Academic Journal of Computing & Information Science, 2025, 8(1); doi: 10.25236/AJCIS.2025.080101.
Haohang Li
Dipont Huayao Collegiate School Kunshan, Suzhou City, Jiangsu Province, 215300, China
The rapid advancements in deep learning have significantly transformed the field of image processing, particularly in feature extraction and classification tasks. This paper explores the application of deep learning techniques, primarily Convolutional Neural Networks (CNNs), in automating the extraction of meaningful features from images and performing accurate image classification. Traditional image processing methods, such as edge detection and handcrafted feature extraction, are limited by their reliance on domain-specific expertise. In contrast, deep learning models can learn hierarchical features from raw data, offering superior performance across various applications, including medical imaging, autonomous vehicles, and surveillance. This paper also examines the challenges associated with deep learning models, such as overfitting, the need for large labeled datasets, and high computational costs. Finally, the paper discusses the future directions of deep learning in image processing, including the integration of explainable AI, self-supervised learning, and edge computing, which could further enhance model efficiency and accessibility.
Deep Learning, Image Feature Extraction, Image Classification, Convolutional Neural Networks (CNN), Medical Imaging, Autonomous Vehicles, Data Preprocessing, Model Evaluation, Transfer Learning, Overfitting
Haohang Li. Research on image feature extraction and classification based on deep learning. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 1: 1-7. https://doi.org/10.25236/AJCIS.2025.080101.
[1] Wang, N. (2024). Research on Lung Nodule CT Image Classification Based on Deep Learning and Multi-modal Feature Fusion. Anhui University of Science and Technology. DOI: 10.26918/d.cnki. ghngc. 2024.000030.
[2] Wang, Z. (2024). Research on Vibration Signal Feature Extraction and Classification Algorithms Based on Deep Learning. Shijiazhuang Tiedao University. DOI: 10.27334/d.cnki.gstdy.2024.000331.
[3] Liu, S. (2024). Research on Text Classification Models Based on Deep Learning Feature Fusion Extraction. Yantai University. DOI: 10.27437/d.cnki.gytdu.2024.000425.
[4] Zhang, J. (2023). Research on Hyperspectral Image Classification Methods Based on Deep Feature Representation Learning. Xi'an University of Posts and Telecommunications. DOI: 10.27712/d.cnki. gxayd.2 023.000093.
[5] Shen, H. (2023). Research on Multi-feature Fusion Sea Ice Classification Method for Heterogeneous Remote Sensing Images Based on Deep Learning. Shanghai Ocean University. DOI: 10.27314/d.cnki. gsscu.2023.000463.
[6] Jingmin, L. (2022). Research on Electronic Information Feature Extraction and Classification Recognition Algorithm Based on Deep Learning. Security and Communication Networks, 2022.
[7] Shanshan, L., Ruo, H., Jianfang, W., et al. (2021). Research on Data Classification and Feature Fusion Method of Cancer Nuclei Image Based on Deep Learning. International Journal of Imaging Systems and Technology, 32(3), 969-981.
[8] Dongsheng, J., Zhujun, Z., Yanzhong, Z., et al. (2021). Research on Classification of COVID-19 Chest X-Ray Image Modal Feature Fusion Based on Deep Learning. Journal of Healthcare Engineering, 2021, 20216799202-6799202.