Skip to content

In this project, we present the development and analysis of a Convolutional Neural Network (CNN) for the task of multi-class image classification. The study leverages the powerful combination of TensorFlow and Keras to create an effective deep-learning model for multi-class image classification.

License

Notifications You must be signed in to change notification settings

abhishtmishra/Computer-Vision-Multi-Class-Image-Classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Computer-Vision-Multi-Class-Image-Classification

In this project, we present the development and analysis of a Convolutional Neural Network (CNN) for the task of multi-class image classification. The study leverages the powerful combination of TensorFlow and Keras to create an effective deep-learning model. The methodology comprises several key components, including data preprocessing, model architecture design, and rigorous training.

Data preprocessing involves the extraction of image data from specified directories and the normalization of pixel intensities to a common scale, thus facilitating effective model training and ensuring robustness. The model architecture is constructed as a Sequential neural network, consisting of three convolutional layers with progressively increasing filter sizes. These convolutional layers are interleaved with batch normalization and max-pooling operations to enhance feature extraction and spatial reduction. Following the convolutional layers, three fully connected layers are incorporated, each with Rectified Linear Unit (ReLU) activation functions. Additionally, dropout layers are inserted after the first two dense layers to mitigate overfitting.

The CNN model is then compiled using the binary cross-entropy loss function and the Adam optimizer. Extensive training is performed on the preprocessed data, and the model’s performance is monitored across multiple epochs. Training accuracy and validation accuracy are visualized to assess the learning progress, while training loss and validation loss curves help in detecting overfitting. These visualizations play a crucial role in understanding the model’s convergence and generalization capabilities.

In conclusion, the presented research paper outlines a comprehensive implementation of a CNN for binary image classification using TensorFlow and Keras. The results and analysis highlight the model’s effectiveness in handling binary classification tasks, and the visualization of accuracy and loss metrics provides insights into its training behaviour. This work serves as a valuable reference for practitioners and researchers seeking to leverage CNNs for similar image classification tasks and offers a foundation for further advancements in the field of deep learning and computer vision.

About

In this project, we present the development and analysis of a Convolutional Neural Network (CNN) for the task of multi-class image classification. The study leverages the powerful combination of TensorFlow and Keras to create an effective deep-learning model for multi-class image classification.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published