This paper presents comparisons of two learning-based super-resolution algorithms as well as standard interpolation methods. To allow quality assessment of results, a comparison of a variety of image quality measures is also performed. Results show that a MRF-based super-resolution algorithm improves a previously interpolated image. The estimated degree of improvement varies both according to the quality measure chosen for the comparison as well as the image class.
{"title":"Comparison of Super-Resolution Algorithms Using Image Quality Measures","authors":"I. Bégin, F. Ferrie","doi":"10.1109/CRV.2006.23","DOIUrl":"https://doi.org/10.1109/CRV.2006.23","url":null,"abstract":"This paper presents comparisons of two learning-based super-resolution algorithms as well as standard interpolation methods. To allow quality assessment of results, a comparison of a variety of image quality measures is also performed. Results show that a MRF-based super-resolution algorithm improves a previously interpolated image. The estimated degree of improvement varies both according to the quality measure chosen for the comparison as well as the image class.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130502271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Onay Urfahuglo, Thorsten Thormählen, Hellward Broszio, Patrick Mikulastik
Feature points for camera parameter estimation are detected in noisy images. Therefore, the feature points and also the camera parameters can only be estimated with limited accuracy. In case of collinear feature points, it is possible to benefit from this geometrical regularity which results in an increased accuracy of the camera parameters. In this paper, a complete theoretical covariance propagation starting from the error of the feature points up to the error of the estimated camera parameters is performed. Additionally, by determining the Fisher information matrix the Cramer-Rao bounds for the covariance of the corrected feature point positions are determined. To demonstrate the impact of collinearity on the accuracy of the camera parameters, a covariance propagation is performed with varying feature point error covariances.
{"title":"Error Analysis of Camera Parameter Estimation based on Collinear Features","authors":"Onay Urfahuglo, Thorsten Thormählen, Hellward Broszio, Patrick Mikulastik","doi":"10.1109/CRV.2006.30","DOIUrl":"https://doi.org/10.1109/CRV.2006.30","url":null,"abstract":"Feature points for camera parameter estimation are detected in noisy images. Therefore, the feature points and also the camera parameters can only be estimated with limited accuracy. In case of collinear feature points, it is possible to benefit from this geometrical regularity which results in an increased accuracy of the camera parameters. In this paper, a complete theoretical covariance propagation starting from the error of the feature points up to the error of the estimated camera parameters is performed. Additionally, by determining the Fisher information matrix the Cramer-Rao bounds for the covariance of the corrected feature point positions are determined. To demonstrate the impact of collinearity on the accuracy of the camera parameters, a covariance propagation is performed with varying feature point error covariances.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134081228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A new method of video object extraction is proposed to accurately obtain the object of interest from actively acquired videos. Traditional video object extraction techniques often operate under the assumption of homogeneous object motion and extract various parts of the video that are motion consistent as objects. In contrast, the proposed active video object extraction (AVOE) paradigm assumes that the object of interest is being actively tracked by a non-calibrated camera under general motion and classifies the possible movements of the camera that result in the 2D motion patterns as recovered from the image sequence. Consequently, the AVOE method is able to extract the single object of interest from the active video. We formalize the AVOE process using notions from Gestalt psychology. We define a new Gestalt factor called "shift and hold" and present 2D object extraction algorithms. Moreover, since an active video sequence naturally contains multiple views of the object of interest, we demonstrate that these views can be combined to form a single 3D object regardless of whether the object is static or moving in the video.
{"title":"Object Extraction and Reconstruction in Active Video","authors":"Ye Lu, Ze-Nian Li","doi":"10.1109/CRV.2006.53","DOIUrl":"https://doi.org/10.1109/CRV.2006.53","url":null,"abstract":"A new method of video object extraction is proposed to accurately obtain the object of interest from actively acquired videos. Traditional video object extraction techniques often operate under the assumption of homogeneous object motion and extract various parts of the video that are motion consistent as objects. In contrast, the proposed active video object extraction (AVOE) paradigm assumes that the object of interest is being actively tracked by a non-calibrated camera under general motion and classifies the possible movements of the camera that result in the 2D motion patterns as recovered from the image sequence. Consequently, the AVOE method is able to extract the single object of interest from the active video. We formalize the AVOE process using notions from Gestalt psychology. We define a new Gestalt factor called \"shift and hold\" and present 2D object extraction algorithms. Moreover, since an active video sequence naturally contains multiple views of the object of interest, we demonstrate that these views can be combined to form a single 3D object regardless of whether the object is static or moving in the video.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126836495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Georegistration of digital elevation maps is a vital step in fusing sensor data. In this paper, we present an entropic registration method using Morse singularities. The core idea behind our proposed approach is to encode an elevation map into a set of Morse singular points. Then an information-theoretic dissimilarity measure between the Morse features of the target and the reference maps is maximized to bring the elevation data into alignment. We also show that maximizing this divergence measure leads to minimizing the total length of the joint minimal spanning tree of both elevation data maps. Illustrating experimental results are presented to show the robustness and the georegistration accuracy of the proposed approach.
{"title":"An Information-Theoretic Approach to Georegistration of Digital Elevation Maps","authors":"Miguel Aguilera, A. Hamza","doi":"10.1109/CRV.2006.11","DOIUrl":"https://doi.org/10.1109/CRV.2006.11","url":null,"abstract":"Georegistration of digital elevation maps is a vital step in fusing sensor data. In this paper, we present an entropic registration method using Morse singularities. The core idea behind our proposed approach is to encode an elevation map into a set of Morse singular points. Then an information-theoretic dissimilarity measure between the Morse features of the target and the reference maps is maximized to bring the elevation data into alignment. We also show that maximizing this divergence measure leads to minimizing the total length of the joint minimal spanning tree of both elevation data maps. Illustrating experimental results are presented to show the robustness and the georegistration accuracy of the proposed approach.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126976168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The hierarchical segmentation of SAR (Synthetic Aperture Radar) images is greatly complicated by the presence of coherent speckle. We are exploring the utilization of spatial constraints and contour shapes in order to improve the segmentation results. With standard merging criterion, the high noise level of SAR images results in the production of regions that have variable mean and variance values and irregular shapes. If the first segments are not correctly delimited then the following steps will merge segments from different fields. In examining the evolution of the initial segments, we see that the merging should take into account spatial aspects. Particularly, the segment contours should have good shapes. In this paper, we examine how the pseudo-convex envelope of a region can be used to evaluate the region contour. We present a pseudo-convex measure adapted to the geometry of image lattice. We show how the pseudo-convex envelope can be calculated. We present measures comparing contour shapes and using the perimeter, the area and the boundary length of segments. We use a hierarchical segmentation algorithm based upon stepwise optimization. A stepwise merging criterion is derived from the multiplicative speckle noise model. The shape measures are combined with the merging criterion in order to guide correctly the segment merging process. The new criterion produces good segmentation of SAR images. This is illustrated by synthetic and real image results.
{"title":"Pseudo-convex Contour Criterion for Hierarchical Segmentation of SAR Images","authors":"Jean-Marie Beaulieu","doi":"10.1109/CRV.2006.58","DOIUrl":"https://doi.org/10.1109/CRV.2006.58","url":null,"abstract":"The hierarchical segmentation of SAR (Synthetic Aperture Radar) images is greatly complicated by the presence of coherent speckle. We are exploring the utilization of spatial constraints and contour shapes in order to improve the segmentation results. With standard merging criterion, the high noise level of SAR images results in the production of regions that have variable mean and variance values and irregular shapes. If the first segments are not correctly delimited then the following steps will merge segments from different fields. In examining the evolution of the initial segments, we see that the merging should take into account spatial aspects. Particularly, the segment contours should have good shapes. In this paper, we examine how the pseudo-convex envelope of a region can be used to evaluate the region contour. We present a pseudo-convex measure adapted to the geometry of image lattice. We show how the pseudo-convex envelope can be calculated. We present measures comparing contour shapes and using the perimeter, the area and the boundary length of segments. We use a hierarchical segmentation algorithm based upon stepwise optimization. A stepwise merging criterion is derived from the multiplicative speckle noise model. The shape measures are combined with the merging criterion in order to guide correctly the segment merging process. The new criterion produces good segmentation of SAR images. This is illustrated by synthetic and real image results.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129916437","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent years, region-based active contour models have gained great popularity in solving image segmentation problem. Those models usually share two assumptions regarding the image pixel properties: 1) within each region/ object, the intensity values conform to a Gaussian distribution; 2) the "global mean" (average intensity value) for different regions are distinct, therefore can be used in discriminating pixels. These two assumptions are often violated in reality, which results in segmentation leakage or misclassification. In this paper, we propose a robust segmentation framework that overcomes the above mentioned drawback existing in most region-based active contour models. Our framework consists of two components: 1) instead of using a global average intensity value (mean) to represent certain region, we use local medians as the region representative measure to better characterize the local property of the image; 2) median and sum of absolute values (L1 norm) is used to formulate the energy minimization functional for better handling intensity variations and outliers. Experiments are conducted on several real images, and we compare our solution with a popular region-based model to show the improvements.
{"title":"Robust Image Segmentation using Local Median","authors":"Jundong Liu","doi":"10.1109/CRV.2006.60","DOIUrl":"https://doi.org/10.1109/CRV.2006.60","url":null,"abstract":"In recent years, region-based active contour models have gained great popularity in solving image segmentation problem. Those models usually share two assumptions regarding the image pixel properties: 1) within each region/ object, the intensity values conform to a Gaussian distribution; 2) the \"global mean\" (average intensity value) for different regions are distinct, therefore can be used in discriminating pixels. These two assumptions are often violated in reality, which results in segmentation leakage or misclassification. In this paper, we propose a robust segmentation framework that overcomes the above mentioned drawback existing in most region-based active contour models. Our framework consists of two components: 1) instead of using a global average intensity value (mean) to represent certain region, we use local medians as the region representative measure to better characterize the local property of the image; 2) median and sum of absolute values (L1 norm) is used to formulate the energy minimization functional for better handling intensity variations and outliers. Experiments are conducted on several real images, and we compare our solution with a popular region-based model to show the improvements.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125402101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We are developing a helper robot that carries out tasks ordered by users through speech. The robot needs a vision system to recognize the objects appearing in the orders. It is, however, difficult to realize vision systems that can work in various conditions. They may find many objects and cannot determine which is the target. We have proposed a method of using a conversation with the user to solve this problem. The robot asks questions which the user can easily answer and whose answer can efficiently reduce the number of candidate objects. In our previous system, however, we assumed that there was no occlusion among objects. This paper presents an extended system that can detect target objects in occlusion cases. It is difficult to obtain some features in occlusion cases. To compensate the system for this shortage of features, we propose to use reference systems to express the positional relationships of objects. Experimental results show that the robot can efficiently detect objects through user-friendly conversation.
{"title":"Use of Spatial Reference Systems in Interactive Object Recognition","authors":"R. Kurnia, Md. Altab Hossain, Y. Kuno","doi":"10.1109/CRV.2006.82","DOIUrl":"https://doi.org/10.1109/CRV.2006.82","url":null,"abstract":"We are developing a helper robot that carries out tasks ordered by users through speech. The robot needs a vision system to recognize the objects appearing in the orders. It is, however, difficult to realize vision systems that can work in various conditions. They may find many objects and cannot determine which is the target. We have proposed a method of using a conversation with the user to solve this problem. The robot asks questions which the user can easily answer and whose answer can efficiently reduce the number of candidate objects. In our previous system, however, we assumed that there was no occlusion among objects. This paper presents an extended system that can detect target objects in occlusion cases. It is difficult to obtain some features in occlusion cases. To compensate the system for this shortage of features, we propose to use reference systems to express the positional relationships of objects. Experimental results show that the robot can efficiently detect objects through user-friendly conversation.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"312 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122814113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Nomad 200 and the Nomad SuperScouts are among the most popular platforms used, for research in robotics. Built in the early 1990’s they were the base of choice for many mobile robotics researchers. Unfortunately, lack of support and proper documentation to enhance the computing power on these robots has meant that they have slowly faded away into oblivion. We at York University have four robots from Nomadic Technologies Inc., and rather than allowing our old robots to just rust away, we decided to breathe new life back into them. In this paper we present the techniques we used to resurrect our old Nomads.
{"title":"The Nomad 200 and the Nomad SuperScout: Reverse engineered and resurrected","authors":"A. Chopra, Mark Obsniuk, M. Jenkin","doi":"10.1109/CRV.2006.76","DOIUrl":"https://doi.org/10.1109/CRV.2006.76","url":null,"abstract":"The Nomad 200 and the Nomad SuperScouts are among the most popular platforms used, for research in robotics. Built in the early 1990’s they were the base of choice for many mobile robotics researchers. Unfortunately, lack of support and proper documentation to enhance the computing power on these robots has meant that they have slowly faded away into oblivion. We at York University have four robots from Nomadic Technologies Inc., and rather than allowing our old robots to just rust away, we decided to breathe new life back into them. In this paper we present the techniques we used to resurrect our old Nomads.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"136 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114753958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An efficient algorithm for detecting duplicate regions is proposed in this paper. The basic idea is to segment the input image into blocks and search for blocks with similar intensity patterns using matching techniques. To improve the efficiency, the blocks are sorted based on the concept of k-dimensional tree. The sorting process groups blocks with similar patterns and hence the number of matching operations required for finding the duplicated blocks can be significantly reduced. The matching block detection results are encoded as a color image. This makes it possible to use a set of colour-based morphological operations to remove isolated mismatches, as well as to fill in missing matches. The experiments conducted show the effectiveness of the proposed algorithm.
{"title":"An Efficient Match-based Duplication Detection Algorithm","authors":"A. Langille, Minglun Gong","doi":"10.1109/CRV.2006.9","DOIUrl":"https://doi.org/10.1109/CRV.2006.9","url":null,"abstract":"An efficient algorithm for detecting duplicate regions is proposed in this paper. The basic idea is to segment the input image into blocks and search for blocks with similar intensity patterns using matching techniques. To improve the efficiency, the blocks are sorted based on the concept of k-dimensional tree. The sorting process groups blocks with similar patterns and hence the number of matching operations required for finding the duplicated blocks can be significantly reduced. The matching block detection results are encoded as a color image. This makes it possible to use a set of colour-based morphological operations to remove isolated mismatches, as well as to fill in missing matches. The experiments conducted show the effectiveness of the proposed algorithm.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124163508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A set of microarray images were acquired by a sequence of biological experiments which were scanned via a high resolution scanner. For each spot corresponding to a gene, the ratio of Cy3 and Cy5 fluorescent signal intensities was obtained and which may be normalied based on piecewise linear regression such as lowess method. In this study, we computed from 55 microarray images to get an M × N genematrix, A, with N = 55 patients and M = 13574 effected genes in each microarray. We start with our gene discovery from a genematrix A epsillon R^M×N, M = 13574, N = 55, including N1 = 29 patients of hepatitis B virus (HBV), N2 = 21 patients of hepatitis C virus (HCV), 1 patient clinically diagnosed to be infected with HCV as well as HBV, and 4 patients were neither HCV nor HBV infected. Simple software was developed to solve the following problems: (i) Detect differentially expressed genes and (ii) Select a subset of genes which best distinguishes HBV patients from HCV ones.
{"title":"Simple Software for Microarray Image Analysis","authors":"Chaur-Chin Chen, Cheng-Yan Kao, Chun-Fan Chang, Hsueh-Ting Chu, Chiung-Nien Chen","doi":"10.1109/CRV.2006.65","DOIUrl":"https://doi.org/10.1109/CRV.2006.65","url":null,"abstract":"A set of microarray images were acquired by a sequence of biological experiments which were scanned via a high resolution scanner. For each spot corresponding to a gene, the ratio of Cy3 and Cy5 fluorescent signal intensities was obtained and which may be normalied based on piecewise linear regression such as lowess method. In this study, we computed from 55 microarray images to get an M × N genematrix, A, with N = 55 patients and M = 13574 effected genes in each microarray. We start with our gene discovery from a genematrix A epsillon R^M×N, M = 13574, N = 55, including N1 = 29 patients of hepatitis B virus (HBV), N2 = 21 patients of hepatitis C virus (HCV), 1 patient clinically diagnosed to be infected with HCV as well as HBV, and 4 patients were neither HCV nor HBV infected. Simple software was developed to solve the following problems: (i) Detect differentially expressed genes and (ii) Select a subset of genes which best distinguishes HBV patients from HCV ones.","PeriodicalId":369170,"journal":{"name":"The 3rd Canadian Conference on Computer and Robot Vision (CRV'06)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131602038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：