Image Classification using Keras on CIFAR-10 dataset
Image Classification using Keras on CIFAR-10 dataset
Image classification is a fascinating deep learning project. Specifically, image classification comes under the computer vision project category.
In this project, we will build a convolution neural network in Keras with python on a CIFAR-10 dataset. First, we will explore our dataset, and then we will train our neural network using python and Keras.
CIFAR-10 is a very popular computer vision dataset. This dataset is well studied in many types of deep learning research for object recognition.
This dataset consists of 60,000 images divided into 10 target classes, with each category containing 6000 images of shape 3232. This dataset contains images of low resolution (3232), which allows researchers to try new algorithms. The 10 different classes of this dataset are:
CIFAR-10 dataset is already available in the datasets module of Keras. We do not need to download it; we can directly import it from keras.datasets.
The prerequisite to develop and execute image classification project is Keras and Tensorflow installation.
All the required steps can be referred from source.py or Image_Classification_Cifar-10.ipynb file.
The model acheives an accuracy of 84% in just 15 epochs and by visualizing the Classifiaction Accuracy graph it can be estimated that the model can acheive about 90% accuracy if it is trained for 50-60 epochs.