Introduction

The current Lloyd and Bleek collection consists of notebooks which contain stories in the |xam and !kun Bushman languages. In some of these notebooks there are English translations along-side the Bushman words but in many cases those translations do not exist or are unclear. Given that the BOLD Dictionary contains images which have both an English word as well as its |xam translation it is natural to assume that people may want to make use of the BOLD dictionary to translate words which appear in the Lloyd and Bleek collection. The problem exists, however, that it is not practical to do this by hand simply due to the size of the BOLD dictionary. A content based image retrieval (CBIR) system was built to try and overcome this impractibility by allowing users to select a word in one of the notebooks as a search key and have words with a similar appearance returned. A CBIR system is a system which searches for images which are similar to another image which is known as the search key.

 

The goal of the BOLD Translator is to match words in the notebooks to their corresponding dictionary entries

 

Motivation

The motivation for the BOLD Translator is to make the translation of Wilhelm Bleek’s notebooks a practical undertaking. The BOLD Translator has the potential to be of assistance to researchers who are studying Bushman language and culture.

 

Aim of Research

The research question posed is:

Can image-based searching be done accurately and efficiently?

 

The BOLD Translator

by Kyle Williams

 

 

Menu

 

Introduction

 

Design and Implementation

 

Evaluation

 

Conclusion

 

Future Work