scispace - formally typeset
Search or ask a question

Showing papers by "Santanu Chaudhury published in 2018"


Journal ArticleDOI
TL;DR: Experimental evaluations show that the proposed DRNN outperforms the state-of-the-art despeckling approaches in terms of the structural similarity index measure, peak signal to noise ratio, edge preservation index, and speckle region's signal-to- noise ratio.
Abstract: In this letter, we aim to develop a deep adversarial despeckling approach to enhance the quality of ultrasound images. Most of the existing approaches target a complete removal of speckle, which produces oversmooth outputs and results in loss of structural details. In contrast, the proposed approach reduces the speckle extent without altering the structural and qualitative attributes of the ultrasound images. A despeckling residual neural network (DRNN) is trained with an adversarial loss imposed by a discriminator. The discriminator tries to differentiate between the despeckled images generated by the DRNN and the set of high-quality images. Further to prevent the developed network from oversmoothing, a structural loss term is used along with the adversarial loss. Experimental evaluations show that the proposed DRNN outperforms the state-of-the-art despeckling approaches in terms of the structural similarity index measure, peak signal to noise ratio, edge preservation index, and speckle region's signal to noise ratio.

44 citations


Journal ArticleDOI
TL;DR: In the subjective evaluation, performed by the expert radiologists, the proposed filter’s outputs are preferred for the improved contrast and sharpness of the object boundaries and the proposed filtering framework is suitable to reduce the unwanted speckle and improve the quality of the ultrasound images.
Abstract: Anisotropic diffusion filters are one of the best choices for speckle reduction in the ultrasound images. These filters control the diffusion flux flow using local image statistics and provide the desired speckle suppression. However, inefficient use of edge characteristics results in either oversmooth image or an image containing misinterpreted spurious edges. As a result, the diagnostic quality of the images becomes a concern. To alleviate such problems, a novel anisotropic diffusion-based speckle reducing filter is proposed in this paper. A probability density function of the edges along with pixel relativity information is used to control the diffusion flux flow. The probability density function helps in removing the spurious edges and the pixel relativity reduces the oversmoothing effects. Furthermore, the filtering is performed in superpixel domain to reduce the execution time, wherein a minimum of 15% of the total number of image pixels can be used. For performance evaluation, 31 frames of three synthetic images and 40 real ultrasound images are used. In most of the experiments, the proposed filter shows a better performance as compared to the state-of-the-art filters in terms of the speckle region’s signal-to-noise ratio and mean square error. It also shows a comparative performance for figure of merit and structural similarity measure index. Furthermore, in the subjective evaluation, performed by the expert radiologists, the proposed filter’s outputs are preferred for the improved contrast and sharpness of the object boundaries. Hence, the proposed filtering framework is suitable to reduce the unwanted speckle and improve the quality of the ultrasound images.

26 citations


Proceedings ArticleDOI
27 May 2018
TL;DR: A pipelined network comprising of a convolutional neural network followed by unsupervised clustering is proposed to perform vessel segmentation in liver ultrasound images, motivated by the tremendous success of CNNs in object detection and localization.
Abstract: Vascular region segmentation in ultrasound images is necessary for applications like automatic registration, and surgical navigation. In this paper, a pipelined network comprising of a convolutional neural network (CNN) followed by unsupervised clustering is proposed to perform vessel segmentation in liver ultrasound images. The work is motivated by the tremendous success of CNNs in object detection and localization. CNN here is trained to localize vascular regions, which are subsequently segmented by the clustering. The proposed network results in 99.14% pixel accuracy and 69.62% mean region intersection over union on 132 images. These values are better than some existing methods.

19 citations


Journal ArticleDOI
TL;DR: The effectiveness of the proposed scheme is demonstrated by comparing it with eighteen state-of-the-art local change detection algorithms and the performance is carried out using one subjective and three quantitative evaluation measures.
Abstract: In this article, a local change detection technique for underwater video sequences is proposed to detect the positions of the moving objects. The proposed change detection scheme integrates the Mixture of Gaussian (MoG) process in a Wronskian framework. It uses spatiotemporal modes (an integration of spatio-contextual and temporal modes) arising over the underwater video sequences to detect the local changes. The Wronskian framework takes care of the spatio-contextual modes whereas MoG models the temporal modes arising due to inter-dependency of a pixel in a video. The proposed scheme follows two steps: background construction and background subtraction. It takes initial few frames to construct a background model and thereby detection of the moving objects in the subsequent frames. During background construction stage; the linear dependency test between the region of supports/ local image patch in the target image frame and the reference background model are carried out using the Wronskian change detection model. The pixel values those are linearly dependent are assumed to be generated from an MoG process and are modeled using the same. Once the background is constructed, then the background subtraction and update process starts from the next frame. The efficiency of the proposed scheme is validated by testing it on two benchmark underwater video databases: fish4knowledge and underwaterchangedetection and one large scale outdoor video database: changedetection.net. The effectiveness of the proposed scheme is demonstrated by comparing it with eighteen state-of-the-art local change detection algorithms. The performance of the proposed scheme is carried out using one subjective and three quantitative evaluation measures.

17 citations


Book ChapterDOI
08 Sep 2018
TL;DR: A novel automated pipeline leveraging deep convolutional neural networks for stomata detection and its quantification shows a superior performance in contrast to the existing stomATA detection methods in terms of precision and recall.
Abstract: Analysis of stomata density and its configuration based on scanning electron microscopic (SEM) image of a leaf surface, is an effective way to characterize the plant’s behaviour under various environmental stresses (drought, salinity etc.). Existing methods for phenotyping these stomatal traits are often based on manual or semi-automatic labeling and segmentation of SEM images. This is a low-throughput process when large number of SEM images is investigated for statistical analysis. To overcome this limitation, we propose a novel automated pipeline leveraging deep convolutional neural networks for stomata detection and its quantification. The proposed framework shows a superior performance in contrast to the existing stomata detection methods in terms of precision and recall, 0.91 and 0.89 respectively. Furthermore, the morphological traits (i.e. length & width) obtained at stomata quantification step shows a correlation of 0.95 and 0.91 with manually computed traits, resulting in an efficient and high-throughput solution for stomata phenotyping.

15 citations


Journal ArticleDOI
TL;DR: A novel anatomically informed (AI) score that evaluates fitness of a given connectivity structure to both DTI and fMRI data simultaneously is proposed that is employed in structure learning of DBN given the data.
Abstract: Objective: Effective connectivity (EC) is the methodology for determining functional-integration among the functionally active segregated regions of the brain. By definition [1] EC is “the causal influence exerted by one neuronal group on another” which is constrained by anatomical connectivity (AC) (axonal connections). AC is necessary for EC but does not fully determine it, because synaptic communication occurs dynamically in a context-dependent fashion. Although there is a vast emerging evidence of structure–function relationship using multimodal imaging studies, till date only a few studies have done joint modeling of the two modalities: functional MRI (fMRI) and diffusion tensor imaging (DTI). We aim to propose a unified probabilistic framework that combines information from both sources to learn EC using dynamic Bayesian networks (DBNs). Method: DBNs are probabilistic graphical temporal models that learn EC in an exploratory fashion. Specifically, we propose a novel anatomically informed (AI) score that evaluates fitness of a given connectivity structure to both DTI and fMRI data simultaneously. The AI score is employed in structure learning of DBN given the data. Results: Experiments with synthetic-data demonstrate the face validity of structure learning with our AI score over anatomically uninformed counterpart. Moreover, real-data results are cross-validated by performing classification-experiments. Conclusion: EC inferred on real fMRI-DTI datasets is found to be consistent with previous literature and show promising results in light of the AC present as compared to other classically used techniques such as Granger-causality. Significance: Multimodal analyses provide a more reliable basis for differentiating brain under abnormal/diseased conditions than the single modality analysis.

13 citations


Journal ArticleDOI
TL;DR: A latent Dirichlet allocation (LDA) based model is proposed to represent temporal behaviour of mobile subscribers as compact and interpretable profiles and makes use of the structural regularity within the observable data corresponding to a large number of user profiles and relaxes the strict temporal ordering of user preferences in TBS clustering.
Abstract: Customer segmentation based on temporal variation of subscriber preferences is useful for communication service providers (CSPs) in applications such as targeted campaign design, churn prediction, and fraud detection. Traditional clustering algorithms are inadequate in this context, as a multidimensional feature vector represents a subscriber profile at an instant of time, and grouping of subscribers needs to consider variation of subscriber preferences across time. Clustering in this case usually requires complex multivariate time series analysis-based models. Because conventional time series clustering models have limitations around scalability and ability to accurately represent temporal behaviour sequences (TBS) of users, that may be short, noisy, and non-stationary, we propose a latent Dirichlet allocation (LDA) based model to represent temporal behaviour of mobile subscribers as compact and interpretable profiles. Our model makes use of the structural regularity within the observable data corresponding to a large number of user profiles and relaxes the strict temporal ordering of user preferences in TBS clustering. We use mean-shift clustering to segment subscribers based on their discovered profiles. Further, we mine segment-specific association rules from the discovered TBS clusters, to aid marketers in designing intelligent campaigns that match segment preferences. Our experiments on real world data collected from a popular Asian communication service provider gave encouraging results.

12 citations


Journal ArticleDOI
TL;DR: A novel camera-based traffic monitoring and prediction scheme without identifying or tracking vehicles that learns the road state pattern using dynamic Bayesian network and predicts the future road traffic state within a specific time delay is proposed.
Abstract: The varied road conditions, chaotic and unstructured traffic, lack of lane discipline and wide variety of vehicles in countries like India, Pakistan and so on pose a need for a novel traffic monitoring system. In this study, the authors propose a novel camera-based traffic monitoring and prediction scheme without identifying or tracking vehicles. Spatial interest points (SIPs) and spatio-temporal interest points (STIPs) are extracted from the video stream of road traffic. SIP represents the number of vehicles and STIP represents the number of moving vehicles. The distributions of these features are then classified using Gaussian mixture model. In the proposed method, they learn the road state pattern using dynamic Bayesian network and predict the future road traffic state within a specific time delay. The predicted road state information can be used for traffic planning. The proposed method is computationally light, yet very powerful and efficient. The algorithm is tested for different weather conditions as well. They have validated their algorithm using Synchro Studio simulator and got 95.7% as average accuracy and on real-time video we got an accuracy of 84%.

11 citations


Journal ArticleDOI
TL;DR: Assessment of the correlation between genotypic and phenotypic traits that can contribute towards the emerging field of rice phenomics finds that there is a notable difference in gene expression of OsPIP2;5 and OsNip2;1 in various indica varieties of rice at different time periods of stress.

9 citations


Proceedings ArticleDOI
01 Aug 2018
TL;DR: A novel automated framework for stomata quantification is proposed based on a hybrid approach where the candidateStomata region is first detected by a convolutional neural network and the occlusion is dealt with an inpainting algorithm to solve the problem of shape, scale and Occlusion in an end-to-end manner.
Abstract: Stomatal morphology is a key phenotypic trait for plants' response analysis under various environmental stresses (e.g. drought, salinity etc.). Stomata exhibit diverse characteristics with respect to orientation, size, shape and varying degree of papillae occlusion. Thus, the biologists currently rely on manual or semi-automatic approaches to accurately compute its morphological traits based on scanning electron microscopic (SEM) images of leaf surface. In contrast to these subjective and low-throughput methods, we propose a novel automated framework for stomata quantification. It is realized based on a hybrid approach where the candidate stomata region is first detected by a convolutional neural network (CNN) and the occlusion is dealt with an inpainting algorithm. In addition, we propose stomata segmentation based quantification framework to solve the problem of shape, scale and occlusion in an end-to-end manner. The performance of the proposed automated frameworks is evaluated by comparing the derived traits with manually computed morphological traits of stomata. With no prior information about its size and location, the hybrid and end-to-end machine learning frameworks shows a correlation of 0.94 and 0.93, respectively on rice stomata images. Furthermore, they successfully enable wheat stomata quantification showing generalizability in terms of cultivars.

7 citations


Proceedings ArticleDOI
10 Oct 2018
TL;DR: Continuous video stream of data captured by CCTV cameras can be processed on the fly to give real-time alerts to concerned authorities and these alerts can be disseminated using e-mail, text messaging, on-screen alerts and alarms.
Abstract: In this paper, we have proposed an ontology-based context-aware framework for providing intelligent services such as smart surveillance, which employ IoT technologies to ensure better quality of life in a smart city. An IoT network such as a smart surveillance system combines the working of Closed-circuit television (CCTV) cameras and various sensors to perform real-time computation for identifying threats and critical situations with the help of valuable context information. This information is perceptual in nature and needs to be converted into higher-level abstractions that can further be used for reasoning to recognize situations. Semantic abstractions for perceptual inputs are possible with the use of a multimedia ontology encoded using Multimedia Web Ontology Language (MOWL) that helps to define concepts, properties and structure of a possible environment. MOWL also allows for a dynamic modeling of real-time situations by employing Dynamic Bayesian networks (DBN), which suits the requirements of a intelligent IoT system. In this paper, we show the application of this framework in a smart surveillance system. Surveillance is enhanced by not only helping to analyze past events, but by predicting dangerous situations for which preventive actions can be taken. In our proposed approach, continuous video stream of data captured by CCTV cameras can be processed on the fly to give real-time alerts to concerned authorities. These alerts can be disseminated using e-mail, text messaging, on-screen alerts and alarms.

Book ChapterDOI
07 Dec 2018
TL;DR: There is a scope for improvement in the reading comprehension, by changing the physical properties of the document without changing its content, when the same document is read in different font type.
Abstract: In this world of digitization, screen reading has grown immensely due to the availability of affordable display devices. Most of the people prefer to read on display devices as compared to the print media. To make the reading experience of the reader pleasant and comfortable, the font designers strive hard to choose suitable typographical properties of the text such as font type, font size etc. Some of the researchers suggest that the typography of the text affects the reading performance of the readers to some extent. However, the research focusing on the effect of typography on the reading behavior of the readers is limited and it is hardly touched upon for the Indian scripts. Therefore, the proposed paper aims to find out the effect of Devanagari font type on the reading performance, especially reading comprehension of the readers. In addition to this, a method to reduce the error in the gaze estimation of the eye tracker is also proposed. In order to understand the reading behavior, an eye tracking experiment is performed on 14 participants asking them to read 22 pages, in 3 different font types, presented on the screen of the eye tracker. The performance of the readers is analyzed in terms of total reading time, comprehension score, number of fixations, fixation duration and number of regressions. Our results show that there is a significant difference in the fixation duration, a number of fixations and the comprehension score, when the same document is read in different font type. Thus, there is a scope for improvement in the reading comprehension, by changing the physical properties of the document without changing its content. These findings might be useful to understand the readers’ preference for the font and to design a proper font type for online reading.

Proceedings ArticleDOI
01 Aug 2018
TL;DR: This paper proposes a three level hierarchical estimation framework that employs heavy-tailed priors for block sparse modeling and variational inference for Bayesian estimation and demonstrates the applicability of this framework in telemonitoring of Fetal Electrocardiogram.
Abstract: This paper addresses the problem of Bayesian Block Sparse Modeling when coefficients within the blocks are correlated. In contrast to the current hierarchical methods which do not exploit correlation structure within the blocks, we propose a three level hierarchical estimation framework. It employs heavy-tailed priors for block sparse modeling and variational inference for Bayesian estimation. This paper also describes the relationship between proposed framework and some of the existing Block Sparse Bayesian Learning (SBL) methods and show that these SBL methods can be viewed as its special cases. Extensive experimental results for synthetic signals are provided, demonstrating the superior performance of the proposed framework in terms of failure rate, relative reconstruction error, to name a few. We also demonstrate the applicability of this framework in telemonitoring of Fetal Electrocardiogram.

Proceedings ArticleDOI
01 Aug 2018
TL;DR: A convolutional neural network is developed which learns to remove speckle from US images using the outputs of these classical approaches and is able to outperform the state-of-the-art despeckling approaches and produces the outputs even better than the ensembles for certain images.
Abstract: Ultrasound (US) image despeckling is a problem of high clinical importance. Machine learning solutions to the problem are considered impractical due to the unavailability of speckle-free US image dataset. On the other hand, the classical approaches, which are able to provide the desired outputs, have limitations like input dependent parameter tuning. In this work, a convolutional neural network (CNN) is developed which learns to remove speckle from US images using the outputs of these classical approaches. It is observed that the existing approaches can be combined in a complementary manner to generate an output better than their individual outputs. Thus, the CNN is trained using the individual outputs as well as the output ensembles. It eliminates the cumbersome process of parameter tuning required by the existing approaches for every new input. Further, the proposed CNN is able to outperform the state-of-the-art despeckling approaches and produces the outputs even better than the ensembles for certain images.

Posted Content
TL;DR: This paper addresses the problem of imposing desired modal properties on the generated space using a latent distribution, engineered in accordance with themodal properties of the true data distribution, by training a latent space inversion network in tandem with the generative network using a divergence loss.
Abstract: Generative adversarial networks (GANs) have shown remarkable success in generation of unstructured data, such as, natural images. However, discovery and separation of modes in the generated space, essential for several tasks beyond naive data generation, is still a challenge. In this paper, we address the problem of imposing desired modal properties on the generated space using a latent distribution, engineered in accordance with the modal properties of the true data distribution. This is achieved by training a latent space inversion network in tandem with the generative network using a divergence loss. The latent space is made to follow a continuous multimodal distribution generated by reparameterization of a pair of continuous and discrete random variables. In addition, the modal priors of the latent distribution are learned to match with the true data distribution using minimal-supervision with negligible increment in number of learnable parameters. We validate our method on multiple tasks such as mode separation, conditional generation, and attribute discovery on multiple real world image datasets and demonstrate its efficacy over other state-of-the-art methods.

Proceedings ArticleDOI
18 Dec 2018
TL;DR: Results show that the proposed temporal model is highly useful in processing SSVEP-EEG signals irrespective of the recognition algorithms used.
Abstract: Compressed Sensing (CS) has emerged as an alternate method to acquire high dimensional signals effectively by exploiting the sparsity assumption. However, owing to non-sparse and non-stationary nature, it is extremely difficult to process Electroencephalograph (EEG) signals using CS paradigm. The success of Bayesian algorithms in recovering non-sparse signals has triggered the research in CS based models for neurophysiological signal processing. In this paper, we address the problem of Temporal Modeling of EEG Signals using Block Sparse Variational Bayes (SVB) Framework. Temporal correlation of EEG signals is modeled blockwise using normal variance scale mixtures parameterized via some random and deterministic parameters. Variational inference is exploited to infer the random parameters and Expectation Maximization (EM) is used to obtain the estimate of deterministic parameters. To validate the framework, we present experimental results for benchmark State Visual Evoked Potential (SSVEP) dataset with 40-target Brain-Computer Interface (BCI) speller using two frequency recognition algorithms viz. Canonical Correlation Analysis (CCA) and L1-regularized Multiway CCA. Results show that the proposed temporal model is highly useful in processing SSVEP-EEG signals irrespective of the recognition algorithms used.

Posted Content
08 Nov 2018
TL;DR: This paper proposes a method that can generate samples conditioned on the properties of a latent distribution engineered in accordance with a certain data prior, and demonstrates that its model, despite being fully unsupervised, is effective in learning meaningful representations through its mode matching property.
Abstract: Generative adversarial networks (GANs) have shown remarkable success in generation of unstructured data, such as, natural images. However, discovery and separation of modes in the generated space, essential for several tasks beyond naive data generation, is still a challenge. In this paper, we address the problem of imposing desired modal properties on the generated space using a latent distribution, engineered in accordance with the modal properties of the true data distribution. This is achieved by training a latent space inversion network in tandem with the generative network using a divergence loss. The latent space is made to follow a continuous multimodal distribution generated by reparameterization of a pair of continuous and discrete random variables. In addition, the modal priors of the latent distribution are learned to match with the true data distribution using minimal-supervision with negligible increment in number of learnable parameters. We validate our method on multiple tasks such as mode separation, conditional generation, and attribute discovery on multiple real world image datasets and demonstrate its efficacy over other state-of-the-art methods.

Posted Content
TL;DR: In this article, an unsupervised deep adversarial approach was used to address the despeckling problem using an adversarial loss imposed by a discriminator to differentiate between the deseckled images generated by the DRNN and the set of high-quality images.
Abstract: Contrast and quality of ultrasound images are adversely affected by the excessive presence of speckle However, being an inherent imaging property, speckle helps in tissue characterization and tracking Thus, despeckling of the ultrasound images requires the reduction of speckle extent without any oversmoothing In this letter, we aim to address the despeckling problem using an unsupervised deep adversarial approach A despeckling residual neural network (DRNN) is trained with an adversarial loss imposed by a discriminator The discriminator tries to differentiate between the despeckled images generated by the DRNN and the set of high-quality images Further to prevent the developed DRNN from oversmoothing, a structural loss term is used along with the adversarial loss Experimental evaluations show that the proposed DRNN is able to outperform the state-of-the-art despeckling approaches

Book ChapterDOI
10 Oct 2018
TL;DR: This research focuses on an ontology-based context-aware framework for providing services such as smart surveillance and intelligent traffic monitoring, which employ IoT technologies to ensure better quality of life in a smart city.
Abstract: This research focuses on an ontology-based context-aware framework for providing services such as smart surveillance and intelligent traffic monitoring, which employ IoT technologies to ensure better quality of life in a smart city. An IoT network combines the working of Closed-circuit television (CCTV) cameras and various sensors to perform real-time computation for identifying threats, traffic conditions and other such situations with the help of valuable context information. This information is perceptual in nature and needs to be converted into higher-level abstractions that can further be used for reasoning to recognize situations. Semantic abstractions for perceptual inputs are possible with the use of a multimedia ontology which helps to define concepts, properties and structure of a possible environment. We have used Multimedia Web Ontology Language (MOWL) for semantic interpretation and handling inherent uncertainties in multimedia observations linked with the system. MOWL also allows for a dynamic modeling of real-time situations by employing Dynamic Bayesian networks (DBN), which suits the requirements of an intelligent IoT system. In this paper, we show the application of this framework in a smart surveillance system for traffic monitoring. Surveillance is enhanced by not only helping to analyze past events, but by predicting anomalous situations for which preventive actions can be taken. In our proposed approach, continuous video stream of data captured by CCTV cameras can be processed on-the-fly to give real-time alerts to security agencies. These alerts can be disseminated via e-mail, text messaging, on-screen and alarms not only to pedestrians and drivers in the locality but also the nearest police station and hospital in order to prevent and decrease the loss incurred by any event.

Posted Content
TL;DR: The proposed framework is extended to the registration of CT and US images, which is free from any data specific assumptions and has better generalization capability as compared to the existing rule-based/classical approaches.
Abstract: Image registration is a process of aligning two or more images of same objects using geometric transformation. Most of the existing approaches work on the assumption of location invariance. These approaches require object-centric images to perform matching. Further, in absence of intensity level symmetry between the corresponding points in two images, the learning based registration approaches rely on synthetic deformations, which often fail in real scenarios. To address these issues, a combination of convolutional neural networks (CNNs) to perform the desired registration is developed in this work. The complete objective is divided into three sub-objectives: object localization, segmentation and matching transformation. Object localization step establishes an initial correspondence between the images. A modified version of single shot multi-box detector is used for this purpose. The detected region is cropped to make the images object-centric. Subsequently, the objects are segmented and matched using a spatial transformer network employing thin plate spline deformation. Initial experiments on MNIST and Caltech-101 datasets show that the proposed model is able to produce accurate matching. Quantitative evaluation performed using dice coefficient (DC) and mean intersection over union (mIoU) show that proposed method results in the values of 79% and 66%, respectively for MNIST dataset and the values of 94% and 90%, respectively for Caltech-101 dataset. The proposed framework is extended to the registration of CT and US images, which is free from any data specific assumptions and has better generalization capability as compared to the existing rule-based/classical approaches.

Proceedings ArticleDOI
18 Dec 2018
TL;DR: Bayesian Sparse Signal Recovery (SSR) for Multiple Measurement Vectors, when elements of each row of solution matrix are correlated, is addressed and it can be seen that by exploiting temporal correlation information present in the successive image samples, the proposed framework can reconstruct the data with less linear random measurements with high fidelity.
Abstract: Bayesian Sparse Signal Recovery (SSR) for Multiple Measurement Vectors, when elements of each row of solution matrix are correlated, is addressed in the paper. We propose a standard linear Gaussian observation model and a three-level hierarchical estimation framework, based on Gaussian Scale Mixture (GSM) model with some random and deterministic parameters, to model each row of the unknown solution matrix. This hierarchical model induces heavy-tailed marginal distribution over each row which encompasses several choices of distributions viz. Laplace distribution, Student's t distribution and Jeffery prior. Automatic Relevance Determination (ARD) phenomenon introduces sparsity in the model. It is interesting to see that Block Sparse Bayesian Learning framework is a special case of the proposed framework when induced marginal is Jeffrey prior. Experimental results for synthetic signals are provided to demonstrate its effectiveness. We also explore the possibility of using Multiple Measurement Vectors to model Dynamic Hand Posture Database which consists of sequence of temporally correlated hand posture sequence. It can be seen that by exploiting temporal correlation information present in the successive image samples, the proposed framework can reconstruct the data with less linear random measurements with high fidelity.

Book ChapterDOI
20 Aug 2018
TL;DR: The effect of the threshold to prune out variance parameters of algorithms corresponding to several choices of marginals, viz. multivariate Jeffery prior, multivariate Laplace distribution and multivariate Student’s t distribution is discussed.
Abstract: We present some details of Bayesian block sparse modeling using hierarchical prior having deterministic and random parameters when entries within the blocks are correlated. In particular, the effect of the threshold to prune out variance parameters of algorithms corresponding to several choices of marginals, viz. multivariate Jeffery prior, multivariate Laplace distribution and multivariate Student’s t distribution, is discussed. We also provide details of experiments with Electroencephalograph (EEG) data which shed some light on the possible applicability of the proposed Sparse Variational Bayes framework.

Posted Content
01 May 2018
TL;DR: The proposed framework is extended to the registration of CT and US images, which is free from any data specific assumptions and has better generalization capability as compared to the existing rule based/classical approaches.
Abstract: Image registration requires simultaneous processing of multiple images to match the keypoints or landmarks of the contained objects. These images often come from different modalities for example CT and Ultrasound (US), and pose the challenge of establishing one to one correspondence. In this work, a novel pipeline of convolutional neural networks is developed to perform the desired registration. The complete objective is divided into three parts: localization of the object of interest, segmentation and matching transformation. Most of the existing approaches skip the localization step and are prone to fail in general scenarios. We overcome this challenge through detection which also establishes initial correspondence between the images. A modified version of single shot multibox detector is used for this purpose. The detected region is cropped and subsequently segmented to generate a mask corresponding to the desired object. The mask is used by a spatial transformer network employing thin plate spline deformation to perform the desired registration. Initial experiments on MNIST and Caltech-101 datasets show that the proposed model is able to accurately match the segmented images. The proposed framework is extended to the registration of CT and US images, which is free from any data specific assumptions and has better generalization capability as compared to the existing rule based/classical approaches.