Object Extraction Using Topological Models from Complex Scene Image

2017 ◽  
pp. 335-357
Author(s):  
Uday Pratap Singh ◽  
Sanjeev Jain
Keyword(s):  
Object Recognition ◽  
Object Segmentation ◽  
Multimedia Data ◽  
Region Merging ◽  
Complex Scene ◽  
Object Contour ◽  
Merging Process ◽  
Feasible Solutions ◽  
Neighbourhood Graph ◽  
Topological Models

Efficient and effective object recognition from a multimedia data are very complex. Automatic object segmentation is usually very hard for natural images; interactive schemes with a few simple markers provide feasible solutions. In this chapter, we propose topological model based region merging. In this work, we will focus on topological models like, Relative Neighbourhood Graph (RNG) and Gabriel graph (GG), etc. From the Initial segmented image, we constructed a neighbourhood graph represented different regions as the node of graph and weight of the edges are the value of dissimilarity measures function for their colour histogram vectors. A method of similarity based region merging mechanism (supervised and unsupervised) is proposed to guide the merging process with the help of markers. The region merging process is adaptive to the image content and it does not need to set the similarity threshold in advance. To the validation of proposed method extensive experiments are performed and the result shows that the proposed method extracts the object contour from the complex background.

Object Segmentation Based on a Nonparametric Snake with Motion Prediction in Video

2012 ◽  
pp. 86-107
Author(s):  
Sang-Myoung Ye ◽  
Rae-Hong Park ◽  
Dong-Kyu Lee
Keyword(s):  
Object Segmentation ◽  
Computation Time ◽  
Absolute Difference ◽  
Motion Prediction ◽  
Surveillance Systems ◽  
Gradient Vector ◽  
Snake Model ◽  
Object Contour ◽  
The Mean ◽  
Initial Object

Object segmentation in video sequence is a basic and important task in video applications such as surveillance systems and video coding. Nonparametric snake algorithms for object segmentation have been proposed to overcome the drawback of conventional snake algorithms: the dependency on several parameters. In this chapter, a new object segmentation algorithm for video, based on a nonparametric snake model with motion prediction, is proposed. Object contour is initialized by using the mean absolute difference of intensity between input and previous frames. And in order to convert initial object contours into more exact object contours, the gradient vector flow snake is used. Finally object contour is predicted using a Kalman filter in successive frames. The proposed object segmentation method for video can provide more detailed and improved object segmentation results than the conventional methods. Various experimental results show the effectiveness of the proposed method in terms of the pixel-based quality measure and the computation time.

Content-Aware Image Retargeting Based on Semantic Analysis

2014 ◽  
Vol 577 ◽  
pp. 777-781 ◽  
Author(s):  
Hao Cheng ◽  
Wei Li ◽  
Pei Min Zhong
Keyword(s):  
Object Recognition ◽  
Energy Loss ◽  
Energy Function ◽  
Semantic Analysis ◽  
Triangular Mesh ◽  
Image Retargeting ◽  
Complex Scene ◽  
Important Object ◽  
Least Energy ◽  
Content Aware

This paper presents a retargeting approach based on semantic analysis. Our approach has better performance in images that have comlicated backgroud or multi-objects. Our aim is to protect important object in retargeting process. So we brought in object recognition and image importance calculation for retargeting. This approach consists of three parts: object recognition, the importance of image calculation and image retargeting. (i) Firstly we optimized semantic texton forest (STF)[11]and got much better results of object recognition.(ii) Secondly we presented a method called dynamically adjust importance image calculation. (iii) Thirdly we give a retargeting method based on triangular mesh. According to importance image we cover Delaunay triangular mesh on image and solve optimization mesh transforming based on energy function which subjects to least energy loss and boundary restraint. Compared with previous approaches, our method has better result in some complex scene images.

scholarly journals Iterated Region for Interactive Image Segmentation on Dental Panoramic Radiograph

2019 ◽  
Vol 12 (1) ◽  
pp. 13
Author(s):  
Biandina Meidyani ◽  
Lailly S. Qolby ◽  
Ahmad Miftah Fajrin ◽  
Agus Zainal Arifin ◽  
Dini Adni Navastara
Keyword(s):  
Image Segmentation ◽  
Final Analysis ◽  
Panoramic Radiograph ◽  
Region Merging ◽  
Low Contrast ◽  
Analysis Process ◽  
Merging Process ◽  
Dental Panoramic Radiograph ◽  
Background Segmentation ◽  
Radiograph Image

Image Segmentation is a process to separate between foreground and background. Segmentation process in low contrast image such as dental panoramic radiograph image is not easily determined. Image segmentation accuracy determines the success or failure of the final analysis process. The process of segmentation can occur ambiguity. This ambiguity is due to an ambiguous area if it is not selected as a region so it may have occurred cluster errors. To solve this ambiguity, we proposed a new region merging by iterated region merging process on dental panoramic radiograph image. The proposed method starts from the user marking and works iteratively to label the surrounding regions. In each iteration, the minimal gray-levels value is merged so the unknown regions significantly reduced. This experiment shows that the proposed method is effective with an average of ME and RAE of 0.04% and 0.06%.

scholarly journals MULTISCALE SEGMENTATION OF POLARIMETRIC SAR IMAGE BASED ON SRM SUPERPIXELS

2016 ◽  
Vol XLI-B7 ◽  
pp. 283-287
Author(s):  
F. Lang ◽  
J. Yang ◽  
L. Wu ◽  
D. Li
Keyword(s):  
Object Oriented ◽  
Test Site ◽  
Region Merging ◽  
Polarimetric Sar ◽  
Multi Scale ◽  
Single Scale ◽  
Binary Partition ◽  
Segmentation Algorithms ◽  
Merging Process ◽  
Statistical Region Merging

Multi-scale segmentation of remote sensing image is more systematic and more convenient for the object-oriented image analysis compared to single-scale segmentation. However, the existing pixel-based polarimetric SAR (PolSAR) image multi-scale segmentation algorithms are usually inefficient and impractical. In this paper, we proposed a superpixel-based binary partition tree (BPT) segmentation algorithm by combining the generalized statistical region merging (GSRM) algorithm and the BPT algorithm. First, superpixels are obtained by setting a maximum region number threshold to GSRM. Then, the region merging process of the BPT algorithm is implemented based on superpixels but not pixels. The proposed algorithm inherits the advantages of both GSRM and BPT. The operation efficiency is obviously improved compared to the pixel-based BPT segmentation. Experiments using the Lband ESAR image over the Oberpfaffenhofen test site proved the effectiveness of the proposed method.

Automatic RGBD Object Segmentation Based on MSRM Framework Integrating Depth Value

2020 ◽  
Vol 29 (07n08) ◽  
pp. 2040009
Author(s):  
Guoqing Li ◽  
Guoping Zhang ◽  
Chanchan Qin ◽  
Anqin Lu
Keyword(s):  
Object Segmentation ◽  
Natural Image ◽  
Depth Information ◽  
Region Merging ◽  
Accurate Localization ◽  
Complex Image ◽  
Depth Cue ◽  
Simple Linear Iterative Clustering ◽  
Rgb Images ◽  
Depth Feature

In this paper, an automatic RGBD object segmentation method is described. The method integrates depth feature with the cues from RGB images and then uses maximal similarity based region merging (MSRM) method to obtain the segmentation results. Firstly, the depth information is fused to the simple linear iterative clustering (SLIC) method so as to produce superpixels whose boundaries are well adhered to the edges of the natural image. Meanwhile, the depth prior is also incorporated into the saliency estimation, which helps a more accurate localization of representative object and background seeds. By introducing the depth cue into the region merging rule, the maximal geometry weighted similarity (MGWS) is considered, and the resulting segmentation framework has the ability to handle the complex image with similar colour appearance between object and background. Extensive experiments on public RGBD image datasets show that our proposed approach can reliably and automatically provide very promising segmentation results.

scholarly journals How to Generate Image Dataset based on 3D Model and Deep Learning Method

2018 ◽  
Vol 7 (3.34) ◽  
pp. 221
Author(s):  
Sooyoung Cho ◽  
Sang Geun Choi ◽  
Daeyeol Kim ◽  
Gyunghak Lee ◽  
Chae BongSohn
Keyword(s):  
Computer Vision ◽  
Deep Learning ◽  
Object Recognition ◽  
Object Tracking ◽  
3D Model ◽  
Object Segmentation ◽  
Learning Method ◽  
Data Set ◽  
Image Dataset ◽  
Deep Learning Model

Performances of computer vision tasks have been drastically improved after applying deep learning. Such object recognition, object segmentation, object tracking, and others have been approached to the super-human level. Most of the algorithms were trained by using supervised learning. In general, the performance of computer vision is improved by increasing the size of the data. The collected data was labeled and used as a data set of the YOLO algorithm. In this paper, we propose a data set generation method using Unity which is one of the 3D engines. The proposed method makes it easy to obtain the data necessary for learning. We classify 2D polymorphic objects and test them against various data using a deep learning model. In the classification using CNN and VGG-16, 90% accuracy was achieved. And we used Tiny-YOLO of YOLO algorithm for object recognition and we achieved 78% accuracy. Finally, we compared in terms of virtual and real environments it showed a result of 97 to 99 percent for each accuracy.

Study on Shape Recognition Based on Contour

2014 ◽  
Vol 644-650 ◽  
pp. 4240-4243
Author(s):  
Wei Zhang
Keyword(s):  
Feature Extraction ◽  
Object Recognition ◽  
Fourier Transformation ◽  
Recursive Algorithm ◽  
Shape Recognition ◽  
Experimental Results ◽  
Discrete Fourier Transformation ◽  
Object Contour ◽  
Important Character

Shape is the inherence characteristic of an object in the image, and it is the important character used for the object recognition. So it is significant for object recognition based on shape. This paper presents a contour-based method of feature extraction and shape recognition. First the object contour is translated into a 1-D contour curve. Secondly the curve is smoothed to restrain the noise. The number of peaks of the curve is achieved as well as the areas which contained between adjacent peak-valley, then the latter is followed by Discrete Fourier Transformation (DFT). Then two kinds of features ate extracted which are invariant to translation, scaling and rotation transformations. By using the features, a two-stake recursive algorithm for recognition is proposed. Experimental results show that this method is simple and efficient.

Sign in / Sign up

Export Citation Format

Share Document