Extracting useful features at multiple scales is a crucial task in computer vision. The emergence of deep-learning techniques and the advancements in convolutional neural networks (CNNs) have facilitated effective multiscale feature extraction that results in stable performance improvements in numerous real-life applications. However, currently available state-of-the-art methods primarily rely on a parallel multiscale feature extraction approach, and despite exhibiting competitive accuracy, the models lead to poor results in efficient computation and low generalization on small-scale images. Moreover, efficient and lightweight networks cannot appropriately learn useful features, and this causes underfitting when training with small-scale images or datasets with a limited number of samples. To address these problems, we propose a novel image classification system based on elaborate data preprocessing steps and a carefully designed CNN model architecture. Specifically, we present a consecutive multiscale feature-learning network (CMSFL-Net) that employs a consecutive feature-learning approach based on the usage of various feature maps with different receptive fields to achieve faster training/inference and higher accuracy. In the conducted experiments using six real-life image classification datasets, including small-scale, large-scale, and limited data, the CMSFL-Net exhibits an accuracy comparable with those of existing state-of-the-art efficient networks. Moreover, the proposed system outperforms them in terms of efficiency and speed and achieves the best results in accuracy-efficiency trade-off.
Extracting useful features at multiple scales is one of the most crucial tasks in computer vision. Emergence of deep learning(DL) techniques and progress of convolutional neural networks (CNNs) allow powerful multi-scale feature extraction that result in stable performance increase in numerous real-life applications. However, currently available state-of-the-art powerful methods primarily rely on parallel multi-scale feature extraction approach and despite providing competitive accuracy, the models lead to poor results in efficient computation and lack at generalizing on small-scale images. On the other hand, efficient and lightweight networks cannot learn useful features properly and result in underfitting problem in case of training with small-scale images or datasets with limited number of samples. Considering these problems, we proposed a novel image classification system based on elaborate data pre-processing step and carefully designed CNN model architecture. Specifically, consecutive multiscale feature learning network (CMSFL-Net) employs consecutive feature learning approach based on usage of various feature maps with different receptive fields to obtain faster training/inference and higher accuracy. In the conducted experiments using 6 real-life image classification datasets, including small-scale, large-scale, and limited data, the proposed system obtained comparable accuracy when compared with the existing state-of-the-art powerful and efficient networks by outperforming them in terms of efficiency and speed. Moreover, the proposed system achieved the best results in accuracy-efficiency tradeoff in comparison with the currently available models.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.