An improved contour detection method based on level set and watershed transform

2009 ◽  
Author(s):  
Shuang Wang ◽  
Shenggao Fu ◽  
Licheng Jiao ◽  
Xiaojing Zhang
Keyword(s):  
Level Set ◽  
Detection Method ◽  
Contour Detection ◽  
Watershed Transform

A Contour Detection Method Based on Image Saliency

2020 ◽  
Author(s):  
Yongcai Pan ◽  
Yuwei Zhang ◽  
Qingzheng Liu ◽  
Zhaobin Wu ◽  
Wei Hu ◽  
...  
Keyword(s):  
Detection Method ◽  
Contour Detection ◽  
Image Saliency

Optimization of Doppler Echocardiographic Velocity Measurements Using an Automatic Contour Detection Method

2010 ◽  
Vol 36 (9) ◽  
pp. 1513-1524 ◽  
Author(s):  
Emmanuel Gaillard ◽  
Lyes Kadem ◽  
Marie-Annick Clavel ◽  
Philippe Pibarot ◽  
Louis-Gilles Durand
Keyword(s):  
Detection Method ◽  
Contour Detection ◽  
Velocity Measurements ◽  
Automatic Contour

scholarly journals Saliency Contour Extraction Based on Multi-level Feature Channel Optimization Coding

2021 ◽  
Vol 7 (4) ◽  
pp. 117
Author(s):  
Linling Fang ◽  
Yingle Fan
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Network Model ◽  
Detection Method ◽  
Gabor Filter ◽  
Contour Detection ◽  
Frequency Separation ◽  
Experimental Sample ◽  
Multi Level ◽  
End To End

<p>A biomimetic vision computing model based on multi-level feature channel optimization coding is proposed and applied to image contour detection, combining the end-to-end detection method of full convolutional neural network and the traditional contour detection method based on biological vision mechanism. Considering the effectiveness of the Gabor filter in perceiving the scale and direction of the image target, the Gabor filter is introduced to simulate the multi-level feature response on the visual path. The optimal scale and direction of the Gabor filter are obtained based on the similarity index, and they are used as the frequency separation parameter of the NSCT transform. The contour sub-image obtained by the NSCT transform is combined with the original image for feature enhancement and fusion to realize the primary contour response. The low-dimensional and low-redundancy primary contour response is used as the input sample of the network model to relieve network pressure and reduce computational complexity. A fully improved convolutional neural network model is constructed for multi-scale training, through feature encoder to feature decoder, to achieve end-to-end pixel prediction, and obtain a complete and continuous detection image of the subject contour. Using the BSDS500 atlas as the experimental sample, the average accuracy index is 0.85, which runs on the device CPU at a detection rate of 20+ FPS to achieve a good balance between training efficiency and detection effect.</p>

Sign in / Sign up

Export Citation Format

Share Document