Image Classification Test phase, srs, design phase and source code final deliverable

Image Classification Test phase, srs, design phase and source code final deliverable

Project Domain / Category

Artificial Intelligence + Image Processing.

Abstract / Introduction

The domain of image processing deals with enhancing images and extracting useful information. Examples of image processing techniques are object detection, image classification, face recognition, forgery detection, and many more. Image classification is the task of assigning an input image one label from a fixed set of categories.

The cat and dog image classification problem sounds simple and effectively addressed through advanced neural network concepts. The main goal of this project is to develop a robust model to classify from a large dataset of cats and dogs to classify whether the image of a cat or dog. In this project, Students are required to create a real-time system to classify cat and dog images using the Dogs vs Cats Dataset. This dataset includes 25000 images of cats and dogs. The dataset link has been shared below from which students can download it by visiting it. There will be 2 modules to complete this project. Students are required to complete all requirements of these modules.

Module 1:

• The system will first Import the image dataset of cats and dogs from described link.
• The system will preprocess the images.
• System will Split the image dataset into train and test directories.
• System will use Neural Network to train the data.
  • System needs to use Convolutional Neural Network (CNN) model for training.
  • System needs to use any one transfer learning (Inception/Xception/VGG/ResNet) technique for training.
• After completing the training process, system will evaluate the trained model on the test data.
• System will save the model for future use.
• System will take unknown random images to classify whether it is a cat image or dog image.

Module 2:

• User interaface is required for input images.
• The interface should provide the user an option to interact with the system, by first entering image and predicting whether the image is cat or dog.

Dataset:

• https://www.kaggle.com/c/dogs-vs-cats/data

Important links and Tutorials:

• Python
• https://www.w3schools.com/python/
• https://www.tutorialspoint.com/python/index.htm
• Image processing
• https://regenerativetoday.com/some-basic-image-preprocessing-operations-forbeginners-in-python/
• https://www.section.io/engineering-education/image-preprocessing-in-python/ l https://www.tensorflow.org/tutorials/load_data/images
• Deep Learning
• https://www.simplilearn.com/tutorials/deep-learning-tutorial/guide-to-buildingpowerful-keras-image-classification-models
• https://www.analyticsvidhya.com/blog/2020/02/learn-image-classification-cnnconvolutional-neural-networks-3-datasets/
• Transfer learning
• https://towardsdatascience.com/transfer-learning-for-image-classification-usingtensorflow-71c359b56673
• https://www.kaggle.com/code/kmkarakaya/transfer-learning-for-imageclassification/notebook

Hardware Requirement:

• Processor –Core i3
• Hard Disk – 160 GB
• Memory – 12GB RAM l Monitor

Tools:

Language: Python (Only python language)

Framework: Anaconda

IDE: JupyterNotebook, Pycharm, Spyder, Visual Studio Code, etc.

Supervisor:

Name: Madiha Faqir Hussain