Colorectal Polyp Image Segmentation Based on DeepLab-RCP
DOI:
https://doi.org/10.54097/s7s0x240Keywords:
Polyp Segmentation, Deeplab, Multi-scale Feature Fusion, ResNeXt50, DSConvAbstract
Colorectal cancer early screening and polyp resection are of great significance for prevention. With the development of deep learning technologies, computer-aided detection has become a mainstream method for polyp segmentation. However, challenges such as variations in polyp size, complex shapes, and blurred boundaries remain. To address these issues, this paper proposes an improved polyp segmentation model based on the DeepLabv3+ framework, named DeepLab-RCP (DeepLab with ResNeXt, Cross-Dense Fusion, and Parallel Swin-ASPP). The model designs a Cross-Dense Fusion (CDF) module to tackle size variations and a Parallel Swin Transformer and Atrous Spatial Pyramid Pooling (PS-ASPP) module to handle shape complexity. Additionally, ResNeXt-50 and Depthwise Separable Convolution (DSConv) are employed to optimize boundary segmentation. Experimental results demonstrate that the model achieves mean Dice coefficients and mean intersection over union of 92.8% and 86.9%, respectively, on the Kvasir-SEG dataset, and 95.5% and 91.6% on the CVC-ClinicDB dataset, validating its effectiveness.
Downloads
References
[1] Y. S. B, A. R, and M. A, "Deep learning applications in medical image segmentation: overview, approaches, and challenges," in Plastics, 2nd ed. vol. 3, J. Peters, Ed. Hoboken, NJ: John Wiley & Sons, 2025, pp. 15–64.
[2] J. Long, E. Shelhamer, and T. Darrell, "Fully convolutional networks for semantic segmentation," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, Boston, MA, 2015, pp. 3431–3440.
[3] O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," in *Proc. 18th Int. Conf. Medical Image Computing and Computer-Assisted Intervention*, Munich, Germany, 2015, pp. 234–241.
[4] K. Wang, Y. T. Lu, J. Liu, et al., "GAML: Geometry-aware mutual learning for polyp segmentation," Biomedical Signal Processing and Control, vol. 108, p. 107965, 2025.
[5] V. Badrinarayanan, A. Kendall, and R. Cipolla, "SegNet: A deep convolutional encoder-decoder architecture for image segmentation," IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 12, pp. 2481–2495, Dec. 2017.
[6] T. K. Dutta, et al., "SAM-Mamba: Mamba guided SAM architecture for generalized zero-shot polyp segmentation," in Proc. IEEE Winter Conf. Applications of Computer Vision, Waikoloa, HI, 2025, pp. 4655–4664.
[7] H. Zhao, J. Shi, X. Qi, et al., "Pyramid scene parsing network," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, Honolulu, HI, 2017, pp. 2881–2890.
[8] L. C. Chen, Y. Zhu, G. Papandreou, et al., "Encoder-decoder with atrous separable convolution for semantic image segmentation," in Proc. European Conf. Computer Vision, Munich, Germany, 2018, pp. 801–818.
[9] M. Yang, K. Yu, C. Zhang, et al., "DenseASPP for semantic segmentation in street scenes," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, Salt Lake City, UT, 2018, pp. 3684–3692.
[10] C. H. Huang, H. Y. Wu, and Y. L. Lin, "HarDNet-MSEG: A simple encoder-decoder polyp segmentation neural network that achieves over 0.9 mean dice and 86 fps," IEEE Trans. Pattern Anal. Mach. Intell., to be published.
[11] J. Hu, L. Shen, and G. Sun, "Squeeze-and-excitation networks," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, Salt Lake City, UT, 2018, pp. 7132–7141.
[12] S. Woo, J. Park, J. Y. Lee, et al., "CBAM: Convolutional block attention module," in Proc. European Conf. Computer Vision, Munich, Germany, 2018, pp. 3–19.
[13] A. Srivastava, D. Jha, S. Chanda, et al., "MSRF-Net: A multi-scale residual fusion network for biomedical image segmentation," IEEE J. Biomed. Health Inform., vol. 26, no. 5, pp. 2252–2263, May 2021.
[14] D. Jha, P. H. Smedsrud, M. A. Riegler, et al., "Kvasir-SEG: A segmented polyp dataset," in Proc. 26th Int. Conf. MultiMedia Modeling, Daejeon, South Korea, 2020, pp. 451–462.
[15] D. Vázquez, J. Bernal, F. J. Sánchez, et al., "A benchmark for endoluminal scene segmentation of colonoscopy images," J. Healthcare Eng., vol. 2017, p. 4037190, 2017.
[16] Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, et al., "UNet++: A nested U-Net architecture for medical image segmentation," in Proc. 4th Int. Workshop Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, Granada, Spain, 2018, pp. 3–11.
[17] H. Zhao, J. Shi, X. Qi, et al., "Pyramid scene parsing network," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, Honolulu, HI, 2017, pp. 2881–2890.
[18] T. K. Dutta, et al., "SAM-Mamba: Mamba guided SAM architecture for generalized zero-shot polyp segmentation," in Proc. IEEE Winter Conf. Applications of Computer Vision, Waikoloa, HI, 2025, pp. 4655–4664.
[19] L. C. Chen, Y. Zhu, G. Papandreou, et al., "Encoder-decoder with atrous separable convolution for semantic image segmentation," in Proc. European Conf. Computer Vision, Munich, Germany, 2018, pp. 801–818.
[20] D. Fan, Q. Yang, X. Zhang, et al., "PraNet: Parallel reverse attention network for polyp segmentation," in *Proc. 23rd Int. Conf. Medical Image Computing and Computer-Assisted Intervention*, Lima, Peru, 2020, pp. 555–563.
[21] J. Liu, K. Wang, Y. T. Lu, et al., "CaraNet: Contextual axial reverse attention network for small medical object segmentation," IEEE J. Biomed. Health Inform., vol. 27, no. 8, pp. 3608–3617, Aug. 2023.
[22] D. Jha, P. H. Smedsrud, M. A. Riegler, et al., "Kvasir-SEG: A segmented polyp dataset," in Proc. 26th Int. Conf. MultiMedia Modeling, Daejeon, South Korea, 2020, pp. 451–462.
[23] D. Vázquez, J. Bernal, F. J. Sánchez, et al., "A benchmark for endoluminal scene segmentation of colonoscopy images," J. Healthcare Eng., vol. 2017, p. 4037190, 2017.
[24] J. Moorthy and U. D. Gandhi, "A survey on medical image segmentation based on deep learning techniques," Big Data and Cognitive Computing, vol. 6, no. 4, p. 117, 2022.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
