, image analysis can rapidly provide maps of remote territory to assist emergency operations, save lives, reduce costs and support insurance and reinsurance services.
Attempts to automate image analysis and information extraction have been continuing for decades. But despite increases in computational power and imaging capabilities, fundamental advances in automated image analysis have been limited. Looking at a satellite image, it is easy, even for an untrained eye, to distinguish a river from a lake. But traditional automated image classification tools often fail to recognize features that are obvious to any person because they only use information contained within individual pixels and their immediate neighbors.
From the perspective of an image analyst, there is so much more to an image than the color and intensity of pixels. There is shape, texture, area, scale and the relationships of the objects, all attributes which analysts use to understand the content of an image and data set.
Image analysis in Infrastructure Planning
Accurate mapping and change detection are essential tools in the sustainable planning of urban infrastructure and traffic corridors. They are vital in reducing the cost of constructing and operating new oil and gas pipelines or telecoms infrastructure. They support urban planning should be supported by helping to ensure that land is used appropriately and that quality of life is enhanced by reducing noise and pollution.
Image analysis in Natural Resource Management
An accurate and up-to-date picture of land cover and of the surface of the ocean is vital for assessing the health of the planet. It is used to measure the impact of climate change, assess biodiversity in natural habitats, and support sustainable farming and forest management. It is also invaluable in finding and exploiting natural resources like oil and gas fields.
Image analysis in Security and Emergency Response
Rapid mapping and data fusion are necessary to create detailed maps of any territory. Object detection needs to be linked with feature extraction in order to understand fully the situation on the ground. For example, in order to assess damage after disasters and monitor reconstruction, change detection is an essential tool. Similarly, monitoring and identifying changes over time supports geo-professionals in making timely and well-informed decisions.
Definiens enables organizations involved in Earth Sciences to extract accurate and reliable geo-information rapidly, from any kind of remote sensing imagery. Enterprises and organizations all along the value chain – data providers, service providers and end users – rely on our software to bridge the gap between Earth Observation and remote sensing on the one hand and Geographical Information Systems (GIS) on the other. By correlating image analysis with other geospatial information, Earth Sciences organizations can generate, with minimal manual intervention, accurate GIS-ready information.