An Improved Canny Algorithm Based on Median Filtering and Adaptive Threshold in Robot Training System
- 10.2991/icmse-15.2015.78How to use a DOI?
- surgical robot, virtual training system, canny algorithm, salt and pepper noise
The system of robot virtual operation training requires effective soft-tissue organ models, of which the shape and size shall be obtained according to patients’ medical image. The way of cutting required organ image from CT scans with noise background is essentially important. This paper proposes a kind of improved algorithm that can effectively reduce salt-and-pepper noise in image, according to traditional Canny edge detection algorithm’s weak effect in reduction of salt-and-pepper noise in image and disadvantage of manual selection of double thresholds. Firstly, improved Gaussian weighted median filtering is applied by the algorithm on smooth processing of medical images, well reducing salt-and-pepper noise and effectively protecting edge detail information of medical images. Secondly, modified adaptive OTSU algorithm is used to acquire appropriate threshold value, according to the interclass variance value between the object and the background in medical image, so as to accurately locate medical image edge. The simulation result reveals the improved algorithm’s detection effect under background of salt-and-pepper noise is better than traditional canny edge detection algorithm and the published literature methods, thus the effectiveness of the algorithm are proved.
- © 2015, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Yidong Bao AU - Dongmei Wu PY - 2015/12 DA - 2015/12 TI - An Improved Canny Algorithm Based on Median Filtering and Adaptive Threshold in Robot Training System BT - Proceedings of the 2015 6th International Conference on Manufacturing Science and Engineering PB - Atlantis Press SP - 423 EP - 429 SN - 2352-5401 UR - https://doi.org/10.2991/icmse-15.2015.78 DO - 10.2991/icmse-15.2015.78 ID - Bao2015/12 ER -