H
Hang Li
Researcher at Shenyang Normal University
Publications - 41
Citations - 399
Hang Li is an academic researcher from Shenyang Normal University. The author has contributed to research in topics: Image segmentation & Feature extraction. The author has an hindex of 7, co-authored 39 publications receiving 155 citations.
Papers
More filters
Journal ArticleDOI
Airport Detection Based on Improved Faster RCNN in Large Scale Remote Sensing Images
Shoulin Yin,Hang Li,Lin Teng +2 more
TL;DR: The proposed improved faster region-based convolutional neural network detection method for airport detection in large scale remote sensing images can accurately detect different airports under complex background with high detection rate, low false alarm rate and short running time.
Journal ArticleDOI
Hot Region Selection Based on Selective Search and Modified Fuzzy C-Means in Remote Sensing Images
Shoulin Yin,Hang Li +1 more
TL;DR: This work creates a Gaussian curvature filter to preprocess large scale remote sensing images and adopts an enhanced selective search method to establish well-defined boundaries for the HRS and to improve the immunity to noise.
Journal ArticleDOI
DMCNN: A Deep Multiscale Convolutional Neural Network Model for Medical Image Segmentation.
Lin Teng,Hang Li,Shahid Karim +2 more
TL;DR: A novelty model for medical image segmentation based on deep multiscale convolutional neural network (CNN) has strong robustness compared with other segmentation methods and can boost the segmentation accuracy.
Journal ArticleDOI
Active contour modal based on density-oriented BIRCH clustering method for medical image segmentation
TL;DR: A new medical image segmentation method is proposed, which adopts density-oriented BIRCH (balanced iterative reducing and clustering using hierarchies) clustering method to modify active contour model and improve the robustness of noise.
Journal ArticleDOI
A Network Intrusion Detection Method Based on Deep Multi-scale Convolutional Neural Network
TL;DR: DMCNN has a high intrusion detection accuracy and a low false alarm rate, which overcomes the limitations of using the traditional detection methods and makes the new approach an attractive one for practical intrusion detection.