Skip to content

Latest commit

 

History

History
117 lines (99 loc) · 2.6 KB

README.md

File metadata and controls

117 lines (99 loc) · 2.6 KB

HotDog-Nothotdog

"What would you say if I told you there is a app on the market that tell you if you have a hotdog or not a hotdog.

Alt text Alt text

Collect the data

Training Data

1- HotDogs images

2- NotHotdogs Images(Random images)

We can use some food image datasets Like

  http://mmspg.epfl.ch/food-image-datasets
  
  https://www.vision.ee.ethz.ch/datasets_extra/food-101/
  
  or use imagenet

How to train it?

Convolutional Neural Networks (ConvNets or CNNs) are a category of Neural Networks that have proven very effective in areas such as image recognition and classification. ConvNets have been successful in identifying faces, objects and traffic signs apart from powering vision in robots and self driving cars.

![Alt text](/CNN.jpg?raw=true "Optional Title")

Model, That I've used!! Alt text

You can use this model too

Layer Description
Input 128x128x1 Gray scale image
Convolution 8x8 4x4 subsampling
ELU
Convolution 5x5 2x2 subsampling
ELU
Convolution 5x5 2x2 subsampling
Flatten
Dropout .2 dropout probability
ELU
Fully connected output 512
Dropout .5 dropout probability
ELU
Fully connected output 2
Softmax output 2

(use this pre-trained model -- model.h5)-- thanks to Kevin