EECS-738-P3

Convolutional Neural Network Based on AlexNet for Image Prediction

Team Members

• Ian Yake
• Jules Garrett
• Jian Shen
• Dinesh Dandamudi
• Brian Quiroz

Overview

We developed a CNN (Convolutional Neural Network) based on AlexNet and used it to classify a dataset of images into 200 categories corresponding to the object shown in the image. After being trained on the ImageNet dataset, our model outputs confidence levels for each category with number percent accuracy.

Goal

Our goal is to write a Convolutional Neural Network that, given an image, classifies it.

Approach

Results

Accuracy

TBD

Requisites

The code was written from scratch using Python 3.7(?) and the following modules:

• Numpy
• Tensorflow
• OpenCV
• Pickle

References

https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf