Image Boundary Detection Algorithm Based on Snake Model

Wang Heng,Chen Shaoqiang

2007-01-01

Abstract:Some Snake Models are described in this paper, and the present research tries to combine Snake Algorithm and DP Algorithm to obtain the characteristic edge points of the image. The result shows that to use Imp roved Snake Algorithm can get a continuous and close edge, thus the target object can be extracted from the image well.

What problem does this paper attempt to address?

A New Algorithm To Detect Edge For Infrared Image Based On Two-Dimension Continuous Wavelet Transform

Cj Zhang,Xd Wang,Hr Zhang,Hj Feng

2005-01-01

Abstract:A new kind of algorithm to detect edge for infrared image is proposed employing the two-dimension continuous wavelet transforms. Direct calculation of the two-dimension continuous wavelet transform (TDCWT) for an image requires great amount of time and computer resources. FFT and inverse FFT are employed to compute convolution allowed to make such calculations practically possible and to reduce its time considerably. The two-dimension continuous wavelet transform is used to detect edge for infrared image here. Isotropy Mexican hat wavelet is selected as analyzing wavelet, which is able to meet our purpose because little texture information which is sensitive to directions in the infrared image. The Mexican hat wavelet has an excellent selectivity in position. It is quite efficient for a pointwise analysis, in particular the detection of local singularities (contours, corners). A new threshold algorithm is used directly to positive or negative wavelet coefficients in every resolution level at all scales so as to determine primary edge point sets. "Voting" method is proposed to integrate all the edge information at all scales. The new algorithm needn't connect edge points after threshold. Continuous and smooth edge can be obtained. The noise in the image can be suppressed to some extent, too. Experimental results show that the new algorithm can detect the whole contours and detail information in the infrared image well. It is more excellent than a few traditional edge detection operators in performance.
Initial Model of Snake and Improved Approaches

Jin Xue-jun

2006-01-01

Abstract:SNAKE model is mostly applied in medical image manipulation,motion track,shape recognition,and so on.An approach in tracking moving objects in video frequency is proposed and modified in this paper.Firstly,an initial contour line will be appointed close to the object interested.Then the contour line will be evolved and the object edge will be locked.When the initial contour line of the object is obtained,its motion should be tracked in subsequent frames without any input parameter from user.
Application of Snake Model for Edge Extraction Breast Tumor in Ultrasound Images

赵暖,陈亚青,余建国,王威琪

DOI: https://doi.org/10.3969/j.issn.1008-617x.2005.01.003

2005-01-01

Abstract:Objective To extract the edge of breast tumors in ultrasound images. Methods Because of the low signal-to-noise ratio (SNR) of medical ultrasound images, it was hard to get optimal result using classical edge extraction algorithms. As an effective edge extraction algorithm based on advanced information, Snake model attracts more and more attention. In this paper, Snake model was improved with the prior knowledge of ultrasound images from the original model. Results Fifteen ultrasound images of the breast tumor were collected from Shanghai Sixth People's Hospital. After a lot of pre-processing like gray-level segmentation and shape filtering,the improved Snake model was applied to the ultrasound images of breast tumor. This new method could extract the edge of the tumor in most test images and get better result than some classical methods. Conclusions The improved Snake model can extract the edge of breast tumors in ultrasound images and play an important part in computer-aided diagnosis algorithm of breast tumors.
Face Detection under Complex Background

Min Li,Bo Wang

2013-01-01

Abstract:Face detection and feature extraction are widely used in face recognition, expression recognition, video-phone conference, human-computer interface, and intelligent monitoring system and so on. In order to solve the problem of low accuracy in the use of the traditional face detection technology in the detection of human face under complex background, in this paper, we propose a new view for face detection method, which is Snake model base on shape restricted. Consider character of the human facial shape is similar to an ellipse, the traditional Snake model is modified by adding the shape energy item in its energy function, which strong the relationship among the Snake points so as to improve the accuracy. Experiments show that it is an effective approach for extracting facial contour under complex background.
A New Elliptic Contour Extraction Method for Reference Hole Detection in Robotic Drilling

Biao Mei,Weidong Zhu,Guorui Yan,Yinglin Ke

DOI: https://doi.org/10.1007/s10044-014-0394-6

IF: 2.307

2014-01-01

Pattern Analysis and Applications

Abstract:In robotic drilling of aircraft structures, reference holes are pre-drilled on aircraft structures and then detected by vision systems in the drilling process to compensate for the relative positioning errors between the robot tool center point and the workpiece, thus achieving improved position accuracy of drilled holes. In this paper, a novel elliptic contour extraction method based on salient region detection and optimization with snakes model is proposed for reference hole detection. Firstly, salient region detection is used for segmenting the region of reference hole from the background, and the resultant image of this operation is used for contours retrieving. Secondly, the initial contour of the reference hole is obtained from the retrieved contours using the voting method. Then the initial contour of the reference hole is further refined with the snakes model through energy minimizing of the snake. Finally, the elliptical parameters of the reference hole are computed by fitting an ellipse to the evolving result of the snake. The robustness and accuracy of reference hole detection with respect to noise and environmental disturbance are enhanced significantly through saliency estimation and optimization with the snakes model. Experimental results reveal that the proposed method can be applied to detect reference holes accurately and robustly in the jamming environment of aircraft assembly.
Deep Snake for Real-Time Instance Segmentation

Sida Peng,Jin Wen,Huaijin Pi,Xiuli Li,Hujun Bao,Xiaowei Zhou

DOI: https://doi.org/10.1109/cvpr42600.2020.00856

2020-01-01

Abstract:This paper introduces a novel contour-based approach named deep snake for real-time instance segmentation. Unlike some recent methods that directly regress the coordinates of the object boundary points from an image, deep snake uses a neural network to iteratively deform an initial contour to match the object boundary, which implements the classic idea of snake algorithms with a learning-based approach. For structured feature learning on the contour, we propose to use circular convolution in deep snake, which better exploits the cycle-graph structure of a contour compared against generic graph convolution. Based on deep snake, we develop a two-stage pipeline for instance segmentation: initial contour proposal and contour deformation, which can handle errors in object localization. Experiments show that the proposed approach achieves competitive performances on the Cityscapes, KINS, SBD and COCO datasets while being efficient for real-time applications with a speed of 32.3 fps for 512 × 512 images on a 1080Ti GPU. The code is available at https://github.com/zju3dv/snake/.
Progressive deep snake for instance boundary extraction in medical images

Zixuan Tang,Bin Chen,An Zeng,Mengyuan Liu,Shen Zhao

DOI: https://doi.org/10.1016/j.eswa.2024.123590

IF: 8.5

2024-03-27

Expert Systems with Applications

Abstract:Boundary extraction is meaningful in medical image analysis since it explicitly extracts the tissue/lesions boundary coordinates, which benefits the follow-up diagnosis processes. However, accurately extracting human tissue/lesions is challenging due to unclear target boundaries or pseudo boundaries, which could lead to inferior performance of existing boundary extraction and/or semantic segmentation methods. We propose a Progressive Deep Snake (PDS), which is designed with a multi-step contour evolution (MCE) and a sequential label generation module (SLGM) to explicitly extract the smooth and continuous boundary of different tissue/lesions even if they are unclear, with irregular shape, or occluded by pseudo boundaries. (1) MCE is designed to initialize and evolve the contours to the target boundaries, where a multi-step deep supervision strategy is designed to improve flexibility and accuracy. (2) SLGM is designed to generate easy-to-difficult supervision signals for the contour evolution, which provides targets of moderate difficulty in each step for better guiding the contour evolution. Experiments on a challenging verbal segment datasets show that our method achieves state-of-the-art performance (MR_AVBCE dataset: mIoU: 0.9234, mDice: 0.9598, mBoundF: 0.9241, mWcov: 0.9704). Our code is available at https://github.com/sysu19351118/PDS-pytorch .

computer science, artificial intelligence,engineering, electrical & electronic,operations research & management science
A multi-object segmentation algorithm for snakes using inflating balloon model

Hong-yu LV,Jing-feng MA,Shang-lian BAO

DOI: https://doi.org/10.3321/j.issn:1003-3289.2006.03.040

2006-01-01

Abstract:Objective: To develop an effective scheme for using Snakes or parametric active contours to identify the multi-objects in medical image. Methods: The approach simulates the balloon inflating in a close space containing multi-objects to realize the topological change of the snake. A burning area is utilized to trace the snake curve while the fire front keeps a small distance to the snake contour. Its function is to give the stopping criteria for the snake. Then omitting the stopping criteria, it can form a connected region by evolving the snake and continuing burning. The contours extracted from the burned region represent the child snake of each object respectively. Results: The proposed methods can outline the boundaries of the multi-objects of medical image correctly. Conclusion: The principle of the approach is explicit because its techniques are model based. The algorithm can be implemented easily and is effective for practical applications.
Coronary Artery Edge Detection Based on an Orthogonal T-Snake Model

ZHOU Jie,JIANG Xiaohua,CHENG Liuquan,SHANG Uhua

DOI: https://doi.org/10.3321/j.issn:1000-0054.2009.01.012

2009-01-01

Abstract:Coronary artery edges must be accurately identified for proper diagnosis of heart conditions.This paper presents a T-Snake model based on orthogonal grids to semi-automatically detect the edges of coronary arteries.A mathematical morphology transform is used to remove the fake edges caused by asymmetries of the gray scale in the blood vessel and in the larger tissues.The method then forms the edge contour along the nodes of an orthogonal grid and calculates the contour based on the node energy.The energy function does not include the elastic energy,but include an area energy added as a part of the ‘inflation’ energy.Clinical results with coronary artery images show that this method is effective with both magnetic resonance imaging(MRI) and X-ray computed tomography(CT) images.
A Model for Target Acquisition and Edge Detection under Complex Scenes

Fei Wang,Jihong Zhu

DOI: https://doi.org/10.1504/ijcat.2019.103304

2019-01-01

International Journal of Computer Applications in Technology

Abstract:For the needs of target acquisition and edge extraction under complex scenes, a method based on layer by layer segmentation is proposed. First of all, morphological reconstruction technology is combined with ant colony edge detection method to execute pre-segmentation to get the foreground that targets locate. Secondly, line scanning technique is used to divide foreground into several parts for further segmentation. And then, mean difference characteristic is adopted to determine seed point, with which region growing algorithm is applied to find true targets. Finally, active contour method is used to find out the edge of targets. The experimental results show that our method is effective in finding the target and extracting its edge under complex scenes.
Neighborhood Aided Implicit Active Contours

Huafeng Liu,Yunmei Chen,Wufan Chen

DOI: https://doi.org/10.1109/CVPR.2006.205

2006-01-01

Abstract:We have developed a geometric deformable model that employs neighborhood influence to achieve robust segmentation for noisy and broken edges. The fundamental power of this strategy rests with the explicitly combination of regional inter-point constraints, image forces, and a priori boundary information for each geometric contour point within its adaptively determined local influence domain. This formulation thus naturally unifies the essences of the geometric and parametric snakes through automatic local scale selection, and exhibits their respective fundamental strengths of allowing stable boundary detection when the edge information is weak and possibly discontinuous, while maintaining the abilities to handle topological changes during front evolution. In particular, this paper presents an implementation of the method through local integration of the level set function and the image/prior-driven evolution forces, where the resulting partial differential equation is solved numerically using standard finite difference method. Experimental results on synthetic and real images demonstrate its superior performance.
Active Canny: Edge Detection and Recovery with Open Active Contour Models

Muhammet Bastan,S. Saqib Bukhari,Thomas M. Breuel

DOI: https://doi.org/10.48550/arXiv.1609.03415

2016-09-12

Abstract:We introduce an edge detection and recovery framework based on open active contour models (snakelets). This is motivated by the noisy or broken edges output by standard edge detection algorithms, like Canny. The idea is to utilize the local continuity and smoothness cues provided by strong edges and grow them to recover the missing edges. This way, the strong edges are used to recover weak or missing edges by considering the local edge structures, instead of blindly linking them if gradient magnitudes are above some threshold. We initialize short snakelets on the gradient magnitudes or binary edges automatically and then deform and grow them under the influence of gradient vector flow. The output snakelets are able to recover most of the breaks or weak edges, and they provide a smooth edge representation of the image; they can also be used for higher level analysis, like contour segmentation.

Computer Vision and Pattern Recognition,Multimedia
An Application on an Updated Snake Model in Vehicle Image Tracking System

Qian Jin,Liu Da-cheng,Yang Wen-zhe,Xing Tao,Li Xiao-lu,Yang Qing-ge,Luo Shi-tu

DOI: https://doi.org/10.1117/12.970298

2012-01-01

Abstract:It is a key of all template matching techniques to construct an initial target template. The paper presents an update Snake model, and applies it on the initial template construct process in vehicle image tracking system. The update Snake model is improved mainly on two aspects: one is reconstructing its internal and external energy functions, and the order is adaptively modifying the weights. The update Snake model stretches demand for selecting of initial control points, and quickens the contour searching on a certain extent so as to meet the system's demand for real-time. Besides, the number of contour feature points can adaptively change with the size and complexity of the target, in order to the target contour more precise.
Learned snakes for 3D image segmentation

Lihong Guo,Yueyun Liu,Yu Wang,Yuping Duan,Xue-Cheng Tai

DOI: https://doi.org/10.1016/j.sigpro.2021.108013

IF: 4.729

2021-06-01

Signal Processing

Abstract:<p>Snakes or active contour models are classical methods for boundary detection and segmentation, which deform an initial contour (for 2D image) or a surface (for 3D image) towards the boundary of the desired object. Such snakes models are ideal choices for handling medical image segmentation problems since they are very efficient and require fewer memories by solely tracking the explicit curves or surfaces. However, traditional snakes models solved by the level set method suffer from numerical instabilities and are usually difficult to deal with topological changes. In this paper, we propose a learned snakes model for 3D medical image segmentation, where both the initial and final surfaces are estimated using deep neural networks in end-to-end regimes. The merit of our learned snakes model is that we can realize 3D segmentation by finding a 2D surface based on 2D convolutional neural networks rather than using 3D networks or cutting the volume into 2D slices. Experiments on the Medical Segmentation Decathlon spleen dataset against both 2D- and 3D-based networks demonstrate our model achieving the state-of-the-art accuracy and efficiency, which not only enjoys a 1% higher DSC but also saves more than 90% computational time compared to the well-established elastic boundary projection model [1].</p>

engineering, electrical & electronic
An Edge Detection Model Based on W-Transform Single Pixel Imaging

Guangyao Chen,Yuanping Xu,Chao Kong,Zhijie Xu,Yanlong Cao,Kaiwei Wang,Chaolong Zhang

DOI: https://doi.org/10.1109/icoim60566.2023.10491524

2023-01-01

Abstract:To efficiently reduce the required measurement count for single-pixel imaging edge detection, this study devises an edge detection model based on W-transform single-pixel imaging. In this model, the convolution of the DoG (Difference of Gaussians) operator and W-transform basis patterns as modulation patterns are employed to directly extract edge information from the spectral domain of the target object. Different from traditional methods, it achieves the extraction of edge details from an object without imaging. Numerical simulations and experimental results demonstrate that this model extracts edges with higher signal-to-noise ratios compared to the edge extraction using phase-shifted sinusoidal patterns. Additionally, the proposed model reduces the number of modulation patterns required by half, so as to gain double efficiency. The integration of W-transform single-pixel imaging with edge detection offers a novel approach for edge detections without the need for imaging.
An Extended Image Force Model of Snakes for Medical Image Segmentation and Smoothing

Hongyu Lu,Shanglian Bao

DOI: https://doi.org/10.1109/icosp.2006.345613

2006-01-01

Abstract:A scheme of the image force for Snakes, named the eSnake Model, has been introduced based on the electrostatic theory. To improve the ability for segmenting medical images with uneven grayscale intensity, a novel extended eSnake model, eeSnake, is developed to get the desirable image force distribution for Snakes. Breaking through the limitation of the Coulomb law, it can work better with the elasticity and balloon forces of Snakes than eSnake. The pre-calculated data stencil of distance functions is adopted to accelerate the computing speed of the potential field. The analysis points out that the eeSnake model can also function as a smoothing filter. Experiments show that the proposed eeSnake model is satisfactory in both segmentation and noise reduction applications
Region Labeling Algorithm Based on Boundary Tracking for Binary Image

Chen Li,Yang,Cen Zhaofeng,Li Xiaotong

DOI: https://doi.org/10.1117/12.869443

2010-01-01

Abstract:Region labeling for binary image is an important part of image processing. For the special use of small and multi-objects labeling, a new region labeling algorithm based on boundary tracking is proposed in this paper. Experiments prove that our algorithm is feasible and efficient, and even faster than some of other algorithms.
A Segmentation and Inpainting Algorithm for Overlapped Cells Based on Cell Scatter Plot and Modified Snake

ZOU Gang,YAO Wei,AO Yong-Hong

DOI: https://doi.org/10.3969/j.issn.0258-8021.2010.01.007

2010-01-01

Abstract:The overlapping phenomenon often appears in the medical microscopic images.A segmentation algorithm for overlapped cell images based on cell scatter plot and modified Snake was proposed.Firstly,an iterative corrosion method with a dynamic corrosion structure was used to get cell scatter plot image and separation cell image,then the initialization outline was obtained by the edge information.Finally,the modified Snake algorithm with constraint energy was performed to get the cell overlapped boundary.In order to get the mask image based on the overlapped region,a fast image inpainting algorithm was presented based on adaptive iterative convolution to solve the information deterioration resulted from overlapping.Experiment results showed that the proposed algorithm was effective for automatic cell image segmentation and available in getting the cell boundary.The inpainting for the cell overlapped plot was rapid on the basis of mask image.
Edge Detection Algorithm Based on Valley Lines and Watershed

Bo Wang,Zhiying Yang

DOI: https://doi.org/10.1088/1742-6596/2010/1/012034

2021-01-01

Journal of Physics Conference Series

Abstract:The purpose of edge detection algorithm is to highlight the edge in the image, which is also the premise and key of image segmentation. However, the traditional edge detection algorithm will amplify the influence of noise in the process of derivative operation, and the noisy edges will often appear, which is not conducive to the accurate extraction of edge. To solve this problem, we propose an edge detection algorithm combining valley lines and watershed algorithm. In this algorithm, the bilateral filter is used to replace the Gaussian filter to achieve edge preserving and denoising. The rewritten Canny operator is used to calculate the low value points and get the valley lines of the image. This process can achieve the first denoising. Then, the valley lines are used to replace the discrete points of water injection and the watershed algorithm is used to calculate the segmentation results of the image. At this time, there are many pseudo edges around the main boundary, and the pseudo edges are deleted by setting the threshold, so as to achieve the effect of secondary denoising. The experimental results show that our method has good performance in noise resistance and edge continuity compared with other and traditional edge detection algorithms.
Optimization of Edge Extraction Algorithm for Objects in Complex Background

Fengyun Huang,Yu Jiang

DOI: https://doi.org/10.1088/1742-6596/1880/1/012027

2021-04-01

Journal of Physics: Conference Series

Abstract:Abstract In the field of artificial intelligence, machine vision is expected because of its low cost and easy popularization. Feature extraction is one of the core steps of machine vision. When we use Canny algorithm, which extracts edges by gray level change, we cannot directly separate the object from the background. The edge information of the object will also be lost when the Gaussian filter smoothes the image. The research proposes an improved edge extraction algorithm based on Canny algorithm. We use the image difference method and object envelopment method to separate the object from the background, so the data to be processed is greatly reduced and the speed and accuracy of the algorithm are improved. Then we use bilateral filtering instead of Gaussian filtering, which increases the weight of gray difference on the basis of Gaussian filtering, so we can to retain more edge information. Finally, we improve the gradient operator, by determining the gradient operator coefficient according to the inverse ratio of the Euclidean distance between the pixel and the center pixel. The average performance of the new gradient operator in the center of the picture is better than that of Sobel operator. Experimental simulation shows that the algorithm has good detection accuracy. Compared with traditional algorithms, the improved algorithm is not affected by complex backgrounds, and reduce the influence of light. The improved algorithm has clear edges in the center area of the image.

Image Boundary Detection Algorithm Based on Snake Model

A New Algorithm To Detect Edge For Infrared Image Based On Two-Dimension Continuous Wavelet Transform

Initial Model of Snake and Improved Approaches

Application of Snake Model for Edge Extraction Breast Tumor in Ultrasound Images

Face Detection under Complex Background

A New Elliptic Contour Extraction Method for Reference Hole Detection in Robotic Drilling

Deep Snake for Real-Time Instance Segmentation

Progressive deep snake for instance boundary extraction in medical images

A multi-object segmentation algorithm for snakes using inflating balloon model

Coronary Artery Edge Detection Based on an Orthogonal T-Snake Model

A Model for Target Acquisition and Edge Detection under Complex Scenes

Neighborhood Aided Implicit Active Contours

Active Canny: Edge Detection and Recovery with Open Active Contour Models

An Application on an Updated Snake Model in Vehicle Image Tracking System

Learned snakes for 3D image segmentation

An Edge Detection Model Based on W-Transform Single Pixel Imaging

An Extended Image Force Model of Snakes for Medical Image Segmentation and Smoothing

Region Labeling Algorithm Based on Boundary Tracking for Binary Image

A Segmentation and Inpainting Algorithm for Overlapped Cells Based on Cell Scatter Plot and Modified Snake

Edge Detection Algorithm Based on Valley Lines and Watershed

Optimization of Edge Extraction Algorithm for Objects in Complex Background