Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310124
Stefan Schrunner, Olivia Bluder, Anja Zernig, Andre Kästner, Roman Kern
In semiconductor industry it is of paramount importance to check whether a manufactured device fulfills all quality specifications and is therefore suitable for being sold to the customer. The occurrence of specific spatial patterns within the so-called wafer test data, i.e. analog electric measurements, might point out on production issues. However, the shape of these critical patterns is unknown. In this paper different kinds of process patterns are extracted from wafer test data by an image processing approach using Markov Random Field models for image restoration. The goal is to develop an automated procedure to identify visible patterns in wafer test data to improve pattern matching. This step is a necessary precondition for a subsequent root-cause analysis of these patterns. The developed pattern extraction algorithm yields a more accurate discrimination between distinct patterns, resulting in an improved pattern comparison than in the original dataset. In a next step pattern classification will be applied to improve the production process control.
{"title":"Markov random fields for pattern extraction in analog wafer test data","authors":"Stefan Schrunner, Olivia Bluder, Anja Zernig, Andre Kästner, Roman Kern","doi":"10.1109/IPTA.2017.8310124","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310124","url":null,"abstract":"In semiconductor industry it is of paramount importance to check whether a manufactured device fulfills all quality specifications and is therefore suitable for being sold to the customer. The occurrence of specific spatial patterns within the so-called wafer test data, i.e. analog electric measurements, might point out on production issues. However, the shape of these critical patterns is unknown. In this paper different kinds of process patterns are extracted from wafer test data by an image processing approach using Markov Random Field models for image restoration. The goal is to develop an automated procedure to identify visible patterns in wafer test data to improve pattern matching. This step is a necessary precondition for a subsequent root-cause analysis of these patterns. The developed pattern extraction algorithm yields a more accurate discrimination between distinct patterns, resulting in an improved pattern comparison than in the original dataset. In a next step pattern classification will be applied to improve the production process control.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126341745","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310079
Justin Strait, S. Kurtek
An important problem in statistical shape analysis is the matching of geometric features across shapes, known as registration. In short, given two objects, one wants to know the correspondence of points on one shape to points on another. Such a matching problem, with various levels of complexity, is present regardless of the shape's mathematical representation. A recent framework for shape analysis of n-dimensional curves combines an infinite-dimensional functional curve representation with landmark information encoding important curve features. In this setting, shape matching is performed by minimizing an objective function with constraints, which respect landmark correspondences. Currently, the minimizer in this approach is found using piecewise dynamic programming; this does not respect the smoothness requirement of the matching function. Thus, the solution is not really a member of the group of registration functions. In this work, we present a landmark-constrained gradient descent algorithm, which searches for a smooth matching function and respects landmark locations. We compare the proposed method to the previously used approach using examples from the MPEG-7 dataset.
{"title":"A novel algorithm for optimal matching of elastic shapes with landmark constraints","authors":"Justin Strait, S. Kurtek","doi":"10.1109/IPTA.2017.8310079","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310079","url":null,"abstract":"An important problem in statistical shape analysis is the matching of geometric features across shapes, known as registration. In short, given two objects, one wants to know the correspondence of points on one shape to points on another. Such a matching problem, with various levels of complexity, is present regardless of the shape's mathematical representation. A recent framework for shape analysis of n-dimensional curves combines an infinite-dimensional functional curve representation with landmark information encoding important curve features. In this setting, shape matching is performed by minimizing an objective function with constraints, which respect landmark correspondences. Currently, the minimizer in this approach is found using piecewise dynamic programming; this does not respect the smoothness requirement of the matching function. Thus, the solution is not really a member of the group of registration functions. In this work, we present a landmark-constrained gradient descent algorithm, which searches for a smooth matching function and respects landmark locations. We compare the proposed method to the previously used approach using examples from the MPEG-7 dataset.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130262856","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310136
Y. Guerbai, Y. Chibani, Bilal Hadjadji
Handwriting gender recognition becomes considerable matter for the document analysis community, due to its effective use in practical applications. This paper addresses the problem of classifying handwriting data with respect to gender. From the state of the art, only a few studies have been carried out in this field. Thus, we propose a new framework for classifying the gender from the handwriting document using the curvelet transform and a classification method based on One-Class Support Vector Machine (OC-SVM). In order to improve the robustness of the proposed system, multiple OC-SVM classifiers are combined according to the type of distance used into the kernel. Experimental results conducted on IAM datasets show the effective use of the OC-SVM for handwriting gender recognition comparatively to the state of the art.
{"title":"Handwriting gender recognition system based on the one-class support vector machines","authors":"Y. Guerbai, Y. Chibani, Bilal Hadjadji","doi":"10.1109/IPTA.2017.8310136","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310136","url":null,"abstract":"Handwriting gender recognition becomes considerable matter for the document analysis community, due to its effective use in practical applications. This paper addresses the problem of classifying handwriting data with respect to gender. From the state of the art, only a few studies have been carried out in this field. Thus, we propose a new framework for classifying the gender from the handwriting document using the curvelet transform and a classification method based on One-Class Support Vector Machine (OC-SVM). In order to improve the robustness of the proposed system, multiple OC-SVM classifiers are combined according to the type of distance used into the kernel. Experimental results conducted on IAM datasets show the effective use of the OC-SVM for handwriting gender recognition comparatively to the state of the art.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122899791","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310134
Jorge Calvo-Zaragoza, Gabriel Vigliensoni, Ichiro Fujinaga
Content within musical documents not only contains music symbol but also include different elements such as staff lines, text, or frontispieces. Before attempting to automatically recognize components in these layers, it is necessary to perform an analysis of the musical documents in order to detect and classify each of these constituent parts. The obstacle for this analysis is the high heterogeneity amongst music collections, especially with ancient documents, which makes it difficult to devise methods that can be generalizable to a broader range of sources. In this paper we propose a data-driven document analysis framework based on machine learning that focuses on classifying regions of interest at pixel level. For that, we make use of Convolutional Neural Networks trained to infer the category of each pixel. The main advantage of this approach is that it can be applied regardless of the type of document provided, as long as training data is available. Since this work represents first efforts in that direction, our experimentation focuses on reporting a baseline classification using our framework. The experiments show promising performance, achieving an accuracy around 90% in two corpora of old music documents.
{"title":"Pixelwise classification for music document analysis","authors":"Jorge Calvo-Zaragoza, Gabriel Vigliensoni, Ichiro Fujinaga","doi":"10.1109/IPTA.2017.8310134","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310134","url":null,"abstract":"Content within musical documents not only contains music symbol but also include different elements such as staff lines, text, or frontispieces. Before attempting to automatically recognize components in these layers, it is necessary to perform an analysis of the musical documents in order to detect and classify each of these constituent parts. The obstacle for this analysis is the high heterogeneity amongst music collections, especially with ancient documents, which makes it difficult to devise methods that can be generalizable to a broader range of sources. In this paper we propose a data-driven document analysis framework based on machine learning that focuses on classifying regions of interest at pixel level. For that, we make use of Convolutional Neural Networks trained to infer the category of each pixel. The main advantage of this approach is that it can be applied regardless of the type of document provided, as long as training data is available. Since this work represents first efforts in that direction, our experimentation focuses on reporting a baseline classification using our framework. The experiments show promising performance, achieving an accuracy around 90% in two corpora of old music documents.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128589680","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310153
Z. Hamici
In this paper a novel signal processing algorithm for secure image communication is proposed. It is a genetic algorithm that combines both gene fusion and Horizontal Gene Transfer (HGT) inspired from the spread of antibiotic resistance in bacteria. The symmetric encryption key playing the role of a pathogen is generated by controlled multi-layer random sequences. The principal key is then absorbed by the organism genome represented by the image. The process of encryption starts by a full uptake of the key-agent or pathogen which has the size of the whole image; the image and the principal key are merged together in a gene fusion process. A second phase of encryption is an obfuscation process produced by an HGT where the genes are pixels and the chromosomes are the rows and columns of the image. The obfuscation process as a second layer of encryption is realized by two sub-keys from which the principal key matrix is constructed. The whole process is repeated recursively L rounds. A Salt extracted from the image hash-value is used against chosen-plaintext cryptanalysis, therefore, even a modification of one pixel will generate different encryption keys adding a stealthy-key feature to the cipher. The key generation process with the two layers of encryption of the genetic algorithm are fully described. Results of the signal processing algorithm based on gene fusion and HGT show a great promise in genetic inspired data security.
{"title":"Image cryptography based on the imitation of gene fusion and horizontal gene transfer","authors":"Z. Hamici","doi":"10.1109/IPTA.2017.8310153","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310153","url":null,"abstract":"In this paper a novel signal processing algorithm for secure image communication is proposed. It is a genetic algorithm that combines both gene fusion and Horizontal Gene Transfer (HGT) inspired from the spread of antibiotic resistance in bacteria. The symmetric encryption key playing the role of a pathogen is generated by controlled multi-layer random sequences. The principal key is then absorbed by the organism genome represented by the image. The process of encryption starts by a full uptake of the key-agent or pathogen which has the size of the whole image; the image and the principal key are merged together in a gene fusion process. A second phase of encryption is an obfuscation process produced by an HGT where the genes are pixels and the chromosomes are the rows and columns of the image. The obfuscation process as a second layer of encryption is realized by two sub-keys from which the principal key matrix is constructed. The whole process is repeated recursively L rounds. A Salt extracted from the image hash-value is used against chosen-plaintext cryptanalysis, therefore, even a modification of one pixel will generate different encryption keys adding a stealthy-key feature to the cipher. The key generation process with the two layers of encryption of the genetic algorithm are fully described. Results of the signal processing algorithm based on gene fusion and HGT show a great promise in genetic inspired data security.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125950426","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310127
Wenqiang Zhan, Changshi Xiao, Haiwen Yuan, Y. Wen
In the maritime environment, real-time and accurate water line detection can help unmanned surface vehicles (USVs) for autonomous navigation. Special features such as cloud clutter, water reflection, water surface glint and background texture in optical images make it very difficult for USV to detect the water line accurately. To address this problem, a novel water line detection approach is proposed for a practical USV system in the inland water. In this paper, land-water and sky-water line are taken into consideration as part of the water line. To minimize the influence of background texture, the local variation method is introduced to smooth the image and remain the background structure, such as the water line. Regarding the linear characteristic, these water line points are constrained near a line by the RANSAC algorithm. The experimental results are given in the end and show that the proposed approach is effective and can be applied to different environments.
{"title":"Effective Waterline detection for unmanned surface vehicles in inland water","authors":"Wenqiang Zhan, Changshi Xiao, Haiwen Yuan, Y. Wen","doi":"10.1109/IPTA.2017.8310127","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310127","url":null,"abstract":"In the maritime environment, real-time and accurate water line detection can help unmanned surface vehicles (USVs) for autonomous navigation. Special features such as cloud clutter, water reflection, water surface glint and background texture in optical images make it very difficult for USV to detect the water line accurately. To address this problem, a novel water line detection approach is proposed for a practical USV system in the inland water. In this paper, land-water and sky-water line are taken into consideration as part of the water line. To minimize the influence of background texture, the local variation method is introduced to smooth the image and remain the background structure, such as the water line. Regarding the linear characteristic, these water line points are constrained near a line by the RANSAC algorithm. The experimental results are given in the end and show that the proposed approach is effective and can be applied to different environments.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125699107","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310087
Bo Li, W. Bouachir, Rafik Gouiaa, R. Noumeir
Inmates in solitary confinement may attempt to harm themselves in many ways, resulting in trivial to mortal injuries. In this context, suicide by hanging is one of the major causes of death among the incarcerated. The Rapid detection of suicide can reduce the mortality rate. Recently, several technologies have been developed to detect suicide by hanging attempts, but most of them use bulky devices, or they are greatly depending on human attention. In this paper, we propose a computer vision based system to automatically detect suicide by hanging attempts. Our method is based on modeling suicidal actions using pose and motion features, by exploiting the body joints' positions. The proposed video surveillance system analyses depth images provided by an RGB-D camera to detect the event of interest in real-time, regardeless of illumination conditions. The experimental results obtained on a realistic dataset demonstrated the high precision of our system in detecting suicide by hanging.
{"title":"Real-time recognition of suicidal behavior using an RGB-D camera","authors":"Bo Li, W. Bouachir, Rafik Gouiaa, R. Noumeir","doi":"10.1109/IPTA.2017.8310087","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310087","url":null,"abstract":"Inmates in solitary confinement may attempt to harm themselves in many ways, resulting in trivial to mortal injuries. In this context, suicide by hanging is one of the major causes of death among the incarcerated. The Rapid detection of suicide can reduce the mortality rate. Recently, several technologies have been developed to detect suicide by hanging attempts, but most of them use bulky devices, or they are greatly depending on human attention. In this paper, we propose a computer vision based system to automatically detect suicide by hanging attempts. Our method is based on modeling suicidal actions using pose and motion features, by exploiting the body joints' positions. The proposed video surveillance system analyses depth images provided by an RGB-D camera to detect the event of interest in real-time, regardeless of illumination conditions. The experimental results obtained on a realistic dataset demonstrated the high precision of our system in detecting suicide by hanging.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128347923","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310139
Adam Novozámský, M. Šorel
The integrity of visual data is important for the credibility of news media and especially when used as an evidence in court or during criminal investigation. The common way to manipulate image content is copying an object and pasting in another location of the same image. In this paper, we describe a new idea for the detection of this type of forgery in JPEG images, where the compression significantly degrades detection by popular algorithms. We derive a JPEG-based constraint that any pair of patches must satisfy to be considered a valid candidate for tampered area. We propose also efficient algorithm to verify the constraint that can be integrated into most existing methods. Experiments show significant improvement of detection, especially for difficult cases, such as small objects, objects covered by textureless areas and repeated patterns.
{"title":"JPEG compression model in copy-move forgery detection","authors":"Adam Novozámský, M. Šorel","doi":"10.1109/IPTA.2017.8310139","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310139","url":null,"abstract":"The integrity of visual data is important for the credibility of news media and especially when used as an evidence in court or during criminal investigation. The common way to manipulate image content is copying an object and pasting in another location of the same image. In this paper, we describe a new idea for the detection of this type of forgery in JPEG images, where the compression significantly degrades detection by popular algorithms. We derive a JPEG-based constraint that any pair of patches must satisfy to be considered a valid candidate for tampered area. We propose also efficient algorithm to verify the constraint that can be integrated into most existing methods. Experiments show significant improvement of detection, especially for difficult cases, such as small objects, objects covered by textureless areas and repeated patterns.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131330525","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310082
A. Rodríguez-Sánchez, D. Chea, G. Azzopardi, Sebastian Stabinger
The image of an object changes dramatically depending on the lightning conditions surrounding that object. Shadows, reflections and highlights can make the object very difficult to be recognized for an automatic system. Additionally, images used in medical applications, such as endoscopic images and videos contain a large amount of such reflective components. This can pose an extra difficulty for experts to analyze such type of videos and images. It can then be useful to detect — and possibly correct — the locations where those highlights happen. In this work we designed a Convolutional Neural Network for that task. We trained such a network using a dataset that contains groundtruth highlights showing that those reflective elements can be learnt and thus located and extracted. We then used that trained network to localize and correct the highlights in endoscopic images from the El Salvador Atlas Gastrointestinal videos obtaining promising results.
{"title":"A deep learning approach for detecting and correcting highlights in endoscopic images","authors":"A. Rodríguez-Sánchez, D. Chea, G. Azzopardi, Sebastian Stabinger","doi":"10.1109/IPTA.2017.8310082","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310082","url":null,"abstract":"The image of an object changes dramatically depending on the lightning conditions surrounding that object. Shadows, reflections and highlights can make the object very difficult to be recognized for an automatic system. Additionally, images used in medical applications, such as endoscopic images and videos contain a large amount of such reflective components. This can pose an extra difficulty for experts to analyze such type of videos and images. It can then be useful to detect — and possibly correct — the locations where those highlights happen. In this work we designed a Convolutional Neural Network for that task. We trained such a network using a dataset that contains groundtruth highlights showing that those reflective elements can be learnt and thus located and extracted. We then used that trained network to localize and correct the highlights in endoscopic images from the El Salvador Atlas Gastrointestinal videos obtaining promising results.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131380444","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}
Pub Date : 2017-11-01DOI: 10.1109/IPTA.2017.8310155
H. Rehouma, R. Noumeir, P. Jouvet, W. Bouachir, S. Essouri
This paper presents a novel computer vision method to measure the breathing pattern in intensive care environment. The proposed system uses depth information captured by two RGB-D cameras in order to reconstruct a 3D surface of a patient's torso with a high spatial coverage. The optimal positioning for the sensors is a key step to perform an accurate 3D reconstruction without interfering with patient care. In this context, our hardware setup meets the clinical requirements while allowing accurate estimation of respiratory parameters including respiratory rate, tidal volume and inspiratory time. Our system provides the motion information not only for the top of the torso surface but also for its both lateral sides. Our method was tested in an environment designed for critically ill children, where it was compared to the gold standard method currently used in intensive care units. The performed experiments yielded high accuracy and showed significant agreement with gold standard method.
{"title":"A computer vision method for respiratory monitoring in intensive care environment using RGB-D cameras","authors":"H. Rehouma, R. Noumeir, P. Jouvet, W. Bouachir, S. Essouri","doi":"10.1109/IPTA.2017.8310155","DOIUrl":"https://doi.org/10.1109/IPTA.2017.8310155","url":null,"abstract":"This paper presents a novel computer vision method to measure the breathing pattern in intensive care environment. The proposed system uses depth information captured by two RGB-D cameras in order to reconstruct a 3D surface of a patient's torso with a high spatial coverage. The optimal positioning for the sensors is a key step to perform an accurate 3D reconstruction without interfering with patient care. In this context, our hardware setup meets the clinical requirements while allowing accurate estimation of respiratory parameters including respiratory rate, tidal volume and inspiratory time. Our system provides the motion information not only for the top of the torso surface but also for its both lateral sides. Our method was tested in an environment designed for critically ill children, where it was compared to the gold standard method currently used in intensive care units. The performed experiments yielded high accuracy and showed significant agreement with gold standard method.","PeriodicalId":316356,"journal":{"name":"2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116530990","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：