Loading...
Search for:
segmentation
0.008 seconds
Total 331 records
Class Attention Map Distillation In Semantic Segmentation
, M.Sc. Thesis Sharif University of Technology ; Kasaei, Shohreh (Supervisor)
Abstract
Semantic segmentation is the tash of labeling each pixel of an input image. It is one of the main problems in computer vision and plays an important role in scene understanding. State of the art methods of solving it are based on Convolutional Neural Networhs (CNNs). While many real world tasks like autodriving cars and robot navigation require fast and lightweight models, CNNs inherently tend to give beter accuracy when they are deeper and bigger, and this has raised interest in designing compact networks. Knowledge distillation is one of the popular methods of training compact networhs and helps to transfer a big and powerful network’s knowledge to a small and compact one. In this research...
α-Visibility
, Article Computational Geometry: Theory and Applications ; Vol. 47, issue. 3 PART A , April , 2014 , pp. 435-446 ; ISSN: 09257721 ; ISBN: 9783642311543 ; Maheshwari, A ; Nouri-Baygim, M ; Sack, J. R ; Zarrabi-Zadeh, H ; Sharif University of Technology
Abstract
We study a new class of visibility problems based on the notion of α-visibility. Given an angle α and a collection of line segments S in the plane, a segment t is said to be α-visible from a point p, if there exists an empty triangle with one vertex at p and the side opposite to p on t such that the angle at p is α. In this model of visibility, we study the classical variants of point visibility, weak and complete segment visibility, and the construction of the visibility graph. We also investigate the natural query versions of these problems, when α is either fixed or specified at query time
Investigation of low back pain using system modeling
, Article Advanced Science Letters ; Volume 19, Issue 5 , 2013 , Pages 1260-1264 ; 19366612 (ISSN) ; Malik, A. S ; Xia, L ; Wang, J. L ; Nikkhoo, M ; Parnianpour, M ; Khan, M. I ; Sharif University of Technology
2013
Abstract
This paper focuses on characterization of intervertebral disc behavior upon fatigue stress. It consists in in-vitro experiments to measure intervertebral disc deformation with regard to forces applied to produce fatigues test. In this paper, a system modeling based non-invasive method to assess back pain is proposed which may be used for designing of prediction model to predict the failure of intervertebral disc (IVD) experiencing fatigue loading. An artificial compression fracture was simulated in the lower level of vertebra followed by a poly(methyl methacrylate) (PMMA) bone cement injection. Fatigue loading was applied on the specimens and the load was incrementally increased form 650 N...
Automatic brain tissue detection in MRI images using seeded region growing segmentation and neural network classification
, Article Australian Journal of Basic and Applied Sciences ; Volume 5, Issue 8 , 2011 , Pages 1066-1079 ; 19918178 (ISSN) ; Kasaei, S ; Sharif University of Technology
2011
Abstract
This paper presents a neural network-based method for automatic classification of magnetic resonance images (MRI) of brain under three categories of normal, lesion benign, and malignant. The proposed technique consists of six subsequent stages; namely, preprocessing, seeded region growing segmentation, connected component labeling (CCL), feature extraction, feature Dimension Reduction, and classification. In the preprocessing stage, the enhancement and restoration techniques are used to provide a more appropriate image for the subsequent automated stages. In the second stage, the seeded region growing segmentation is used for partitioning the image into meaningful regions. In the third...
Segmental HMM-based part-of-speech tagger
, Article 2010 International Conference on Audio, Language and Image Processing, ICALIP 2010, Shanghai, 23 November 2010 through 25 November 2010 ; 2010 , Pages 52-56 ; 9781424458653 (ISBN) ; Sameti, H ; Bahrani, M ; Babaali, B ; Sharif University of Technology
2010
Abstract
This paper presents a solution in order to solve the problem of using HMM-based POS tagger in some languages where a word can be comprised of several tokens. Viterbi algorithm is modified in order to support segment of words within a model state. In the other word, the proposed system has a built-in tokenizer where indicates words boundaries as well as its corresponding tag sequence
A model-based approach for estimation of changes in lumbar segmental kinematics associated with alterations in trunk muscle forces
, Article Journal of Biomechanics ; 2017 ; 00219290 (ISSN) ; Arjmand, N ; Meakin, J. R ; Bazrgari, B ; Sharif University of Technology
Elsevier Ltd
2017
Abstract
The kinematics information from imaging, if combined with optimization-based biomechanical models, may provide a unique platform for personalized assessment of trunk muscle forces (TMFs). Such a method, however, is feasible only if differences in lumbar spine kinematics due to differences in TMFs can be captured by the current imaging techniques. A finite element model of the spine within an optimization procedure was used to estimate segmental kinematics of lumbar spine associated with five different sets of TMFs. Each set of TMFs was associated with a hypothetical trunk neuromuscular strategy that optimized one aspect of lower back biomechanics. For each set of TMFs, the segmental...
Covering orthogonal polygons with sliding k-transmitters
, Article Theoretical Computer Science ; Volume 815 , May , 2020 , Pages 163-181 ; Seddighin, S ; Ghodsi, M ; Sharif University of Technology
Elsevier B. V
2020
Abstract
In this paper, we consider a new variant of covering in an orthogonal art gallery problem where each guard is a sliding k-transmitter. Such a guard can travel back and forth along an orthogonal line segment, say s, inside the polygon. A point p is covered by this guard if there exists a point q∈s such that pq‾ is a line segment normal to s, and has at most k intersections with the boundary walls of the polygon. The objective is to minimize the sum of the lengths of the sliding k-transmitters to cover the entire polygon. In other words, the goal is to find the minimum total length of trajectories on which the guards can travel to cover the entire polygon. We prove that this problem is NP-hard...
An optimization based approach embedded in a fuzzy connectivity algorithm for airway tree segmentation
, Article Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology", 20 August 2008 through 25 August 2008, Vancouver, BC ; 2008 , Pages 4011-4014 ; 9781424418152 (ISBN) ; Ahmadian, A. R ; Fatemizadeh, E ; Alirezaie, J ; Sharif University of Technology
2008
Abstract
The main problem with airway segmentation methods which significantly influences their accuracy is leakage into the extra-luminal regions due to thinness of the airway wall during the process of segmentation. This phenomenon potentially makes large regions of lungparenchyma to be wrongly identified as airways. A solution to this problem in the previous methods was based on leak detection followed by reducing leakage during the segmentation process. This has been dealt with adjusting the segmentation parameters and performing the re-segmentation process on the pre-segmented area. This makes the algorithm very exhaustive and more dependent on the user interaction. The method presented here is...
Evolution of multiple states machines for recognition of online cursive handwriting
, Article 2006 World Automation Congress, WAC'06, Budapest, 24 June 2006 through 26 June 2006 ; 2006 ; 1889335339 (ISBN); 9781889335339 (ISBN) ; Shouraki, S. B ; Hassanpour, S ; Sharif University of Technology
IEEE Computer Society
2006
Abstract
Recognition of cursive handwritings such as Persian script is a hard task as there is no fixed segmentation and simultaneous segmentation and recognition is required. This paper presents a novel comparison method for such tasks which is based on a Multiple States Machine to perform robust elastic comparison of small segments with high speed through generation and maintenance of a set of concurrent possible hypotheses, The approach is implemented on Persian (Farsi) language using a typical feature set and a specific tailored genetic algorithm and the recognition and computation time is compared with dynamic programming comparison approach. Copyright - World Automation Congress (WAC) 2006
Development of an integrated model for airflow in building spaces
, Article Renewable Energy ; Volume 31, Issue 4 , 2006 , Pages 401-416 ; 09601481 (ISSN) ; Sattari, S ; Sharif University of Technology
2006
Abstract
In building energy simulation, an integrated modelling of airflow in the building needed. Therefore, in this paper two approaches are used for building energy simulation: zonal network for modelling of the building segments and Computational Fluid Dynamics (CFD) for modelling of the airflow. It is noted that a synchronize solution process is needed for the building and the CFD equation-sets. For this purpose an iterative procedure is used to corresponding solution of these equations. © 2005 Elsevier Ltd. All rights reserved
Large-scale image annotation using prototype-based models
, Article ISPA 2011 - 7th International Symposium on Image and Signal Processing and Analysis ; 2011 , Pages 449-454 ; 9789531841597 (ISBN) ; Jamzad, M ; European Association for Signal Processing (EURASIP); IEEE Signal Processing Society; IEEE Region 8; IEEE Croatia Section; IEEE Croatia Section Signal Processing Chapter ; Sharif University of Technology
Abstract
Automatic image annotation is a challenging problem in the field of image retrieval. Dealing with large databases makes the annotation problem more difficult and therefore an effective approach is needed to manage such databases. In this work, an annotation system has been developed which considers images in separate categories and constructs a profiling model for each category. To describe an image, we propose a new feature extraction method based on color and texture information that describes image content using discrete distribution signatures. Image signatures of one category are partitioned using spectral clustering and a prototype is determined for each cluster by solving an...
Leukocyte's nucleus segmentation using active contour in YCbCr colour space
, Article Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010, 30 November 2010 through 2 December 2010, Kuala Lumpur ; 2010 , Pages 257-260 ; 9781424476008 (ISBN) ; Jahed, M ; Salehian, P ; Eslami, A ; Sharif University of Technology
2010
Abstract
Blood cell segmentation is a crucial part of many medical and laboratory procedures such as cell counting and blood cell disorder diagnosis. Among different types of blood cells, white blood cells are the most important clinically, as they suffer greatest from blood disorders. In this paper we propose a method for automatic segmentation of white blood cells nucleus. A distinctive function is used in YCbCr color space to segment the white blood cells nucleus. Next, the sub-images are extracted which contain the whole body of white blood cell nucleus. Then an active contour method is applied to the sub-images extracted from the previous step to accurately segment the cell nucleus boundary. Our...
ECG segmentation and fiducial point extraction using multi hidden Markov model
, Article Computers in Biology and Medicine ; Volume 79 , 2016 , Pages 21-29 ; 00104825 (ISSN) ; Shamsollahi, M. B ; Sayadi, O ; Armoundas, A. A ; Jutten, C ; Sharif University of Technology
Elsevier Ltd
Abstract
In this paper, we propose a novel method for extracting fiducial points (FPs) of electrocardiogram (ECG) signals. We propose the use of multi hidden Markov model (MultiHMM) as opposed to the traditional use of Classic HMM. In the MultiHMM method, each segment of an ECG beat is represented by a separate ergodic continuous density HMM. Each HMM has different state number and is trained separately. In the test step, the log-likelihood of two consecutive HMMs is compared and a path is estimated, which shows the correspondence of each part of the ECG signal to the HMM with the maximum log-likelihood. Fiducial points are estimated from the obtained path. For performance evaluation, the Physionet...
3D reconstruction of non-rigid surfaces from realistic monocular video
, Article 9th Iranian Conference on Machine Vision and Image Processing, 18 November 2015 through 19 November 2015 ; Volume 2016-February , 2015 , Pages 199-202 ; 21666776 (ISSN) ; 9781467385398 (ISBN) ; Kasaei, S ; Sharif University of Technology
IEEE Computer Society
Abstract
A novel algorithm for recovering the 3D shape of deformable objects purely from realistic monocular video is presented in this paper. Unlike traditional non-rigid structure from motion (NRSfM) methods, which have been studied only on synthetic datasets and controlled lab environments that needs some prior constraints (such as manually segmented objects, limited rotations and occlusions, or full-length trajectories), the proposed method has been described and tested on realistic video sequences, which have been downloaded from some social networks (such as Facebook and Twitter). In order to apply NRSfM to the realistic video sequences, because of no-prior information about the scene and...
Semantic segmentation of RGB-D images using 3D and local neighbouring features
, Article 2015 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2015, 23 November 2015 through 25 November 2015 ; 2015 ; 9781467367950 (ISBN) ; Kasaei, S ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2015
Abstract
3D scene understanding is one of the most important problems in the field of computer vision. Although, in the past decades, considerable attention has been devoted on the 2D scene understanding problem, now with the development of the depth sensors (like Microsoft Kinect), the 3D scene understanding has become a very challenging task. Traditionally, the scene understanding problem was considered as the semantic labeling of each image pixel. Semantic labeling of RGB-D images has not attained a comparable success, as the RGB semantic labeling, due to the lack of a challenging dataset. With the introduction of an RGB-D dataset, called NYU-V2, it became possible to propose a novel method to...
A novel Markov random field model based on region adjacency graph for T1 magnetic resonance imaging brain segmentation
, Article International Journal of Imaging Systems and Technology ; Volume 27, Issue 1 , 2017 , Pages 78-88 ; 08999457 (ISSN) ; Yousefi, S ; Manzuri Shalmani, M. T ; Sharif University of Technology
John Wiley and Sons Inc
2017
Abstract
Tissue segmentation in magnetic resonance brain scans is the most critical task in different aspects of brain analysis. Because manual segmentation of brain magnetic resonance imaging (MRI) images is a time-consuming and labor-intensive procedure, automatic image segmentation is widely used for this purpose. As Markov Random Field (MRF) model provides a powerful tool for segmentation of images with a high level of artifacts, it has been considered as a superior method. But because of the high computational cost of MRF, it is not appropriate for online processing. This article has proposed a novel method based on a proper combination of MRF model and watershed algorithm in order to alleviate...
Bi-directional ConvLSTM U-net with densley connected convolutions
, Article 17th IEEE/CVF International Conference on Computer Vision Workshop, ICCVW 2019, 27 October 2019 through 28 October 2019 ; 2019 , Pages 406-415 ; 9781728150239 (ISBN) ; Asadi Aghbolaghi, M ; Fathy, M ; Escalera, S ; Computer Vision Foundation; IEEE ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2019
Abstract
In recent years, deep learning-based networks have achieved state-of-the-art performance in medical image segmentation. Among the existing networks, U-Net has been successfully applied on medical image segmentation. In this paper, we propose an extension of U-Net, Bi-directional ConvLSTM U-Net with Densely connected convolutions (BCDU-Net), for medical image segmentation, in which we take full advantages of U-Net, bi-directional ConvLSTM (BConvLSTM) and the mechanism of dense convolutions. Instead of a simple concatenation in the skip connection of U-Net, we employ BConvLSTM to combine the feature maps extracted from the corresponding encoding path and the previous decoding up-convolutional...
An efficient uniform-segmented neuron model for large-scale neuromorphic circuit design: Simulation and FPGA synthesis results
, Article IEEE Transactions on Circuits and Systems I: Regular Papers ; Volume 66, Issue 6 , 2019 , Pages 2336-2349 ; 15498328 (ISSN) ; Abolfathi, H ; Ahmadi, A ; Ahmadi, M ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2019
Abstract
Large-scale simulation of spiking neural networks on hardware with a remarkable resemblance to their mathematical models is a key objective of the neuromorphic discipline. This issue is, however, considerably resource-intensive due to the presence of nonlinear terms in neuron models. This paper proposes a novel uniform piecewise linear segmentation approach for nonlinear function evaluations. Employing the proposed approach, we present a uniform-segmented adaptive exponential neuron model capable of accurately producing various responses exhibited by the original model and suitable for efficient large-scale implementation. In contrast to previous nonuniform-segmented neuron models, the...
Color Image Segmentation Using a Fuzzy Inference System
, Article 7th International Conference on Digital Information Processing and Communications, ICDIPC 2019, 2 May 2019 through 4 May 2019 ; 2019 , Pages 78-83 ; 9781728132969 (ISBN) ; Macktoobian, M ; Kasaei, S ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2019
Abstract
A novel method is proposed in the scope of image segmentation that solves this problem by breaking it into two main blocks. The first block's functionality is a method to anticipate the color basis of each segment in segmented images. One of the challenges of image segmentation is the inappropriate distribution of colors in the RGB color space. To determine the color of each segment, after mapping the input image onto the HSI color space, the image colors are classified into some clusters by exploiting the K-Means. Then, the list of cluster centers is winnowed down to a short list of colors based on a set of criteria. The second block of the proposed method defines how each pixel of the...
Personalized computational human phantoms via a hybrid model-based deep learning method
, Article 15th IEEE International Symposium on Medical Measurements and Applications, MeMeA 2020, 1 June 2020 through 3 June 2020 ; July , 2020 ; Bitarafan, A ; Dylov, D. V ; Soleymani Baghshah, M ; Hosseini, S. A ; IEEE; IEEE Instrumentation and Measurement Society; IEEE Sensors Council Italy Chapter; Politecnica di Bari; Politecnico di Torino; Societa Italiana di Analisi del Movimento in Clinica ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2020
Abstract
Computed tomography (CT) simulators are versatile tools for scanning protocol evaluation, optimization of geometrical design parameters, assessment of image reconstruction algorithms, and evaluation of the impact of future innovations attempting to improve the performance of CT scanners. Computational human phantoms (CHPs) play a key role in simulators for the radiation dosimetry and assessment of image quality tasks in the medical x-ray systems. Since the construction of patient-specific CHPs can be both difficult and time-consuming, nominal standard/reference CHPs have been established, yielding significant discrepancies in the special design and optimization demands of patient dose and...