Laser & Optoelectronics Progress, Volume. 57, Issue 24, 241025(2020)

Deep Learning Feature Fusion-Based Retina Image Classification

Tianfu Zhang1, Shuncong Zhong1、*, Chaoming Lian1, Ning Zhou1, and Maosong Xie2
Author Affiliations
  • 1School of Mechanical Engineering and Automation, Fuzhou University, Fuzhou, Fujian 350108, China
  • 2First Affiliated Hospital of Fujian Medical University, Fuzhou, Fujian 350000, China
  • show less

    Aim

    ing at the problems of missed detection and low efficiency in manual classification and diagnosis of optical coherence tomography retina images, a deep learning-based convolutional network classification algorithm is proposed to construct joint multilayer features. First, retinal images are preprocessed using the mean shift and data normalization algorithm. The loss function weighting algorithm is combined to solve the data imbalance problem. Second, a lightweight deep separable convolution rather than an ordinary convolution layer is used to reduce the number of model parameters. Global average pooling replaces fully connected layers to increase spatial robustness, and different convolutional layers are used to build feature fusion layers to enhance feature circulation between layers. Finally, the SoftMax classifier is used for image classification. Experimental results show that the model can achieve 97%, 95%, and 97% in accuracy, precision, and recall, respectively, thereby reducing the recognition time. The proposed deep learning feature fusion-based method performs well in the classification and diagnosis of retinal images.

    Tools

    Get Citation

    Copy Citation Text

    Tianfu Zhang, Shuncong Zhong, Chaoming Lian, Ning Zhou, Maosong Xie. Deep Learning Feature Fusion-Based Retina Image Classification[J]. Laser & Optoelectronics Progress, 2020, 57(24): 241025

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Image Processing

    Received: Apr. 24, 2020

    Accepted: Jun. 9, 2020

    Published Online: Dec. 9, 2020

    The Author Email: Zhong Shuncong (zhongshuncong@hotmail.com)

    DOI:10.3788/LOP57.241025

    Topics