The main objective of the project is to build a Multi-Class Image Classifier to identify the food category, given an image of food.
The Dataset "Food-11 image dataset" is obtained from Kaggle (click here). This dataset contains 16643 food images grouped in 11 major food categories. There are 3 splits in this dataset:
- Training
- Validation
- Evaluation
The following Neural Network Architechture are used to build models using Transfer Learning in Python:
- Inception V3
- VGG 16
The Metric used for evaluation is Accuracy.
The models obtained the following accuracy after fine tuning them
Model | Accuracy |
---|---|
VGG-16 | 86.13% |
Inception V3 | 91.88% |
Inception V3 was selected as the finalized model.