Automatic Cooking Of Porridge Based On Human Visual Perception

Yujuan Sun, Junyu Dong, Lin Qi, Mike J. ChantlerZenodo 2012 · DOI

Abstract

When cooking porridge, humans can easily spot how vigorously the porridge is boiled and accordingly adjust the cooking power. The change of the porridge surface appearance can be regarded as dynamic textures. This paper studies the correlation between several computational descriptors and human perception using image sequences of boiling porridges. By comparing the chosen features with the results of psychophysical experiments, we found that the texture directionality descriptor proposed by Tamura et al and the inverse difference moment descriptor proposed by Haralick et al best coincide with human's perception. Our findings can be further used to judge how vigorously the porridge is boiled and accordingly adjust the cooking power in automatic cooking systems. This record was migrated from the OpenDepot repository service in June, 2017 before shutting down.
BibTeX
@inproceedings{sun2012automatic,
  author = {Yujuan Sun and Junyu Dong and Lin Qi and Mike J. Chantler},
  title = {{Automatic Cooking Of Porridge Based On Human Visual Perception}},
  booktitle = {Zenodo},
  year = {2012},
  doi = {10.5281/zenodo.811663},
}

Loading related content...

Figure