Ponències/Comunicacions de congressos
http://hdl.handle.net/2117/3335
2017-02-26T06:07:23ZImage restoration using HOS and the Radon transform
http://hdl.handle.net/2117/101571
Image restoration using HOS and the Radon transform
Sayrol Clols, Elisa; Chrysostomos, Nikias; Gasull Llampallas, Antoni
The authors propose the use of higher-order statistics (HOS) to study the problem of image restoration. They consider images degraded by linear or zero phase blurring point spread functions (PSF) and additive Gaussian noise. The complexity associated with the combination of two-dimensional signal processing and higher-order statistics is reduced by means of the Radon transform. The projection at each angle is an one-dimensional signal that can be processed by any existing 1-D higher-order statistics-based method. They apply two methods that have proven to attain good one-dimensional signal reconstruction, especially in the presence of noise. After the ideal projections have been estimated, the inverse Radon transform gives the restored image. Simulation results are provided.
2017-02-24T17:05:43ZSayrol Clols, ElisaChrysostomos, NikiasGasull Llampallas, AntoniThe authors propose the use of higher-order statistics (HOS) to study the problem of image restoration. They consider images degraded by linear or zero phase blurring point spread functions (PSF) and additive Gaussian noise. The complexity associated with the combination of two-dimensional signal processing and higher-order statistics is reduced by means of the Radon transform. The projection at each angle is an one-dimensional signal that can be processed by any existing 1-D higher-order statistics-based method. They apply two methods that have proven to attain good one-dimensional signal reconstruction, especially in the presence of noise. After the ideal projections have been estimated, the inverse Radon transform gives the restored image. Simulation results are provided.Unsupervised morphological segmentation for images
http://hdl.handle.net/2117/101488
Unsupervised morphological segmentation for images
Salembier Clairon, Philippe Jean
This paper deals with a morphological approach to unsupervised image segmentation. The proposed technique relies on a multiscale Top-Down approach allowing a hierarchical processing of the data ranging from the most global scale to the most detailed one. At each scale, the algorithm consists of four steps: image simplification, feature extraction, contour localization and quality estimation. The main emphasis of this paper is to discuss the selection of a simplification filter for segmentation. Morphological filters based on reconstruction proved to be very efficient for this purpose. The resulting unsupervised algorithm is very robust and can deal with very different type of images.
2017-02-23T16:19:23ZSalembier Clairon, Philippe JeanThis paper deals with a morphological approach to unsupervised image segmentation. The proposed technique relies on a multiscale Top-Down approach allowing a hierarchical processing of the data ranging from the most global scale to the most detailed one. At each scale, the algorithm consists of four steps: image simplification, feature extraction, contour localization and quality estimation. The main emphasis of this paper is to discuss the selection of a simplification filter for segmentation. Morphological filters based on reconstruction proved to be very efficient for this purpose. The resulting unsupervised algorithm is very robust and can deal with very different type of images.Image restoration using HOS and the Radon transform
http://hdl.handle.net/2117/101471
Image restoration using HOS and the Radon transform
Sayrol Clols, Elisa; Nikias, C L; Gasull Llampallas, Antoni
The authors propose the use of higher-order statistics (HOS) to study the problem of image restoration. They consider images degraded by linear or zero phase blurring point spread functions (PSF) and additive Gaussian noise. The complexity associated with the combination of two-dimensional signal processing and higher-order statistics is reduced by means of the Radon transform. The projection at each angle is an one-dimensional signal that can be processed by any existing 1-D higher-order statistics-based method. They apply two methods that have proven to attain good one-dimensional signal reconstruction, especially in the presence of noise. After the ideal projections have been estimated, the inverse Radon transform gives the restored image. Simulation results are provided.
2017-02-23T13:41:15ZSayrol Clols, ElisaNikias, C LGasull Llampallas, AntoniThe authors propose the use of higher-order statistics (HOS) to study the problem of image restoration. They consider images degraded by linear or zero phase blurring point spread functions (PSF) and additive Gaussian noise. The complexity associated with the combination of two-dimensional signal processing and higher-order statistics is reduced by means of the Radon transform. The projection at each angle is an one-dimensional signal that can be processed by any existing 1-D higher-order statistics-based method. They apply two methods that have proven to attain good one-dimensional signal reconstruction, especially in the presence of noise. After the ideal projections have been estimated, the inverse Radon transform gives the restored image. Simulation results are provided.Stereo Image Analysis using Connected Operators
http://hdl.handle.net/2117/101384
Stereo Image Analysis using Connected Operators
Oliveras Vergés, Albert; Salembier Clairon, Philippe Jean; Garrido Ostermann, Luis
Connected operators are increasingly used in image processing due to their properties of simplifying the image with various criteria, without loosing contour's information. These properties are related to the connected operator approach that either preserves or completely eliminates a determined connected component, according to an established criterion of analysis. In this paper we will define a new connected operator for stereo images. The goal is to simplify one of the images (left) in the sense that the operator will eliminate the image components that are not present at a determined location in the other image (right). This filter let us select in a stereo image, objects as a function of their distance from the observer (for instance used in auto guided vehicles).
2017-02-22T15:22:48ZOliveras Vergés, AlbertSalembier Clairon, Philippe JeanGarrido Ostermann, LuisConnected operators are increasingly used in image processing due to their properties of simplifying the image with various criteria, without loosing contour's information. These properties are related to the connected operator approach that either preserves or completely eliminates a determined connected component, according to an established criterion of analysis. In this paper we will define a new connected operator for stereo images. The goal is to simplify one of the images (left) in the sense that the operator will eliminate the image components that are not present at a determined location in the other image (right). This filter let us select in a stereo image, objects as a function of their distance from the observer (for instance used in auto guided vehicles).Analysis of stereo images using connected operators
http://hdl.handle.net/2117/101327
Analysis of stereo images using connected operators
Oliveras Vergés, Albert; Garrido Ostermann, Luis; Salembier Clairon, Philippe Jean
Connected operators are increasingly used in image processing due to their properties of simplifying the image with various criteria, without loosing contour's information. These properties are related to the connected operator approach that either preserves or completely eliminates a determined connected component, according to an established criterion of analysis. In this paper we will define a new connected operator for stereo images. The goal is to simplify one of the images (left) in the sense that the operator will eliminate the image components that are not present at a determined location in the other image (right). This filter let us select in a stereo image, objects as a function of their distance from the observer (for instance used in auto guided vehicles).
2017-02-21T14:52:27ZOliveras Vergés, AlbertGarrido Ostermann, LuisSalembier Clairon, Philippe JeanConnected operators are increasingly used in image processing due to their properties of simplifying the image with various criteria, without loosing contour's information. These properties are related to the connected operator approach that either preserves or completely eliminates a determined connected component, according to an established criterion of analysis. In this paper we will define a new connected operator for stereo images. The goal is to simplify one of the images (left) in the sense that the operator will eliminate the image components that are not present at a determined location in the other image (right). This filter let us select in a stereo image, objects as a function of their distance from the observer (for instance used in auto guided vehicles).Edge versus contrast estimation of morphological filters
http://hdl.handle.net/2117/101256
Edge versus contrast estimation of morphological filters
Salembier Clairon, Philippe Jean
The estimation properties of morphological filters are considered in terms of edge localization and grey level contrast preservation in two-dimensional spaces. It is shown that, at least in practice, a compromise between these two characteristics has to be made. In a first step, morphological filters are compared with linear and median filters. It is concluded that an efficient edge localization can be achieved with morphological filters with reconstruction but not with median filters or with morphological openings or closings with structuring elements. In the case of strong noise, the good edge preservation provided by the reconstruction filters is obtained at the expense of contrast estimation. In order to be able to tune this tradeoff, a new type of morphological filter, called filters with partial reconstruction is proposed and studied.
2017-02-20T17:13:03ZSalembier Clairon, Philippe JeanThe estimation properties of morphological filters are considered in terms of edge localization and grey level contrast preservation in two-dimensional spaces. It is shown that, at least in practice, a compromise between these two characteristics has to be made. In a first step, morphological filters are compared with linear and median filters. It is concluded that an efficient edge localization can be achieved with morphological filters with reconstruction but not with median filters or with morphological openings or closings with structuring elements. In the case of strong noise, the good edge preservation provided by the reconstruction filters is obtained at the expense of contrast estimation. In order to be able to tune this tradeoff, a new type of morphological filter, called filters with partial reconstruction is proposed and studied.Unsupervised segmentation controlled by morphological contrast ext
http://hdl.handle.net/2117/101255
Unsupervised segmentation controlled by morphological contrast ext
Marqués Acosta, Fernando; Gasull Llampallas, Antoni
A novel approach for unsupervised image segmentation is described. This approach makes use of a Gaussian pyramid as multiresolution decomposition to analyze images. Compound random fields are used to model images at each resolution. The hierarchical image model is formed by a Strauss process in the lower level and a set of white Gaussian random fields in the upper level. This basic image model is adapted to the data present at each resolution. Segmentations at coarse resolutions are used to guide segmentations at finest resolutions. Segmentation quality is controlled, at each level, by means of morphological tools. The control procedure is based on the residue between the original image and a morphological center transform. This procedure checks whether the current segmentation contains all the relevant regions in the scene. If not, the algorithm introduces seeds into the segmented image in order to detect the new regions.
2017-02-20T17:07:12ZMarqués Acosta, FernandoGasull Llampallas, AntoniA novel approach for unsupervised image segmentation is described. This approach makes use of a Gaussian pyramid as multiresolution decomposition to analyze images. Compound random fields are used to model images at each resolution. The hierarchical image model is formed by a Strauss process in the lower level and a set of white Gaussian random fields in the upper level. This basic image model is adapted to the data present at each resolution. Segmentations at coarse resolutions are used to guide segmentations at finest resolutions. Segmentation quality is controlled, at each level, by means of morphological tools. The control procedure is based on the residue between the original image and a morphological center transform. This procedure checks whether the current segmentation contains all the relevant regions in the scene. If not, the algorithm introduces seeds into the segmented image in order to detect the new regions.3D point cloud video segmentation based on interaction analysis
http://hdl.handle.net/2117/101137
3D point cloud video segmentation based on interaction analysis
Lin, Xiao; Casas Pla, Josep Ramon; Pardàs Feliu, Montse
Given the widespread availability of point cloud data from consumer depth sensors, 3D segmentation becomes a promising building block for high level applications such as scene understanding and interaction analysis. It benefits from the richer information contained in actual world 3D data compared to apparent (projected) data in 2D images. This also implies that the classical color segmentation challenges have recently shifted to RGBD data, whereas new emerging challenges are added as 3D information from depth measurements is usually noisy, sparse and unorganized. We present a novel segmentation approach for 3D point cloud video based on low level features and oriented to the analysis of object interactions. A hierarchical representation of the input point cloud is proposed to efficiently segment 3D data at the finer level, and to temporally establish the correspondence between segments, while dynamically managing the object split and merge at the coarser level. Experiments illustrate promising results and its potential application in object interaction analysis.
2017-02-16T11:45:35ZLin, XiaoCasas Pla, Josep RamonPardàs Feliu, MontseGiven the widespread availability of point cloud data from consumer depth sensors, 3D segmentation becomes a promising building block for high level applications such as scene understanding and interaction analysis. It benefits from the richer information contained in actual world 3D data compared to apparent (projected) data in 2D images. This also implies that the classical color segmentation challenges have recently shifted to RGBD data, whereas new emerging challenges are added as 3D information from depth measurements is usually noisy, sparse and unorganized. We present a novel segmentation approach for 3D point cloud video based on low level features and oriented to the analysis of object interactions. A hierarchical representation of the input point cloud is proposed to efficiently segment 3D data at the finer level, and to temporally establish the correspondence between segments, while dynamically managing the object split and merge at the coarser level. Experiments illustrate promising results and its potential application in object interaction analysis.3D point cloud video segmentation oriented to the analysis of interactions
http://hdl.handle.net/2117/101101
3D point cloud video segmentation oriented to the analysis of interactions
Lin, Xiao; Casas Pla, Josep Ramon; Pardàs Feliu, Montse
Given the widespread availability of point cloud data from consumer depth sensors, 3D point cloud segmentation becomes a promising building block for high level applications such as scene understanding and interaction analysis. It benefits from the richer information contained in real world 3D data compared to 2D images. This also implies that the classical color segmentation challenges have shifted to RGBD data, and new challenges have also emerged as the depth information is usually noisy, sparse and unorganized. Meanwhile, the lack of 3D point cloud ground truth labeling also limits the development and comparison among methods in 3D point cloud segmentation. In this paper, we present two contributions: a novel graph based point cloud segmentation method for RGBD stream data with interacting objects and a new ground truth labeling for a previously published data set. This data set focuses on interaction (merge and split between ’object’ point clouds), which differentiates itself from the few existing labeled RGBD data sets which are more oriented to Simultaneous Localization And Mapping (SLAM) tasks. The proposed point cloud segmentation method is evaluated with the 3D point cloud ground truth labeling. Experiments show the promising result of our approach.
2017-02-15T14:58:25ZLin, XiaoCasas Pla, Josep RamonPardàs Feliu, MontseGiven the widespread availability of point cloud data from consumer depth sensors, 3D point cloud segmentation becomes a promising building block for high level applications such as scene understanding and interaction analysis. It benefits from the richer information contained in real world 3D data compared to 2D images. This also implies that the classical color segmentation challenges have shifted to RGBD data, and new challenges have also emerged as the depth information is usually noisy, sparse and unorganized. Meanwhile, the lack of 3D point cloud ground truth labeling also limits the development and comparison among methods in 3D point cloud segmentation. In this paper, we present two contributions: a novel graph based point cloud segmentation method for RGBD stream data with interacting objects and a new ground truth labeling for a previously published data set. This data set focuses on interaction (merge and split between ’object’ point clouds), which differentiates itself from the few existing labeled RGBD data sets which are more oriented to Simultaneous Localization And Mapping (SLAM) tasks. The proposed point cloud segmentation method is evaluated with the 3D point cloud ground truth labeling. Experiments show the promising result of our approach.Morphological detection based on size and contrast criteria Morphological detection based on size and contrast criteria application to cells detection
http://hdl.handle.net/2117/100694
Morphological detection based on size and contrast criteria Morphological detection based on size and contrast criteria application to cells detection
Salembier Clairon, Philippe Jean; Gasull Llampallas, Antoni; Marqués Acosta, Fernando; Sayrol Clols, Elisa
This paper deals with a detection algorithm relying on size and contrast criteria. It is suitable for a large range of applications where a priori information about the size and the contrast of the objects to detect is available. The detection is performed in three separate steps: the first one is a preprocessing which removes unuseful information with a size criterion. The second one performs a feature extraction based on contrast. Finally, the last step is the decision itself. All these steps make use of morphological transformations because of their ability to deal with the criteria of interest and of their low computational cost. As an example, this algorithm is applied to the automatic detection of spermatozoa.
2017-02-08T14:45:00ZSalembier Clairon, Philippe JeanGasull Llampallas, AntoniMarqués Acosta, FernandoSayrol Clols, ElisaThis paper deals with a detection algorithm relying on size and contrast criteria. It is suitable for a large range of applications where a priori information about the size and the contrast of the objects to detect is available. The detection is performed in three separate steps: the first one is a preprocessing which removes unuseful information with a size criterion. The second one performs a feature extraction based on contrast. Finally, the last step is the decision itself. All these steps make use of morphological transformations because of their ability to deal with the criteria of interest and of their low computational cost. As an example, this algorithm is applied to the automatic detection of spermatozoa.