2013 | OriginalPaper | Buchkapitel
A Simple Standalone Sign Based Recognition Translator without Using Super Computer Processing
verfasst von : Sandy Siu Ying Ko, Wei Lun Ng, Chee Kyun Ng, Nor Kamariah Noordin
Erschienen in: Advances in Visual Informatics
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
In this paper, the design and development of a simple standalone sign based recognition translator without the use of supercomputer processing is presented. The main components for digital signal processing (DSP) and data storage in this translator system are peripheral interface controller (PIC) microcontroller and secure digital (SD) card respectively. The PIC 18F4620 is chosen as system’s DSP because it is able to interface with the SD card. The input of sign gestures is being recognized and translated into text message such that people can understand its meanings. A flex sensor is attached to each finger in a data glove which has the capability to measure the finger sign. The five fingers per hand can create many different signs that can be translated into different text messages. The attached PIC 18F4620 microcontroller in the data glove receives the data associated with a given sign. These data are then diagnosed and translated by matching with the predefined data in the SD card data storage device by using deterministic matching model. When the data is matched, the matched predefined data will be sent to the microcontroller to display its text message on an on-board LCD module. The developed standalone translating system is flexible and adaptive as it can be reprogrammable according to the users that have special disabilities without any aid from a supercomputer.