Abstract
In order to improve the effect of glaucoma fundus image classification, a new algorithm based on decision tree and UNet++ was proposed. Firstly, the image is divided into three channels of RGB, and the extracted green channel image is enhanced with the Butterworth parameter function of the fusion power function. Then the improved UNet++ network model is used to extract the texture features of the fundus image, and the residual module is used to enhance the texture features. The results of the experiment show that the average accuracy, the average specificity and the average sensitivity of the improved algorithm increase by 9.2%, 6.4% and 6.5% respectively. The improved algorithm is effective in glaucoma fundus image classification.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Aich, G., Banerjee, G., Debnath, S., Sen, A.: Optical disc segmentation from color fundus image using contrast limited adaptive histogram equalization and morphological operations. In: 2021 International Conference on Smart Generation Computing, Communication and Networking (SMART GENCON), pp. 1ā6 (2021). https://doi.org/10.1109/SMARTGENCON51891.2021.9645808
Li, X.: Research on image analysis and automatic disease diagnosis algorithm based on color fundus image. Beijing Institute of Technology (2017)
He, X., Xu, J., Chen, W.: Research on fundus blood vessel image segmentation based on improved U-Net network. J. Electron. Meas. Instr. 35(10), 202ā208 (2021). https://doi.org/10.13382/j.jemi.b2003781
Deari, S., ĆksĆ¼z, Ä°., Ulukaya, S.: Importance of data augmentation and transfer learning on retinal vessel segmentation. In: 2021 29th Telecommunications Forum (TELFOR), pp. 1ā4 (2021). https://doi.org/10.1109/TELFOR52709.2021.9653400
Yuan X, Guo H, Lu, L., Wei, L., Yu, Z.: High-resolution remote sensing image change detection algorithm based on UNet++ network and attention mechanism. J. Surv. Mapp. Sci. Technol. 38(02), 155ā159 (2021)
Serener, A., Serte, S.: Glaucoma classification via deep learning ensembles. In: 2021 International Conference on INnovations in Intelligent SysTems and Applications (INISTA), pp. 1ā5 (2021). https://doi.org/10.1109/INISTA52262.2021.9548439
Guo, C., Li, W., Zhao, X., Zou, B.: Glaucoma screening method guided by semantic feature map. J. Comput. Aided Design Graph. 33(03), 363ā375 (2021)
Gupta, G., Kulasekaran, S., Ram, K., et al.: Local characterization of neovascularization and identification of proliferative diabetic retinopathy in retinal fundus images. Comput. Med. Imaging Graph. 55(SC), 124ā132 (2017)
Ke, S., Hu, M., Xu, Y.: Research on computer-aided glaucoma diagnosis algorithm based on ensemble learning. J. Beijing Univ. Chem. Technol. (Nat. Sci. Edn.) 46(04), 86ā91 (2019). https://doi.org/10.13543/j.bhxbzr.2019.04.013
Das, P., Nirmala, S.R., Medhi, J.P.: Diagnosis of glaucoma using CDR and NRR area in retina images. Netw. Model. Anal. Health Inform. Bio Inform. 5(1), 1ā14 (2016)
Venugopal, N., Mari, K., Manikandan, G., Sekar, K.R.: Phase quantized polar transformative with cellular automaton for early glaucoma detection. Ain Shams Eng. J. 12(4), 4145 (2021)
Zhu, Z., Li, H., Zhao, L., Wang, H., Tan, J.: Research on image enhancement algorithm of magnetic tile surface defects based on improved homomorphic filtering and SMQT. Optoelectron. Laser 32(08), 818ā825 (2021). https://doi.org/10.16136/J.Joel.2021.08
Hao, M.: Research on underwater image enhancement method based on pyramid fusion. Dalian Maritime University (2020). https://doi.org/10.26989/D.CNKI.GDLHU.2020.000020
Wang, W., Chen, W., Wang, L., Yuan, J.: Research on a method of adjusting the contrast of fundus images. China Med. Equip. 12, 11ā13 (2006)
Zhou, Z., Rahman Siddiquee, M.M., Tajbakhsh, N., Liang, J.: UNet++: a nested U-net architecture for medical image segmentation. In: Stoyanov, D., et al. (eds.) DLMIA/ML-CDS -2018. LNCS, vol. 11045, pp. 3ā11. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-00889-5_1
Fu, Y., Li, Z., Zhang, Y., Pan, D.: A U-net with residual module for detecting the exudation of fundus images. Minicomput. Syst. 42(07), 1479ā1484 (2021)
Chen, H., Zhen, X., Zhao, T.: A small target detection model based on channel-spatial attention mechanism feature fusion [J/OL]. J. Huazhong Univ. Sci. Technol. (Nat. Sci. Edn.) 1ā8 (2022). https://doi.org/10.13245/J.HUST.238491
Chen, X., Well, L., Zhang, Y., Xue, Q.: Post-nonlinear BSS algorithm based on multilayer neural network and PReLU function. J. Mianyang Norm. Univ. 39(02), 25ā30+46 (2020). https://doi.org/10.16276/j.cnki.cn51-1670/g.2020.02.006
Fang, C., Fan, M.: Text detection based on improved convolutional neural network and line features. Microelectron. Comput. 36(08), 77ā82 (2019). https://doi.org/10.19304/j.cnki.issn1000-7180.2019.08.017
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
Ā© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Wang, X., Qiu, Y., Chen, X., Wu, J., Zou, Q., Mu, N. (2023). Decision Tree Fusion andĀ Improved Fundus Image Classification Algorithm. In: Yu, C., Zhou, J., Song, X., Lu, Z. (eds) Green, Pervasive, and Cloud Computing. GPC 2022. Lecture Notes in Computer Science, vol 13744. Springer, Cham. https://doi.org/10.1007/978-3-031-26118-3_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-26118-3_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-26117-6
Online ISBN: 978-3-031-26118-3
eBook Packages: Computer ScienceComputer Science (R0)