Pub Date : 1900-01-01DOI: 10.24132/csrn.2019.2902.2.10
Matthias Trapp, S. Pasewaldt, J. Döllner
This paper presents a GPU-based approach to color quantization by mapping of arbitrary color palettes to input images using Look-Up Tables (LUTs). For it, different types of LUTs, their GPU-based generation, representation, and respective mapping implementations are described and their run-time performance is evaluated and compared.
{"title":"Techniques for GPU-based Color Palette Mapping","authors":"Matthias Trapp, S. Pasewaldt, J. Döllner","doi":"10.24132/csrn.2019.2902.2.10","DOIUrl":"https://doi.org/10.24132/csrn.2019.2902.2.10","url":null,"abstract":"This paper presents a GPU-based approach to color quantization by mapping of arbitrary color palettes to input images using Look-Up Tables (LUTs). For it, different types of LUTs, their GPU-based generation, representation, and respective mapping implementations are described and their run-time performance is evaluated and compared.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"14 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114006516","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 : 1900-01-01DOI: 10.24132/csrn.2019.2901.1.9
Thomas Schult, Uwe Klose, Till-Karsten Hauser, H. Ehricke
3D visualization of volumetric line integral convolution (LIC) datasets has been a field of constant research. So far, most approaches have focused on finding suitable transfer functions and defining appropriate clipping strategies in order to solve the problem of occlusion. In medicine, extensions of the LIC algorithm to diffusion weighted magnetic resonance imaging (dwMRI) have been proposed, allowing highly resolved LIC volumes to be generated. These are used for brain white matter visualization by LIC slice images, depicting fiber structures with good contrast. However, 3D visualization of fiber pathways by volume rendering faces the problem of occlusion of anatomic regions of interest by the dense brain white matter pattern. In this paper, we introduce an anatomy focused LIC algorithm, which allows specific fiber architectures to be visualized by volume rendering. It uses an anatomical atlas, matched to the dwMRI dataset, during the generation of the LIC noise input pattern. Thus,anatomic fiber structures of interest are emphasized, while surrounding fiber tissue is thinned out and its opacity is modulated. Additionally, we present an adaptation of the orientation-dependent transparency rendering algorithm, which recently has been proposed for fiber streamline visualization, to LIC data. The novel methods are evaluated by application to dwMRI datasets from glioma patients, visualizing fiber structures of interest in the vicinity of the lesion.
{"title":"Anatomy-Focused Volume Line Integral Convolution for Brain White Matter Visualization","authors":"Thomas Schult, Uwe Klose, Till-Karsten Hauser, H. Ehricke","doi":"10.24132/csrn.2019.2901.1.9","DOIUrl":"https://doi.org/10.24132/csrn.2019.2901.1.9","url":null,"abstract":"3D visualization of volumetric line integral convolution (LIC) datasets has been a field of constant research. So far, most approaches have focused on finding suitable transfer functions and defining appropriate clipping strategies in order to solve the problem of occlusion. In medicine, extensions of the LIC algorithm to diffusion weighted magnetic resonance imaging (dwMRI) have been proposed, allowing highly resolved LIC volumes to be generated. These are used for brain white matter visualization by LIC slice images, depicting fiber structures with good contrast. However, 3D visualization of fiber pathways by volume rendering faces the problem of occlusion of anatomic regions of interest by the dense brain white matter pattern. In this paper, we introduce an anatomy focused LIC algorithm, which allows specific fiber architectures to be visualized by volume rendering. It uses an anatomical atlas, matched to the dwMRI dataset, during the generation of the LIC noise input pattern. Thus,anatomic fiber structures of interest are emphasized, while surrounding fiber tissue is thinned out and its opacity is modulated. Additionally, we present an adaptation of the orientation-dependent transparency rendering algorithm, which recently has been proposed for fiber streamline visualization, to LIC data. The novel methods are evaluated by application to dwMRI datasets from glioma patients, visualizing fiber structures of interest in the vicinity of the lesion.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128467976","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 : 1900-01-01DOI: 10.24132/csrn.2019.2901.1.12
Kenichi Sugihara
3D building models with roofs are important in several fields, such as urban planning and BIM (Building Information Model). However, enormous time and labor are required to create these 3D models. In order to automate laborious steps, a GIS and CG integrated system is proposed for the automatic generation of 3D building models, based on building polygons (building footprints) on digital maps. The generation is implemented through straight skeleton computation, in which three events (‘Edge’ and ‘Split’, ‘Vertex’ events) were proposed. In the computation process, usually three edges propagate into a node. Often it causes an acute angle shape that is not appropriate for roof boards. To avoid the inappropriate shape, in this paper, methodologies are proposed for adding ‘Line segment’ events besides the conventional events, and monotone polygon nodes sorting.
{"title":"Straight Skeleton Computation Optimized for Roof Model Generation","authors":"Kenichi Sugihara","doi":"10.24132/csrn.2019.2901.1.12","DOIUrl":"https://doi.org/10.24132/csrn.2019.2901.1.12","url":null,"abstract":"3D building models with roofs are important in several fields, such as urban planning and BIM (Building Information Model). However, enormous time and labor are required to create these 3D models. In order to automate laborious steps, a GIS and CG integrated system is proposed for the automatic generation of 3D building models, based on building polygons (building footprints) on digital maps. The generation is implemented through straight skeleton computation, in which three events (‘Edge’ and ‘Split’, ‘Vertex’ events) were proposed. In the computation process, usually three edges propagate into a node. Often it causes an acute angle shape that is not appropriate for roof boards. To avoid the inappropriate shape, in this paper, methodologies are proposed for adding ‘Line segment’ events besides the conventional events, and monotone polygon nodes sorting.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"174 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125798165","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 : 1900-01-01DOI: 10.24132/csrn.2019.2901.1.10
A. Shcherbakov, V. Frolov
{"title":"Dynamic Radiosity","authors":"A. Shcherbakov, V. Frolov","doi":"10.24132/csrn.2019.2901.1.10","DOIUrl":"https://doi.org/10.24132/csrn.2019.2901.1.10","url":null,"abstract":"","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"192 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133563981","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 : 1900-01-01DOI: 10.24132/csrn.2019.2901.1.4
Alexander Leipnitz, T. Strutz, O. Jokisch
Convolutional neural networks are applied successfully for image classification and object detection. Recently, they have been adopted to semantic segmentation tasks and several new network architectures have been proposed. With respect to automotive applications, the Cityscapes dataset is often used as a benchmark. It is one of the biggest datasets in this field and consists of a training, a validation, and a test set. While training and validation allow the optimisation of these nets, the test dataset can be used to evaluate their performance. Our investigations have shown that while these networks perform well for images of the Cityscapes dataset, their segmentation quality significantly drops when applied to new data. It seems that they have limited generalisation abilities. In order to find out whether the image content itself or other image properties cause this effect, we have carried out systematic investigations with modified Cityscapes data. We have found that camera-dependent image properties like brightness, contrast, or saturation can significantly influence the segmentation quality. This papers presents the results of these tests including eight state-of-the-art CNNs. It can be concluded that the out-of-the-boxusage of CNNs in real-world environments is not recommended.
{"title":"Performance Assessment of Convolutional Neural Networks for Semantic Image Segmentation","authors":"Alexander Leipnitz, T. Strutz, O. Jokisch","doi":"10.24132/csrn.2019.2901.1.4","DOIUrl":"https://doi.org/10.24132/csrn.2019.2901.1.4","url":null,"abstract":"Convolutional neural networks are applied successfully for image classification and object detection. Recently, they have been adopted to semantic segmentation tasks and several new network architectures have been proposed. With respect to automotive applications, the Cityscapes dataset is often used as a benchmark. It is one of the biggest datasets in this field and consists of a training, a validation, and a test set. While training and validation allow the optimisation of these nets, the test dataset can be used to evaluate their performance.\u0000Our investigations have shown that while these networks perform well for images of the Cityscapes dataset, their segmentation quality significantly drops when applied to new data. It seems that they have limited generalisation abilities. In order to find out whether the image content itself or other image properties cause this effect, we have carried out systematic investigations with modified Cityscapes data. We have found that camera-dependent image properties like brightness, contrast, or saturation can significantly influence the segmentation quality. This papers presents the results of these tests including eight state-of-the-art CNNs. It can be concluded that the out-of-the-boxusage of CNNs in real-world environments is not recommended.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"171 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115131736","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 : 1900-01-01DOI: 10.24132/csrn.2020.3001.3
J. Valero
The process of filtering digital images represented by complex Cartesian allows to use the available onedimensional (1D) elements (interpixel); however, having those additional 1D elements increases both the volume of data and the time for processing them. The time reduction strategy based on a parallel computing scheme on the number of available central processing units (CPUs) does not consider additional computing resources such as those offered by general purpose graphics processing units (GPUs) of NVIDIA. Parallel computing possibilities provided by the NVIDIA GPUs were explored and, based on them, a computational scheme for the digital image Cartesian complexes filtering task was proposed using the application program interface Open Computing Language (OpenCL) provided for NVIDIA corporation GPUs. The results assessment was established by comparing the response times of the proposed solution compared to those obtained using only CPU resources. The obtained implementation is an alternative to parallelization of the filtering task, which provides response times up to 14 times faster than those obtained with the implementation that uses only the CPU resource. The NVIDIA multicore GPU significantly improves the parallelism, which can be used in conjunction with the available multicore CPU computing capacity, balancing the workload between these two computing powers using both simultaneously.
{"title":"Inter-Pixel Filtrering of Digital Images with CUDA from NVIDIA","authors":"J. Valero","doi":"10.24132/csrn.2020.3001.3","DOIUrl":"https://doi.org/10.24132/csrn.2020.3001.3","url":null,"abstract":"The process of filtering digital images represented by complex Cartesian allows to use the available onedimensional \u0000(1D) elements (interpixel); however, having those additional 1D elements increases both the volume of data and the time for processing them. The time reduction strategy based on a parallel computing scheme on the number of available central processing units (CPUs) does not consider additional computing resources such as \u0000those offered by general purpose graphics processing units (GPUs) of NVIDIA. Parallel computing possibilities provided by the NVIDIA GPUs were explored and, based on them, a computational scheme for the digital image Cartesian complexes filtering task was proposed using the application program interface Open Computing Language \u0000(OpenCL) provided for NVIDIA corporation GPUs. The results assessment was established by comparing the response times of the proposed solution compared to those obtained using only CPU resources. The obtained implementation is an alternative to parallelization of the filtering task, which provides response times up to 14 times faster than those obtained with the implementation that uses only the CPU resource. The NVIDIA multicore GPU significantly improves the parallelism, which can be used in conjunction with the available multicore CPU computing capacity, balancing the workload between these two computing powers using both simultaneously.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123299758","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 : 1900-01-01DOI: 10.24132/csrn.2019.2902.2.11
A. Basov, V. Budak, A. Grimailo
This article shows a new mathematical model for calculation of multiple reflections based on the Stokes vector and Mueller matrices. The global illumination equation and local estimations method were generalized on the polarization case. The results of the calculation of multiple reflections using the local estimations method show a difference of more than 30% between the standard calculation and the polarization-accounting one. A way to describe the surface reflection with polarization account is proposed.
{"title":"The Role of Polarization in The Multiple Reflections Modeling","authors":"A. Basov, V. Budak, A. Grimailo","doi":"10.24132/csrn.2019.2902.2.11","DOIUrl":"https://doi.org/10.24132/csrn.2019.2902.2.11","url":null,"abstract":"This article shows a new mathematical model for calculation of multiple reflections based on the Stokes vector and Mueller matrices. The global illumination equation and local estimations method were generalized on the polarization case. The results of the calculation of multiple reflections using the local estimations method show a difference of more than 30% between the standard calculation and the polarization-accounting one. A way to describe the surface reflection with polarization account is proposed.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"160 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120863828","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 : 1900-01-01DOI: 10.24132/csrn.2020.3001.13
M. Mancosu, S. Czanner, M. Punter
In recent years, the use of Virtual Reality (VR) has skyrocketed in many fields. From videogames to education, the popularity of this technology is well known also thanks to its affordability and the growing number of available content. An immersive approach and the correct use of gamification have been proven to be valuable tools for learning in different fields. There are many examples of the adoption of VR application also in the healthcare setting not only for learning purposes but also for patients’ rehabilitation. This paper is intended to list some existing works and then present our project for an immersive serious game with the aim of enhancing the training of healthcare workers for the assessment of stroke patients.
{"title":"Learning Capacity in Simulated Virtual Neurological Procedures","authors":"M. Mancosu, S. Czanner, M. Punter","doi":"10.24132/csrn.2020.3001.13","DOIUrl":"https://doi.org/10.24132/csrn.2020.3001.13","url":null,"abstract":"In recent years, the use of Virtual Reality (VR) has skyrocketed in many fields. From videogames to education, the popularity of this technology is well known also thanks to its affordability and the growing number of available content. An immersive approach and the correct use of gamification have been proven to be valuable tools for learning in different fields. There are many examples of the adoption of VR application also in the healthcare setting not only for learning purposes but also for patients’ rehabilitation. This paper is intended to list some existing works and then present our project for an immersive serious game with the aim of enhancing the training of healthcare workers for the assessment of stroke patients.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126626261","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 : 1900-01-01DOI: 10.24132/csrn.2020.3001.15
Yuchong Zhang, Y. Ma, Adel Omrani Hamzekalaei, Rahul Yadav, M. Fjeld, M. Fratarcangeli
Inspired by the high performance in image-based medical analysis, this paper explores the use of advanced segmentation techniques for industrial Microwave Tomography (MWT). Our context is the visual analysis of moisture levels in porous foams undergoing microwave drying. We propose an automatic segmentation technique—MWT Segmentation based on K -means (MWTS-KM) and demonstrate its efficiency and accuracy for industrial use. MWTS-KM consists of three stages: image augmentation, grayscale conversion, and K -means implementation. To estimate the performance of this technique, we empirically benchmark its efficiency and accuracy against two well-established alternatives: Otsu and K -means. To elicit performance data, three metrics (Jaccard index, Dice coefficient and false positive) are used. Our results indicate that MWTS-KM outperforms the well-established Otsu and K -means, both in visually observable and objectively quantitative evaluation.
{"title":"Automated Microwave Tomography (MWT) Image Segmentation: State-of-the-Art Implementation and Evaluation","authors":"Yuchong Zhang, Y. Ma, Adel Omrani Hamzekalaei, Rahul Yadav, M. Fjeld, M. Fratarcangeli","doi":"10.24132/csrn.2020.3001.15","DOIUrl":"https://doi.org/10.24132/csrn.2020.3001.15","url":null,"abstract":"Inspired by the high performance in image-based medical analysis, this paper explores the use of advanced segmentation techniques for industrial Microwave Tomography (MWT). Our context is the visual analysis of moisture levels in porous foams undergoing microwave drying. We propose an automatic segmentation technique—MWT Segmentation based on K -means (MWTS-KM) and demonstrate its efficiency and accuracy for industrial use. MWTS-KM consists of three stages: image augmentation, grayscale conversion, and K -means implementation. To estimate the performance of this technique, we empirically benchmark its efficiency and accuracy against two well-established alternatives: Otsu and K -means. To elicit performance data, three metrics (Jaccard index, Dice coefficient and false positive) are used. Our results indicate that MWTS-KM outperforms the well-established Otsu and K -means, both in visually observable and objectively quantitative evaluation.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123508241","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 : 1900-01-01DOI: 10.24132/csrn.2019.2902.2.6
Bouressace Hassina
Recent studies on text line segmentation have not focused on title segmentation in complex structure documents, which may represent the upper rows in each article of a document page. Many methods cannot correctly distinguish between the titles and the text, especially when it contains more than one title. In this paper, we discuss this problem and then present a straightforward and robust title segmentation approach. The proposed method was tested on PATD (Printed Arabic Text Database ) images and we achieved good results.
{"title":"Title Segmentation in Arabic Document Pages","authors":"Bouressace Hassina","doi":"10.24132/csrn.2019.2902.2.6","DOIUrl":"https://doi.org/10.24132/csrn.2019.2902.2.6","url":null,"abstract":"Recent studies on text line segmentation have not focused on title segmentation in complex structure documents,\u0000which may represent the upper rows in each article of a document page. Many methods cannot correctly distinguish\u0000between the titles and the text, especially when it contains more than one title. In this paper, we discuss this problem\u0000and then present a straightforward and robust title segmentation approach. The proposed method was tested on\u0000PATD (Printed Arabic Text Database ) images and we achieved good results.","PeriodicalId":322214,"journal":{"name":"Computer Science Research Notes","volume":"253 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115941673","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：