Proceedings of the 1st International Conference on Informatics, Engineering, Science and Technology, INCITEST 2019, 18 July 2019, Bandung, Indonesia

Research Article

Evaluation of VGG Networks for Semantic Image Segmentation of Malaysian Meals

Download435 downloads
  • @INPROCEEDINGS{10.4108/eai.18-7-2019.2287943,
        author={N  Jamil and N AN N Redzuan and M F Ismail and W AW Ramli},
        title={Evaluation of VGG Networks for Semantic Image Segmentation of Malaysian Meals},
        proceedings={Proceedings of the 1st International Conference on Informatics, Engineering, Science and Technology, INCITEST 2019, 18 July 2019, Bandung, Indonesia},
        publisher={EAI},
        proceedings_a={INCITEST},
        year={2019},
        month={10},
        keywords={food images malaysian meals},
        doi={10.4108/eai.18-7-2019.2287943}
    }
    
  • N Jamil
    N AN N Redzuan
    M F Ismail
    W AW Ramli
    Year: 2019
    Evaluation of VGG Networks for Semantic Image Segmentation of Malaysian Meals
    INCITEST
    EAI
    DOI: 10.4108/eai.18-7-2019.2287943
N Jamil1,*, N AN N Redzuan1, M F Ismail1, W AW Ramli1
  • 1: Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA
*Contact email: njamil@gmail.com

Abstract

This paper evaluates VGG-16 and VGG-19 networks in performing semantic image segmentation of Malaysian meals. This is a preliminary investigation of using transfer learning models to recognize food objects in typical Malaysian meals. Most current works of food recognition system calculate the calories and nutritional content of a meal based on the food object recognition, regardless of the portion size. Our final aim is to develop a food recognition system that considers the portion size in calculating the calories and nutritional content. Therefore, semantic segmentation of the food objects in the meal is a very important stage. Our work also initiated the training datasets for Malaysian meals that will be made available to the public. Using a small training dataset and a basic configuration of the VGG network, our results show inconsistent findings of the performance of VGG-16 and VGG-19. These findings will serve as a fundamental guideline to improve the semantic segmentation of food images.