TensorFlow and Keras-based Convolutional Neural Network in CAT Image Recognition
Abstract
As a part of Computer Vision, image recognition and image classification plays an important role in development of Artificial Intelligence. Deep Learning is a new research area of Machine Learning approaches, which is motivated by building and imitating the natural neural network of human beings. It describes the data through similar approaches of human beings, which include images and sounds. We built up a supervised learning model of the given dataset of 209 pictures in RGB, through convolution layers, pooling layers and dense layers, with ReLU activation function, and the outputting sigmoid activation function. Finally, 90.0% of predictions through our model on test dataset are right.
Keywords
Convolutional neural network, Deep learning, Image recognition
DOI
10.12783/dtcse/cmsam2017/16428
10.12783/dtcse/cmsam2017/16428
Refbacks
- There are currently no refbacks.