This project is object classification using pre-trained model. Two well-known models are used NasNetLarge and Xception.
The code begins by resizing the images to size that fit the model(NasNetLarge or Xception) then the image is pre-processed and then the model classifies the image. if the probability of classification exceeds a certain threshold then the results of the model will be reported otherwise the model failed to classify the image.
- numpy
- keras
- opencv
- glob
- tensorflow