Usually many people don't know much about their own dogs. This app help them to know more details about it. History, Health care, character and behavior, etc. anything related with the dog particular breed. - How to use it? Try to hold the camera still, don't move while aiming a dog and wait until the rectangle's lines gets green and the blue number gets as closer to 1.0 as possible. This number is the level of confidence. Wait for a few seconds, at this point it's trying to get the best guess. It may change a few times from different breeds until it gets stable and the number of confidence reaches the highs score possible.

How to use it?
Try to hold the camera still, don't move while aiming a dog and wait until the rectangle's lines gets green and the blue number gets as closer to 1.0 as possible. This number is the level of confidence.
Wait for a few seconds, at this point it's trying to get the best guess. It may change a few times from different breeds until it gets stable and the number of confidence reaches the highs score possible.

This is a Computer Vision prove of concept App to demonstrate the viability of using powerful Machine Learning Convolutional Neural Networks which can classify amount different breeds of dogs on mobile device. Using Stanford University dataset breeds of dogs to retrain a Convolutional Neural Network with TensorFlow using Python based in Linux and optimized to run with Java for Android OS.