Research Article
Tree-Based Convolutional Neural Networks for Image Classification
@INPROCEEDINGS{10.4108/eai.24-3-2022.2318997, author={Aamir Ahmad Ansari and Saba Raees and Nafisur Rahman}, title={Tree-Based Convolutional Neural Networks for Image Classification}, proceedings={Proceedings of the 3rd International Conference on ICT for Digital, Smart, and Sustainable Development, ICIDSSD 2022, 24-25 March 2022, New Delhi, India}, publisher={EAI}, proceedings_a={ICIDSSD}, year={2023}, month={5}, keywords={convolutional neural networks tree data structure deep learning classification image classification}, doi={10.4108/eai.24-3-2022.2318997} }
- Aamir Ahmad Ansari
Saba Raees
Nafisur Rahman
Year: 2023
Tree-Based Convolutional Neural Networks for Image Classification
ICIDSSD
EAI
DOI: 10.4108/eai.24-3-2022.2318997
Abstract
In this paper, we explore the use of tree-based Convolutional Neural Networks for image classification problems. The main goal of this image classification technique is to identify the traits, in an image, with precision. In the case of image classification, Convolutional Neural Networks (CNNs) are used because of their high precision. We explore the use of tree data structure to design the architect of our CNN model for image classification. Using n-ary trees we deduce the characteristics and compare them with the state-of-the-art Models. We employ the tiny ImageNet data set prepared by Stanford and the CIFAR-10 dataset [1]. This work shows how increasing the depth and width of a tree-structured CNN compares to the state-of-the-art models in terms of computational benefits, efficiency scores, and resource management.