A Deep Feature Fusion Method Based on Dark Channel for Medical Image Segmentation

Hongyou Chen; Xiaodong Wang

doi:10.6180/jase.202012_23(4).0019

A Deep Feature Fusion Method Based on Dark Channel for Medical Image Segmentation

From left to right: original image, ACM-HEF, ACL, proposed method. From top to bottom: Metastatic bronchial carcinoma, Sarcoma, Metastatic cancer of the colon, Meningioma, Metastatic adenocarcinoma.

Hongyou Chen¹ and Xiaodong Wang ¹

¹School of Electrical Engineering, Zhengzhou University of Science and Technology Zhengzhou 450000,China

Received: July 15, 2020
Accepted: July 23, 2020
Publication Date: December 1, 2020

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: ||https://doi.org/10.6180/jase.202012_23(4).0019

ABSTRACT

The level set model is an advanced method in image segmentation, which has a better effect in medical image segmentation. Feature fusion strategy has been widely introduced into the frame work to improve the processing performance of complex images such as high noise and chaotic texture. The traditional image segmentation methods have the problem of uneven image gray scale. Therefore, we propose a deep feature fusion method based on dark channel for medical image segmentation in this paper. For the region feature of the images, the saliency feature of medical image is introduced. A deep edge feature extraction method based on dark channel is proposed. Based on the fusion result of region and deep edge features, the proposed method introduces the distance regularization term to normalize the level set function to enhance the stability of the evolution of the level set function. The new model is applied to medical image segmentation, and the experimental results show that the new model achieves more robust segmentation results and higher segmentation efﬁciency.

Keywords: Medical image segmentation, Deep feature fusion, Dark channel, Level set function, Distance regularization term

REFERENCES

[1]Shoulin Yin, Ye Zhang, Shahid Karim. “Large Scale Remote Sensing Image Segmentation Based on Fuzzy Region Competition and Gaussian Mixture Model”. IEEE Access. 6, pp: 26069-26080. (2018). doi: 1109/ACCESS.2018.2834960
[2]Shoulin Yin, Jie Liu, Hang Li. “A Self-Supervised Learning Method for Shadow Detection in Remote Sensing Imagery”. 3D Research, 9(4) (2018). doi: 10.1007/s13319-018-0204-9
[3]Huang G, Ji H, Zhang W, et al. “Adaptive multilayer level set method for segmenting images with intensity inhomogeneity”. IET Image Processing, 13(10), pp. 1714-1724, (2019). doi: 10.1049/iet-ipr.2019.0315
[4]Doshi T, Caterina G D, Soraghan J, et al. “Combining interpolation and 3D level set method (I+3DLSM) for medical image segmentation”. Electronics Letters, 52(8), pp. 592-594, (2016). doi:
[5]Shoulin Yin, Ye Zhang and Shahid Karim. “Region search based on hybrid convolutional neural network in optical remote sensing images”, International Journal of Distributed Sensor Networks, 15(5), (2019). doi: 10.1177/1550147719852036
[6]Teng Lin, Hang Li and Shoulin Yin. “Modified Pyramid Dual Tree Direction Filter-based Image De-noising via Curvature Scale and Non-local mean multi-Grade remnant multi-Grade Remnant Filter”, International Journal of Communication Systems. 31(16), (2018). doi: 10.1002/dac.3486
[7]Osher S. “Fronts propagating with curvature-dependent speed: Algorithms based on Hamilton-Jacobi formulations”. Journal of Computational Physics, 79(1), pp.12-49, (1988). doi:10.1016/0021-9991(88)90002-2
[8]F. Chan and L. A. Vese. "Active contours without edges." IEEE Transactions on Image Processing, 10(2), pp. 266-277, (2001). doi: 10.1109/83.902291.
[9]Li, C. Kao, J. C. Gore and Z. Ding. "Minimization of Region-Scalable Fitting Energy for Image Segmentation." IEEE Transactions on Image Processing, 17(10), pp. 1940-1949, (2008). doi: 10.1109/TIP.2008.2002304.
[10]Ding K, Xiao L, Weng G. “Active contours driven by region-scalable fitting and optimized Laplacian of Gaussian energy for image segmentation”. Signal Processing, 134(may), pp.224-233, (2017). doi: 10.1016/j.sigpro.2016.12.021
[11]Wang L, Chang Y, Wang H, et al. “An active contour model based on local fitted images for image segmentation”. Information sciences, 418, pp.61-73, (2017). doi: 10.1016/j.ins.2017.06.042
[12]Zhang and J. Song. "An Adaptive Fuzzy Level Set Model With Local Spatial Information for Medical Image Segmentation and Bias Correction." IEEE Access, 7, pp. 27322-27338, (2019). doi: 10.1109/ACCESS.2019.2900089
[13]Nileshsingh V. Thakur, V. R. Parihar. “Graph Theory based Approach for Image Segmentation using Wavelet Transform”. International Journal of Image Processing, 8(5), pp. 255-277, (2014). doi: no
[14]Dai S, Lu K, Dong J, et al. “A novel approach of lung segmentation on chest CT images using graph cuts”. Neurocomputing, 168(nov.30), pp.799-807, (2015). doi: 10.1016/j.neucom.2015.05.044
[15]Shoulin Yin, Jie Liu, Ye Zhang, Lin Teng. “Cuckoo search algorithm based on mobile cloud model”. International Journal of Innovative Computing, Information and Contro 12, pp.1809-1819. (2016). doi: 10.24507/ijicic.12.06.1809
[16]Shoulin Yin, Hang Li, Lin Teng. “Semantics automatic annotation in medical image based on deep learning”. Basic & Clinical Pharmacology & Toxicology. Dec 2018.
[17]Yang Sun, Shoulin Yin,Hang Li, Lin Teng, Shahid Karim. “GPOGC: Gaussian Pigeon-Oriented Graph Clustering Algorithm for Social Networks Cluster”. IEEE Access. 7, pp. 99254 - 99262, (2019). doi: 10.1109/ACCESS.2019.2926816
[18]Wu J, Song S, An W, et al. “Defect Detection and Localization of Nonlinear System Based on Particle Filter with an Adaptive Parametric Model”. Mathematical Problems in Engineering, pp. 1-12, (2015). doi:10.1155/2015/759035
[19]Nabizadeh N, John N. “Brain MRI Tumor Detection using Active Contour Model and Local Image Fitting Energy”, APS March Meeting 2014. American Physical Society, 2014.
[20]Jiang X L, Li B L, Yuan J Y, et al. “Active Contour Driven by Local Gaussian Distribution Fitting and Local Signed Difference Based on Local Entropy”. International Journal of Pattern Recognition and Artificial Intelligence, 30(3), (2016). doi: 10.1142/S0218001416550119
[21]Han B, Wu Y, Basu A. “An adaptive active contour model driven by weighted local and global image fitting constraints for image segmentation”. Signal Image and Video Processing, 14(2), pp. 1-8, (2020). doi:10.1007/s11760-019-01513-5
[22]Lin Teng, Hang Li, Shoulin Yin, Yang Sun. “Improved krill group-based region growing algorithm for image segmentation”. International Journal of Image and Data Fusion. 10(4), pp. 327-341, (2019). doi: 10.1080/19479832.2019.1604574
[23]Fu L H, Guo L, Wu W D. “Salient Region Detection based on Frequency-tuning and Region Contrast”, International Conference on Computer Science & Service System. pp. 732-735, (2014). doi:10.2991/csss-14.2014.171
[24]Sungmin, Lee, Seokmin, et al. “A review on dark channel prior based image dehazing algorithms”. Eurasip Journal on Image & Video Processing, (2016). doi:10.1186/s13640-016-0104-y
[25]Shoulin Yin, Jing Bi. “Medical Image Annotation Based on Deep Transfer Learning”. Journal of Applied Science and Engineering. Vol. 22, No. 2, pp. 385-390, 2019. doi:10.6180/jase.201906_22(2).0020
[26]Lin Teng, Hang Li and Shahid Karim. “DMCNN: A Deep Multiscale Convolutional Neural Network Model for Medical Image Segmentation”. Journal of Healthcare Engineering, (2019). doi: 10.1155/2019/8597606
[27]Lin Teng, Hang Li, Shoulin Yin, Shahid Karim &Yang Sun. “An active contour model based on hybrid energy and fisher criterion for image segmentation”. International Journal of Image and Data Fusion. 11(1), pp. 97-112. (2020). doi: 10.1080/19479832.2019.1649309
[28]Nie D, Shen D. “Adversarial Confidence Learning for Medical Image Segmentation and Synthesis”. International Journal of Computer Vision, 1-20, (2020). doi: 10.1007/s11263-020-01321-2