Malaysia is famous for its variety of cuisines. However, it is hard for tourists to enjoy the food because of a lack of information about it. This study aims to develop a system to help tourists to identify Malaysian cuisine using computer vision. In this study, we found that other country cuisines with a similar look to Malaysian cuisines cause recognition failure. This paper shows how to create training datasets to reduce such loss. Our experimental results show that the system can adequately identify Malaysian cuisines from other country's similar-looking cuisines.