maximum pooling and latest techniques to distinguish between animal images using keras

computer vision

It is very overwhelming as to how our brain can recognize all the different images and differentiate and process one image from the other. We can recognize one form of image or suppose an alphabet however twisted and rusted it is as what it means. Similarly in this paper we explore how neural networks work and how we can get datasets as to train the convolutional neural network to recognize one image from the other. More the data set the better will be the accuracy. In this paper we come to the conclusion and make a thorough study of images to say that neural networks is successful in recognizing one set of object from the other. It uses set of matrices in which the features are recognized by 1 and non features are recognized as zeros. We use python keras in getting the libraries such as pooling.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.