Classification of Image using Deep Neural Networks and SoftMax Classifier with CIFAR datasets

The process of image classification involves the feature extraction from the problem image, processing of those features and classification using the efficient classifier to obtain the necessary output. This work is also doing the same by extracting the features from the input image and the extracted information is processed with the help of Deep Neural Network (DNN) and the image classification is done using SoftMax Classifier. The combination of DNN and the SoftMax Classifier provides the efficient image classification and the data sets that used in this work is CIFAR 10, CIFAR 100 and MNIST to obtain the best results. The training accuracy and the validation accuracy of the image data should be increased and the training loss and the validation loss of the image data should be decreased in the image classification process for efficient results. The accuracy, precision value and those averages are listed, and the macro average and the weighted average of all the recall and precision values with their support values for the CIFAR datasets are also tabled.