CogniVisio: Fusion Image Classifier

Zarinabegam Mundargi, Aryan Shinde, Shravani Bahirat, Shreya Das, Om Rajendra Deokar, Dipasha Shetiya · 2024

CogniVisio is an innovative project that leverages state-of-the-art technologies such as Google Colab, Python, TensorFlow, Convolutional Neural Network (CNN) Inception for developing a robust image classification system. The project addresses the challenge of enhancing image classification accuracy by employing a fusion approach that combines the strengths of image generation and the powerful feature extraction capabilities of CNN Inception. With a dataset comprising 8732 images, our methodology achieves an impressive accuracy of 86.86%, showcasing the effectiveness of the proposed fusion image classifier. The utilization of cloud-based computing through Google Colab provides scalability and accessibility, making CogniVisio a versatile solution for image classification task.

Read the paper · More papers on PaperTik