Research Article
Bipedal Thermographic Segmentation Based on Diabetic Foot Layout Guide for Journal of Physics: Conference Series using Microsoft Word
@INPROCEEDINGS{10.4108/eai.2-12-2022.2328003, author={Zhi Zeng and Zhenjie Cao and Junxia Zhu and Dan Liu}, title={Bipedal Thermographic Segmentation Based on Diabetic Foot Layout Guide for Journal of Physics: Conference Series using Microsoft Word}, proceedings={Proceedings of the 2nd International Conference on Information, Control and Automation, ICICA 2022, December 2-4, 2022, Chongqing, China}, publisher={EAI}, proceedings_a={ICICA}, year={2023}, month={3}, keywords={bipedal thermograms active thermography u2-net network diabetic foot detection}, doi={10.4108/eai.2-12-2022.2328003} }- Zhi Zeng
Zhenjie Cao
Junxia Zhu
Dan Liu
Year: 2023
Bipedal Thermographic Segmentation Based on Diabetic Foot Layout Guide for Journal of Physics: Conference Series using Microsoft Word
ICICA
EAI
DOI: 10.4108/eai.2-12-2022.2328003
Abstract
In this paper we propose an algorithm for segmentation of bipedal thermograms. To obtain the complete segmented foot region, we used a neural network trained using data obtained from active thermography. The foot region was segmented using improved U2-Net network. The results of experimental studies and simulations are given in this paper. The results are as follows. The proposed scheme can effectively segment the foot regions in different situations. The accuracy rate is 0.987, the miss detection rate is 0.006 and the detection speed is increased by 19%. The practical needs of diabetic foot detection can be better met by improving the U2-Net network segmentation algorithm.
Copyright © 2022–2024 EAI
