Convolution Neural Networks by Image Net
Journal Title: International journal of Emerging Trends in Science and Technology - Year 2017, Vol 4, Issue 9
Abstract
We trained a large, deep convolution neural network to classify the 1.2 million high-resolution images in the Image Net LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolution layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way soft ax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overfitting in the fully-connected layers we employed a recently-developed regularization method called ―dropout‖ that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry
Authors and Affiliations
Dr. K. Vikram
Keywords
Related Articles
- EP ID EP245650
- DOI -
- Views 122
- Downloads 0
Image Denoising Using Different Filters (A Comparison of Filters)
Image processing is basically the use of computer algorithms to perform image processing on digital images. Image denoising adds the manipulation of the image to produce a high quality image. The main criteria of Image d...
HASH Mechanism to Secure the AODV Protocol in MANETs
Mobile ad-hoc networks are self forming network without any fixed infrastructure. An ad hoc network does not have fixed infrastructure, it means a network without the standard routing infrastructure like fixed routers an...
Collaborative Filtering Recommender System
There has been an increase in the number of services available in the internet .Datasets are growing at a fast pace as it is being gathered and generated by a number of devices like smartphones, tablets and various infor...
Study and Analysis of Cutting Parameters of Rubber Rollers Using Doe Technique
Rubbers include natural rubbers and synthetic rubbers. Natural rubber is a naturally occurring substance obtained from the exudations of certain tropical plants. Synthetic rubber is artificially derived from petrochemica...
The Performance of Dovetail Halving Joint in Leg-and- Rail Construction Using Chrysophyllum Albidum Case Study: The Working Chair
to the optimal strength and rigidity of the chair joints. This study was set out to assess the performance of dovetail halving joint in leg-and-rail construction. In the study, mortise and tenon joint and dovetail halvin...