Resumen:
|
[EN] Assessing and monitoring river areas is necessary to support river management
and restoration processes. As rivers change over short periods of time
assessment methods need to be quick, reliable and cost/time-effective. ...[+]
[EN] Assessing and monitoring river areas is necessary to support river management
and restoration processes. As rivers change over short periods of time
assessment methods need to be quick, reliable and cost/time-effective. Fixed
point photography has potential, but some challenges remain. A methodology
has been developed in order to obtain useful information from sets of
multitemporal images. First, the images are registered to a common coordinate
system, both automatic and manual methodologies were developed. Second,
vegetation and water features were extracted. Vegetation coverage can be
obtained using a combination of Green Leaf Algorithm (GLA) and texture
characteristics. Water presence-absence is derived from the images using a
combination of different clues, such as hue, saturation, brightness, texture and
ground detection techniques. The extracted features allow the evaluation of
changes over time. However, the accuracy of the results is limited by the
differences in image conditions; such as illumination, orientation, and point of
view location. Another important limitation is water correct segmentation, which
remains an important challenge in RGB images. Therefore, some improvements
to current photo-based monitoring programs are proposed, such as the use of
stereo or NIR images, or measures to allow the photographs to be taken from
the same exact location every time.
[-]
[ES] El proyecto busca investigar cómo podría aplicarse la fotografía de punto fijo para la creación de una línea temporal que muestre los cambios en los ríos desde el punto de vista ecológico y geomorfológico. Se deberán ...[+]
[ES] El proyecto busca investigar cómo podría aplicarse la fotografía de punto fijo para la creación de una línea temporal que muestre los cambios en los ríos desde el punto de vista ecológico y geomorfológico. Se deberán identificar las variables a extraer de las fotografías mediante la revisión de la literatura reciente (ej. Cobertura vegetal, tipo de vegetación, anchura del cauce, sedimento, profundidad del cauce) y desarrollar una metodología para ello de forma tan automatizada como sea posible. El proyecto se realizará en colaboración con la Agencia Medioambiental de Reino Unido (Environment Agency) y se utilizarán las fotografías del programa Springs to Source . Estas imágenes son tomadas mensualmente y pueden complementarse con información de otras fuentes, como estaciones de aforo, si se considera necesario.
[-]
|