The Experts below are selected from a list of 113310 Experts worldwide ranked by ideXlab platform
Zhuowen Tu - One of the best experts on this subject based on the ideXlab platform.
-
holistically nested Edge Detection
International Journal of Computer Vision, 2017Co-Authors: Zhuowen TuAbstract:We develop a new Edge Detection algorithm that addresses two important issues in this long-standing vision problem: (1) holistic image training and prediction; and (2) multi-scale and multi-level feature learning. Our proposed method, holistically-nested Edge Detection (HED), performs image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets. HED automatically learns rich hierarchical representations (guided by deep supervision on side responses) that are important in order to resolve the challenging ambiguity in Edge and object boundary Detection. We significantly advance the state-of-the-art on the BSDS500 dataset (ODS F-score of 0.790) and the NYU Depth dataset (ODS F-score of 0.746), and do so with an improved speed (0.4 s per image) that is orders of magnitude faster than some CNN-based Edge Detection algorithms developed before HED. We also observe encouraging results on other boundary Detection benchmark datasets such as Multicue and PASCAL-Context.
-
Holistically-Nested Edge Detection
International Journal of Computer Vision, 2017Co-Authors: Saining Xie, Zhuowen TuAbstract:We develop a new Edge Detection algorithm that tackles two important issues in this long-standing vision problem: (1) holistic image training and prediction; and (2) multi-scale and multi-level feature learning. Our proposed method, holistically-nested Edge Detection (HED), performs image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets. HED automatically learns rich hierarchical representations (guided by deep supervision on side responses) that are important in order to approach the human ability resolve the challenging ambiguity in Edge and object boundary Detection. We significantly advance the state-of-the-art on the BSD500 dataset (ODS F-score of .782) and the NYU Depth dataset (ODS F-score of .746), and do so with an improved speed (0.4 second per image) that is orders of magnitude faster than some recent CNN-based Edge Detection algorithms.
Saining Xie - One of the best experts on this subject based on the ideXlab platform.
-
Holistically-Nested Edge Detection
International Journal of Computer Vision, 2017Co-Authors: Saining Xie, Zhuowen TuAbstract:We develop a new Edge Detection algorithm that tackles two important issues in this long-standing vision problem: (1) holistic image training and prediction; and (2) multi-scale and multi-level feature learning. Our proposed method, holistically-nested Edge Detection (HED), performs image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets. HED automatically learns rich hierarchical representations (guided by deep supervision on side responses) that are important in order to approach the human ability resolve the challenging ambiguity in Edge and object boundary Detection. We significantly advance the state-of-the-art on the BSD500 dataset (ODS F-score of .782) and the NYU Depth dataset (ODS F-score of .746), and do so with an improved speed (0.4 second per image) that is orders of magnitude faster than some recent CNN-based Edge Detection algorithms.
Zhencheng Chen - One of the best experts on this subject based on the ideXlab platform.
-
Edge Detection Based on Multi-Structure Elements Morphology
2006 6th World Congress on Intelligent Control and Automation, 2006Co-Authors: Yuqian Zhao, Wei-hua Gui, Zhencheng ChenAbstract:Edge Detection is an important pre-processing step in image analysis. Conventionally, mathematical morphology Edge Detection methods use single and symmetrical structure elements. But they are difficult to detect complex Edge feature, because they are only sensitive to image Edge which has the same direction of structure elements. This paper proposed a novel Edge Detection algorithm based on multi-structure elements morphology of eight different directions. We got eight different Edge Detection results by using morphological gradient algorithm respectively, and final Edge result was got by using synthetic weighted method. The experimental results showed that the proposed algorithm was more efficient for Edge Detection than conventional mathematical morphological Edge Detection algorithms and differential Edge Detection operators
-
Edge Detection Based on Multi-Structure
2006Co-Authors: Yuqian Zhao, Zhencheng ChenAbstract:Edge Detection is an important pre-processing step in image analysis. Conventionally, mathematical morphology Edge Detection methods use single and symmetrical structure elements. But they are difficult to detect complex Edge feature, because they are only sensitive to image Edge which has the same direction of structure elements. This paper proposed a novel Edge Detection algorithm based on multi-structure elements morphol- ogy of eight different directions. We got eight different Edge de- tection results by using morphological gradient algorithm respec- tively, and final Edge result was got by using synthetic weighted method. The experimental results showed that the proposed algo- rithm was more efficient for Edge Detection than conventional mathematical morphological Edge Detection algorithms and dif- ferential Edge Detection operators.
Yuqian Zhao - One of the best experts on this subject based on the ideXlab platform.
-
Edge Detection Based on Multi-Structure Elements Morphology
2006 6th World Congress on Intelligent Control and Automation, 2006Co-Authors: Yuqian Zhao, Wei-hua Gui, Zhencheng ChenAbstract:Edge Detection is an important pre-processing step in image analysis. Conventionally, mathematical morphology Edge Detection methods use single and symmetrical structure elements. But they are difficult to detect complex Edge feature, because they are only sensitive to image Edge which has the same direction of structure elements. This paper proposed a novel Edge Detection algorithm based on multi-structure elements morphology of eight different directions. We got eight different Edge Detection results by using morphological gradient algorithm respectively, and final Edge result was got by using synthetic weighted method. The experimental results showed that the proposed algorithm was more efficient for Edge Detection than conventional mathematical morphological Edge Detection algorithms and differential Edge Detection operators
-
Edge Detection Based on Multi-Structure
2006Co-Authors: Yuqian Zhao, Zhencheng ChenAbstract:Edge Detection is an important pre-processing step in image analysis. Conventionally, mathematical morphology Edge Detection methods use single and symmetrical structure elements. But they are difficult to detect complex Edge feature, because they are only sensitive to image Edge which has the same direction of structure elements. This paper proposed a novel Edge Detection algorithm based on multi-structure elements morphol- ogy of eight different directions. We got eight different Edge de- tection results by using morphological gradient algorithm respec- tively, and final Edge result was got by using synthetic weighted method. The experimental results showed that the proposed algo- rithm was more efficient for Edge Detection than conventional mathematical morphological Edge Detection algorithms and dif- ferential Edge Detection operators.
Sos S. Agaian - One of the best experts on this subject based on the ideXlab platform.
-
Logarithmic Edge Detection with Applications
Journal of Computers, 2008Co-Authors: Karen Panetta, Eric J. Wharton, Sos S. AgaianAbstract:<p class="MsoNormal" style="text-align: left; margin: 0cm 0cm 0pt; layout-grid-mode: char;" align="left"><span class="text"><span style="font-family: ";Arial";,";sans-serif";; font-size: 9pt;">In real world machine vision problems, numerous issues such as variable scene illumination make Edge and object Detection difficult. There exists no universal Edge Detection method which works well under all conditions. In this paper, we propose a logarithmic Edge Detection method based on Parameterized Logarithmic Image Processing (PLIP) and a four-directional Sobel method, achieving a higher level of independence from scene illumination. We present experimental results for this method, and compare results of the algorithms against several leading Edge Detection methods, such as Sobel and Canny. To compare results objectively, we use Pratt’s Figure of Merit. We demonstrate the application of the algorithm in conjunction with Edge Preserving Contrast Enhancement (EPCE), which is an image enhancement method dependent on the raw output of an Edge Detection kernel. This shows that the use of this Edge Detection algorithm results in better image enhancement, as quantified by the Logarithmic AME.</span></span><span style="font-family: ";Arial";,";sans-serif";; font-size: 9pt;"></span></p>
-
SMC - Logarithmic Edge Detection with applications
2007 IEEE International Conference on Systems Man and Cybernetics, 2007Co-Authors: Eric J. Wharton, Karen Panetta, Sos S. AgaianAbstract:In real world machine vision problems, issues such as noise and variable scene illumination make Edge and object Detection difficult. There exists no universal Edge Detection method which works under all conditions. In this paper, we propose a logarithmic Edge Detection method. This achieves a higher level of scene illumination and noise independence. We present experimental results for this method, and compare results of the algorithm against several leading Edge Detection methods, such as Sobel and Canny. For an objective basis of comparison, we use Pratt's Figure of Merit. We further demonstrate the application of the algorithm in conjunction with Edge Detection based Image Enhancement (EDIE), showing that the use of this Edge Detection algorithm results in better image enhancement, as quantified by the Logarithmic AME measure.
