Table 3.
Recognition accuracy of the food image recognition method with various feature types. SIFT, scale invariant feature transformation; LBP, local binary patterns.
| SIFT | LBP | Colour | SIFT + LBP + Colour | |
|---|---|---|---|---|
| Beans | 0.33 | 0.53 | 0.19 | 0.43 |
| Carrots | 0.37 | 0.44 | 0.46 | 0.51 |
| Cheese | 0.16 | 0.16 | 0.16 | 0.16 |
| Custards | 0.60 | 0.22 | 0.24 | 0.62 |
| Milk | 0.13 | 0.78 | 0.3 | 0.13 |
| Muscle meat | 0.40 | 0.32 | 0.5 | 0.55 |
| Oranges | 0.23 | 0.44 | 0.5 | 0.57 |
| Peas | 0.65 | 0.85 | 1.0 | 0.93 |
| Tomato | 0.35 | 0.29 | 0.29 | 0.33 |
| Yoghurt | 0.54 | 0.2 | 0.2 | 0.42 |
| Overall | 0.37 | 0.42 | 0.38 | 0.47 |