Unified Hybrid Segmentation: Combining Classical Techniques with State-of-the-Art Deep Learning Models
Keywords:
Image Segmentation, GrabCut, Mask R-CNN, U-Net, Fully Convolutional Network (FCN), Deeplab V3, Hybrid Model, Deep LearningAbstract
In recent years, image segmentation has seen remarkable advancements through the development of various deep learning models. It plays a pivotal role in numerous computer vision applications, such as medical imaging, autonomous driving and scene understanding. This paper presents a novel hybrid segmentation approach that integrates the strengths of GrabCut, Mask R-CNN, U-Net, FCN and DeepLab v3 models to achieve superior segmentation performance. GrabCut provides an efficient graph-cut based foreground extraction, which serves as a refined initial mask for subsequent deep learning models. Mask R-CNN improves object detection and instance segmentation functionalities, while U-Net's encoder-decoder architecture excels in segmenting images with limited annotated data, making it particularly effective for medical imaging tasks. FCN contributes by enabling pixel-wise segmentation, ensuring comprehensive coverage of image details. Finally, DeepLab v3's atrous convolution and spatial pyramid pooling enable capturing multi-scale context, enhancing segmentation accuracy in complex scenes. The proposed hybrid approach is evaluated on multiple benchmark datasets, showing substantial improvements in segmentation accuracy and robustness compared to standalone models. Experimental results demonstrate that our hybrid model surpasses state-of-the-art methods in terms of evaluation metrics. This research covers the way for future advancements in image segmentation by combining the strengths of classical and deep learning-based techniques, offering a comprehensive solution for diverse segmentation challenges.
Downloads
References
Zhaobin Wang, Yongke Lv, Runliang Wu & Yaonan Zhang, "Review of GrabCut in Image Processing," Mathematics, MDPI, vol. 11(8), pp. 1-41, April 2023, https://doi.org/10.3390/math11081965 .
Hao Wu , Yulong Liu, Xiangrong Xu and Yukun Gao “Object Detection Based on the GrabCut Method for Automatic Mask Generation,” Micromachines, vol. 13, no. 12, pp. 2095, 2022. https://doi.org/10.3390/mi13122095.
Hyungjoon Kim , Jae Ho Lee and Suan Lee “Hybrid Image Segmentation Method for Accurate Measurement of Urban Environments,” Electronics, vol. 12, pp. 1845, 2023, https://doi.org/10.3390/electronics 12081845
Tiara Lestari Subaran, Transmissia Semiawan and Nurjannah Syakrani “Mask R-CNN and GrabCut Algorithm for an Image-based Calorie Estimation System,” Journal of Information Systems Engineering and Business Intelligence, vol. 8, no. 1, pp. 1-10, 2022, http://dx.doi.org/10.20473/jisebi.8.1.1-10.
C Rother, V Kolmogorov, A. Blake, “GrabCut -Interactive foreground extraction using iterated graph cuts,” ACM Trans. Graph, vol. 23, pp. 309-314, 2004, https://dl.acm.org/doi/pdf/10.1145/1015706. 1015720.
S Chouhan, A Kaul and U Singh, “Image segmentation using computational intelligence techniques: Review,” Archives of Computational Methods in Engineering, vol. 26, no. 3, pp. 533–596, 2018, https://doi.org/10.1007/s11831-018-9257-4.
Y Li, X Liu, X Gong and M Wang, “A multi-view features hinged siamese U-net for image Co-segmentation,” Multimedia Tools and Applications, vol. 80, no. 15, pp. 22965-22985, 2021. https://doi.org/10.1007/s11042-020-08794-w.
E Shelhamer, J Long and T Darrel, “Fully convolutional networks for semantic segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 4, pp. 640-651, 2017, https://doi.org/10.1109/TPAMI.2016.2572683.
K Liang, Y Jiang, M Long and G Liang, “Adaptive dual threshold based moving target detection algorithm,” IEEE 4th Information Technology and Mechatronics Engineering Conference, pp. 1111-1115, 2018, https://doi.org/10.1109/ITOEC.2018.8740402.
X Chen and L Pan, “A survey of graph cuts/graph search based medical image segmentation,” IEEE Reviews in Biomedical Engineering, vol. 11, pp. 112-124, 2018, https://doi.org/10.1109/RBME.2018.2798701.
F An and J Liu, "Medical image segmentation algorithm based on multilayer boundary perception-self attention deep learning model” Multimedia Tools and Applications, vol. 80, pp. 15017-15039, 2021, https://doi.org/10.1007/s11042-021-10515-w.
A Farahani and H Mohseni, “Medical image segmentation using customized U-net with adaptive activation functions,” Neural Computing and Applications, vol. 33, pp. 6307-6323, 2021, https://doi.org/10.1007/s00521-020-05396-3.
K He, G Gkioxari, P Dollár and R Girshick, Mask R-CNN. IEEE international conference on computer vision, pp. 2961-2969, 2017, doi: 10.1109/ICCV.2017.322.
L C Chen, G Papandreou, I Kokkinos, K Murphy and A L Yuille, “DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs,” IEEE transactions on pattern analysis and machine intelligence, vol. 40, no. 4, pp. 834-848, 2017.
J Long, E Shelhamer and T Darrell, “Fully convolutional networks for semantic segmentation,” IEEE conference on computer vision and pattern recognition, pp. 3431-3440, 2015.
O Ronneberger, P Fischer, and T Brox, “U-Net: Convolutional networks for biomedical image segmentation,” Medical image computing and computer-assisted intervention, pp. 234-241, 2015.
C Rother, V Kolmogorov and A Blake, “GrabCut: Interactive Foreground Extraction using Iterated Graph Cuts,” ACM Transactions on Graphics, vol. 23, no. 3, pp. 309-314, 2004
Xuchao Huang, Shigang Wang, Xueshan Gao , Dingji Luo, Weiye Xu , Huiqing Pang and Ming Zhou, “An H-GrabCut Image Segmentation Algorithm for Indoor Pedestrian Background Removal,” Sensors, vol. 23, pp. 7937, 2023, https://doi.org/10.3390/s23187937.
Peng Shi , Mengmeng Duan , Lifang Yang , Wei Feng , Lianhong Ding and Liwu Jiang, “An Improved U-Net Image Segmentation Method and Its Application for Metallic Grain Size Statistics,” Materials MDPI, vol. 15, pp. 4417, 2022, https://doi.org/10.3390/ma15134417.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
