[1]胡一凡,肖满生,范明凯,等.基于 Mask R-CNN 的胃肠息肉图像增强检测[J].计算机技术与发展,2023,33(03):173-179.[doi:10. 3969 / j. issn. 1673-629X. 2023. 03. 026]
　HU Yi-fan,XIAO Man-sheng,FAN Ming-kai,et al.Mask R-CNN Based Image Enhancement Detection of Gastrointestinal Polyps[J].,2023,33(03):173-179.[doi:10. 3969 / j. issn. 1673-629X. 2023. 03. 026]
点击复制

基于 Mask R-CNN 的胃肠息肉图像增强检测()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
33
期数:
2023年03期
页码:
173-179
栏目:
人工智能
出版日期:
2023-03-10

文章信息/Info

Title:
Mask R-CNN Based Image Enhancement Detection of Gastrointestinal Polyps
文章编号:
1673-629X(2023)03-0173-07
作者:
胡一凡; 肖满生; 范明凯; 吴宇杰
湖南工业大学 计算机学院,湖南 株洲 412007
Author(s):
HU Yi-fan; XIAO Man-sheng; FAN Ming-kai; WU Yu-jie
School of Computer Science,Hunan University of Technology,Zhuzhou 412007,Chin
关键词:
Mask R-CNN; 可变形卷积; 边缘强度; 息肉图像检测; 图像分割
Keywords:
Mask R-CNN; deformable convolution; edge strength; polyp image detection; image segmentation
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2023. 03. 026
摘要:
对胃肠息肉图像进行检测,是医学影像识别中的一个重要部分,在胃肠息肉图像中检测不显著息肉和小目标息肉更是一个难点。 为了提升胃肠息肉图像的检测率,减少对息肉病理的误判,提出了一种基于 Mask R-CNN 的胃肠息肉增强检测模型。 模型网络结构上采用残差网络和特征金字塔网络对图像进行多尺度特征提取,在残差网络的卷积层中改进使用了可变形卷积以扩大模型的采样范围,采样后的特征图进一步输入区域候选网络中对息肉区域筛选。 模型的预测输出部分设计了一个增强图像边缘强度的检测模块,通过滑动窗口二次检测的方式,增强对不显著息肉和小目标息肉的检测。 模型能同时进行息肉图像检测和分割,并在 Kvasir-SEG 与 CVC-ClinicDB 两个息肉数据集上进行了验证,实验结果表明提出的方法能够在原有模型的基础上提升对息肉病灶的检测精确度和分割精确度,检测精确度分别达到了 94. 3% 和98. 6% ,分割精确度分别达到了 93. 7% 和 98. 7% ,优于其他的对比模型。
Abstract:
The detection of gastrointestinal polyps is an important part of medical image recognition, and the detection of non - salientpolyps and small target polyps in gastroin　testinal polyp images is even more difficult. In order to improve the detection rate ofgastrointestinal polyp images and reduce the misclassification of polyp pathology,a Mask R-CNN based enhanced detection model forgastrointestinal polyps is proposed. The model uses a residual network and a feature pyramid network in the network structure for multi-scale feature extraction,and a deformable convolution is improved in the convolution layer of the residual network to extend the samplingrange of the model,and the sampled feature maps are further input into the region proposal network for polyp region screening. Thepredictive output part of the model is designed to enhance the detection of non - salient polyps and small target polyps by means of asliding window secondary detection module with enhanced image edge strength. The model was validated on two polyp datasets,Kvasir-SEG and CVC-ClinicDB,and the experimental results showed that the proposed method could improve the detection and segmentationprecision of polyp lesions based on the original model,the detection precision reached 94. 3% and 98. 6% ,and segmentation precisionreached 93. 7% and 98. 7% ,respectively,outperforming other compared models and validating the effectiveness of the model.

相似文献/References:

[1]姜世浩,齐苏敏,王来花,等.基于 Mask R-CNN 和多特征融合的实例分割[J].计算机技术与发展,2020,30(09):65.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 012]
　JIANG Shi-hao,QI Su-min,WANG Lai-hua,et al.Instance Segmentation Modal Based on Mask R-CNN and Multi-feature Fusion[J].,2020,30(03):65.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 012]
[2]苏 萍,朱晓辉.基于单目视觉的水面目标识别与测距方法研究[J].计算机技术与发展,2021,31(02):80.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 015]
　SU Ping,ZHU Xiao-hui.Research on Water Surface Target Recognition and Ranging Based on Monocular Vision[J].,2021,31(03):80.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 015]
[3]温志强,冯向萍,徐 静.基于 Mask R-CNN 的马匹面部别征识别及分割方法[J].计算机技术与发展,2021,31(06):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 037]
　WEN Zhi-qiang,FENG Xiang-ping,XU Jing.Recognition and Segmentation of Horse Facial FeaturesBased on Mask R-CNN[J].,2021,31(03):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 037]
[4]胡 强,张太红,赵昀杰,等.基于多模型联合的身份证人脸验证应用研究[J].计算机技术与发展,2021,31(08):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 036]
　HU Qiang,ZHANG Tai-hong,ZHAO Yun-jie,et al.Research on Application of ID Face Authentication Based onMulti-model Union[J].,2021,31(03):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 036]
[5]谢斌红,赵金朋,张英俊.结合注意力机制的车型检测算法[J].计算机技术与发展,2021,31(12):78.[doi:10. 3969 / j. issn. 1673-629X. 2021. 12. 014]
　XIE Bin-hong,ZHAO Jin-peng,ZHANG Ying-jun.Vehicle Detection Algorithm Combined with Attention Mechanism[J].,2021,31(03):78.[doi:10. 3969 / j. issn. 1673-629X. 2021. 12. 014]
[6]叶耀光,陈宗楠,陈丽群,等.基于通道注意的可变形金字塔表情识别网络[J].计算机技术与发展,2022,32(11):64.[doi:10. 3969 / j. issn. 1673-629X. 2022. 11. 010]
　YE Yao-guang,CHEN Zong-nan,CHEN Li-qun,et al.Channel-attention-based Deformable Pyramid Network for Facial Expression Recognition[J].,2022,32(03):64.[doi:10. 3969 / j. issn. 1673-629X. 2022. 11. 010]
[7]宋 浩*,张 鸿.基于完全残差的双分支去雨网络[J].计算机技术与发展,2023,33(11):57.[doi:10. 3969 / j. issn. 1673-629X. 2023. 11. 009]
　SONG Hao*,ZHANG Hong.A Dual-branch De-raining Network Based on Fully Residual[J].,2023,33(03):57.[doi:10. 3969 / j. issn. 1673-629X. 2023. 11. 009]
[8]余 咏,吴建平,何旭鑫,等.基于改进 YOLOv4-tiny 的节肢动物目标检测模型[J].计算机技术与发展,2024,34(01):114.[doi:10. 3969 / j. issn. 1673-629X. 2024. 01. 017]
　YU Yong,WU Jian-ping,HE Xu-xin,et al.Arthropod Object Detection Model Based on Improved YOLOv4-tiny[J].,2024,34(03):114.[doi:10. 3969 / j. issn. 1673-629X. 2024. 01. 017]

更新日期/Last Update: 2023-03-10