Publication: Accelerating of visual information for visually impared people
Loading...
Official URL
Full text at PDC
Publication Date
2014
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
La deficiencia visual es una enfermedad común del ojo que puede ocurre a cualquier edad. La pérdida visual viene determinada por un funcionamiento incorrecto del sistema visual. En la actualidad existen aparatos externos que mejoran la percepción visual lo que proporciona al
paciente una mejor calidad de vida. Este tipo de sistemas externos se basan en un receptor que junto a un software permite al paciente incrementar su percepción pese a las enfermedades de baja visión.
Este trabajo considera el uso de los aceleradores hardware hoy en día extendidos en los sistemas de alta computación en el campo de bajo visión. La mayor parte de los sistemas de ayuda a la visión se basan en algoritmos de tratamiento de imágenes para detectar bordes,
realce parcial de imágenes, ect; que conlleva una carga computacional elevada. El objetivo de este trabajo es llevar a cabo un estudio de vialidad para determinar si los aceleradores hardware en su versión portable son buenos candidatos a ser usados en estos dispositivos.
Aunque los sistemas empleados para este estudio pertenecen al campo de sistemas de altas prestaciones que difícilmente pueden ser empleados en sistemas con requerimientos de bajo consumo, su utilización nos permite extrapolar las aceleraciones que pueden obtenerse si se utilicen sus versiones para sistemas empotrados. En la actualidad existen System-on-Chip que incorporan aceleradores hardware con la misma arquitectura que los del estudio por lo que las conclusiones del mismo son igualmente válidas. Además, en este trabajo se proporcionaría la comprensión general de baja visión y una visión general de los enfoques de mejora de visión
baja que podrían ser aún investigados en el futuro.
Visual impairment is a common eye disease that can occurs at any age. There are five main impairments but also a lot of different subtypes of impairments. Visual impairment could happen because some kind of ill in the eye, therefore retina is damaged and eye can’t function properly. In the most cases, those impairments can be enhanced with some medical aids and nowadays scientists trying to develop medical software’s that will help even better.Technology is growing fast and it is possible to develop enhancement software that can test some low vision diseases. Software’s can show how visual impaired people see which will be shown with foveal algorithm and also tested. With that software it is understandable how to design enhancement software’s to enhance vision of visual impaired people. In this master thesis is developed edge detector similar to Canny edge detector which will filter image to binary image with edges. This thesis will also give a convolution algorithm which is primitive operation for image processing. The aim of this work is to test algorithms for image enhancement with different values in the algorithm and to test those algorithms on the different accelerators that could be used for medical aids. The results of this research can show advantages and disadvantages of equipment that is used and also can indicate how to approach in the future work. This paper would provide general understanding of low vision and an overview of low vision enhancement approaches that could be still researched in the future.
Visual impairment is a common eye disease that can occurs at any age. There are five main impairments but also a lot of different subtypes of impairments. Visual impairment could happen because some kind of ill in the eye, therefore retina is damaged and eye can’t function properly. In the most cases, those impairments can be enhanced with some medical aids and nowadays scientists trying to develop medical software’s that will help even better.Technology is growing fast and it is possible to develop enhancement software that can test some low vision diseases. Software’s can show how visual impaired people see which will be shown with foveal algorithm and also tested. With that software it is understandable how to design enhancement software’s to enhance vision of visual impaired people. In this master thesis is developed edge detector similar to Canny edge detector which will filter image to binary image with edges. This thesis will also give a convolution algorithm which is primitive operation for image processing. The aim of this work is to test algorithms for image enhancement with different values in the algorithm and to test those algorithms on the different accelerators that could be used for medical aids. The results of this research can show advantages and disadvantages of equipment that is used and also can indicate how to approach in the future work. This paper would provide general understanding of low vision and an overview of low vision enhancement approaches that could be still researched in the future.
Description
Máster en Investigación en Informática, Facultad de Informática, Departamento de Arquitectura de Computadores y Automática, curso 2013-2014