Books by MALAY KUMAR KUNDU
Papers by MALAY KUMAR KUNDU
Progress In Electromagnetics Research B, 2011
The motivation behind fusing multimodality, multiresolution images is to create a single image wi... more The motivation behind fusing multimodality, multiresolution images is to create a single image with improved interpretability. In this paper, we propose a novel multimodality Medical Image Fusion (MIF) method, based on Ripplet Transform Type-I (RT) for spatially registered, multi-sensor, multi-resolution medical images. RT is a new Multi-scale Geometric Analysis (MGA) tool, capable of resolving two dimensional (2D) singularities and representing image edges more efficiently. The source medical images are first transformed by discrete RT (DRT). Different fusion rules are applied to the different subbands of the transformed images. Then inverse DRT (IDRT) is applied to the fused coefficients to get the fused image. The performance of the proposed scheme is evaluated by various quantitative measures like Mutual Information (MI), Spatial Frequency (SF), and Entropy (EN) etc. Visual and quantitative analysis shows, that the proposed technique performs better compared to fusion scheme based on Contourlet Transform (CNT).
Proceedings of the 2nd International Conference, Feb 26, 2015
We present a novel Content Based Medical Image Retrieval (CBMIR) scheme for color endoscopic imag... more We present a novel Content Based Medical Image Retrieval (CBMIR) scheme for color endoscopic images using Multiscale Geometric Analysis (MGA) of Nonsubsampled Contourlet Transform (NSCT) and the statistical framework based on Generalized Gaussian Density (GGD) model and Kullback-Leibler Distance (KLD). The subband images obtained from the NSCT decomposition are divided into number of blocks and then the coefficients of each block of each subband is modeled with GGD parameters and computing the similarity using the KLD among the model parameters. The retrieval performance of the proposed system is further improved using Least Square-Support Vector Machine (LSSVM) classifier. Extensive experiments were carried out to evaluate the effectiveness of the proposed system on endoscopic image databases consisting of 276 images. Experimental results show that the proposed CBMIR system performs efficiently in image retrieval paradigm.
While surveillance video is the biggest source of unstructured Big Data today, the emergence of h... more While surveillance video is the biggest source of unstructured Big Data today, the emergence of high-efficiency video coding (HEVC) standard is poised to have a huge role in lowering the costs associated with transmission and storage. Among the benefits of HEVC over the legacy MPEG-4 Advanced Video Coding (AVC), is a staggering 40 percent or more bitrate reduction at the same visual quality. Given the bandwidth limitations, video data are compressed essentially by removing spatial and temporal correlations that exist in its uncompressed form. This causes compressed data, which are already de-correlated, to serve as a vital resource for machine learning with significantly fewer samples for training. In this paper, an efficient approach to foreground extraction/segmentation is proposed using novel spatio-temporal decorrelated block features extracted directly from the HEVC compressed video. Most related techniques, in contrast, work on uncompressed images claiming significant storage and computational resources not only for the decoding process prior to initialization but also for the feature selection/extraction and background modeling stage following it. The proposed approach has been qualitatively and quantitatively evaluated against several other state-of-the-art methods.
Recently, Content Based Image Retrieval (CBIR) has emerged as an active research area having appl... more Recently, Content Based Image Retrieval (CBIR) has emerged as an active research area having applications in various fields. There exist several states-of-the art CBIR systems that uses both spatial and transform features as input. However, as hardly any details study reported so far on the effectiveness of different transform domain features in CBIR paradigm. This motivates the current article where we have presented extensive comparative assessment of five different transform domain features considering various filter combinations. Three different feature representation schemes and three different classifiers have been used for this purpose. Extensive experiments on four widely used benchmark image databases (Oliva, Caltech101, Caltech256 and MIRFlickr25000) were conducted to determine the best combination of transform, filters, feature representation and classifier. Furthermore, we have also attempted to discover the optimal features from the best combinations using maximal information compression index (MICI). Both qualitative and quantitative evaluations show that the combination of Least Square Support Vector Machine (LSSVM) as a classifier and the statistical parametric framework based reduced feature representation in Non-Subsampled Contourlet Transform (NSCT) with "pyrexc" and "sinc" filters gives the best retrieval performances.
Content-Based Image Retrieval (CBIR) is an important problem in the domain of digital data manage... more Content-Based Image Retrieval (CBIR) is an important problem in the domain of digital data management.
There is indeed a growing availability of images, but unfortunately the traditional metadata-based search
systems are unable to properly exploit their visual information content. In this article we introduce a
novel CBIR scheme that abstracts each image in the database in terms of statistical features computed
using the Multi-scale Geometric Analysis (MGA) of Non-subsampled Contourlet Transform (NSCT). Noise
resilience is one of the main advantages of this feature representation. To improve the retrieval
performance and reduce the semantic gap, our system incorporates a Relevance Feedback (RF) mechanism
that uses a graph-theoretic approach to rank the images in accordance with the user’s feedback.
First, a graph of images is constructed with edges reflecting the similarity of pairs of images with respect
to the proposed feature representation. Then, images are ranked at each feedback round in terms of the
probability that a random walk on this graph reaches an image tagged as relevant by the user before
hitting a non-relevant one. Experimental analyses on three different databases show the effectiveness
of our algorithm compared to state-of-the-art approaches in particular when the images are corrupted
with different types of noise
The widely used feature representation scheme for magnetic resonance (MR) image classification ba... more The widely used feature representation scheme for magnetic resonance (MR) image classification based on low-frequency subband (LFS) coefficients of wavelet transform (WT) is ineffective in presence of common MR imaging (MRI) artifacts (small rotation, low dynamic range etc.). The directional information present in the high-frequency subbands (HFSs) can be used to improve the performance. Moreover, little attention has been paid to the newly developed multiscale geometric analysis (MGA) tools (curvelet, contourlet, and ripplet etc.) in classifying brain MR images. In this paper, we compare various multiresolution analysis (MRA)/MGA transforms, such as traditional WT, curvelet, contourlet and ripplet, for brain MR image classification. Both the LFS and the high-frequency subbands (HFSs) are used to construct image representative feature vector invariant to common MRI artifacts. The investigations include the effect of different decomposition levels and filters on classification performance. By comparing results, we give the best candidate for classifying brain MR images in presence of common artifacts.
This paper presents a segmentation method, integrating judiciously the merits of rough-fuzzy comp... more This paper presents a segmentation method, integrating judiciously the merits of rough-fuzzy computing and multiresolution image analysis technique, for documents having both text and graphics regions. It assumes that the text and non-text regions of a given document are considered to have different textural properties. The M -band wavelet packet is used to extract the scale-space features, which is able to zoom it onto narrow band high frequency components of a signal. A scale-space feature vector is thus derived, taken at different scales for each pixel in an image. Finally, the rough-fuzzy-possibilistic c-means algorithm is used to address the uncertainty problem of document segmentation. The performance of the proposed technique, along with a comparison with related approaches, is demonstrated on a set of real life document images.
This paper addresses a novel approach to the multisensor, multimodal medical image fusion (MIF) p... more This paper addresses a novel approach to the multisensor, multimodal medical image fusion (MIF) problem, employing multiscale geometric analysis of non-subsampled contourlet transform and fuzzy-adaptive reduced pulse-coupled neural network (RPCNN). The linking strengths of the RPCNNs' neurons are adaptively set by modeling them as the fuzzy membership values, representing their significance in the corresponding source image. Use of RPCNN with less complex structure and having less number of parameters, leads to computational efficiencyan important requirement of point-of-care (POC) health care technologies. The proposed scheme is free from the common shortcomings of the state-of-the-art MIF techniques: contrast reduction, loss of image fine details and unwanted image degradations etc. Subjective and objective evaluations show better performance of this new approach compared to existing techniques.
In this article, we have proposed a blind, fragile and Region of Interest (ROI) lossless medical ... more In this article, we have proposed a blind, fragile and Region of Interest (ROI) lossless medical image watermarking (MIW) technique, providing an all-in-one solution tool to various medical data distribution and management issues like security, content authentication, safe archiving, controlled access retrieval and captioning etc. The proposed scheme combines lossless data compression and encryption technique to embed electronic health record (EHR)/DICOM metadata, image hash, indexing keyword, doctor identification code and tamper localization information in the medical images. Extensive experiments (both subjective and objective) were carried out to evaluate performance of the proposed MIW technique. The findings offer suggestive evidence that the proposed MIW scheme is an effective all-in-one solution tool to various issues of medical information management domain. Moreover, given its relative simplicity, the proposed scheme can be applied to the medical images to serve in many medical applications concerned with privacy protection, safety, and management etc.
The H.264/Advanced Video Coding (AVC) is the industry standard in network surveillance owing to i... more The H.264/Advanced Video Coding (AVC) is the industry standard in network surveillance owing to improved video quality, low bandwidth and latency requirements. This paper presents a novel approach for background subtraction in H.264 encoded bitstreams. Temporal statistics of proposed feature vectors, representing macroblock units in each frame, are used to identify potential candidates containing moving objects. From the set of coarsely localized candidate macroblocks, all foreground pixels are detected by comparing them pixel wise with a background model. The basic difference of the current work compared to the related approaches is that, it allows each macroblock to have a different quantization parameter, in view of the requirements in variable as well as constant bitrate applications. Additionally, for pixel wise comparisons, a new color differencing technique of low complexity is proposed which enables us to obtain pixel-resolution segmentation incurring a negligible cost compared to those of classical pixel domain approaches. Results showing striking comparison against those of proven state-of-theart pixel domain algorithms are presented over a diverse set of standardized surveillance sequences. Index Terms-H.264/AVC, background subtraction, video surveillance, compressed domain algorithm.
We present a novel Content Based Image Retrieval (CBIR) scheme for natural color images using Mul... more We present a novel Content Based Image Retrieval (CBIR) scheme for natural color images using Multi-scale Geometric Analysis (MGA) of Type-I Ripplet Transform (RT) and the statistical framework based on Generalized Gaussian Density (GGD) model and Kullback-Leibler Distance (KLD). The system is based on modeling of the marginal distributions (parameters) of RT coefficients using GGD model and computing the similarity using the KLD among the model parameters. The retrieval performance of the proposed system is further improved using Least Square-Support Vector Machine (LS-SVM) classifier. Extensive experiments were carried out to evaluate the effectiveness of the proposed system on two image databases (DBs) consisting 1000 (Simplicity) and 2788 (Oliva) images, respectively. Experimental results and comparisons show that the proposed CBIR system performs efficiently in image retrieval field.
In this article, we have proposed a novel Content Based Image Retrieval (CBIR) system, where each... more In this article, we have proposed a novel Content Based Image Retrieval (CBIR) system, where each image in the database is represented by a compact image signature which is computed using the Nonsubsampled Contourlet Transform (NSCT) and Fuzzy-C-means (FCM) technique. To improve the retrieval accuracy, the proposed system incorporates Least Square Support Vector Machine (LS-SVM) based classifier, Earth Mover's Distance (EMD) and Relevance Feedback Mechanism (RFM). Extensive experiments were carried out to evaluate the effectiveness of the proposed system on SIMPLIcity image database consisting of 1000 images. Experimental results and comparisons show that the proposed CBIR system performs efficiently in image retrieval domain.
We propose an automatic and accurate technique for classifying normal and abnormal magnetic reson... more We propose an automatic and accurate technique for classifying normal and abnormal magnetic resonance (MR) images of human brain. Ripplet transform Type-I (RT), an efficient multiscale geometric analysis (MGA) tool for digital images, is used to represent the salient features of the brain MR images. The dimensionality of the image representative feature vector is reduced by principal component analysis (PCA). A computationally less expensive support vector machine (SVM), called least square-SVM (LS-SVM) is used to classify the brain MR images. Extensive experiments were carried out to evaluate the performance of the proposed system. Two benchmark MR image datasets and a new larger dataset were used in the experiments, consisting 66, 160 and 255 images, respectively. The generalization capability of the proposed technique is enhanced by 5 × 5 cross validation procedure. For all the datasets used in the experiments, the proposed system shows high classification accuracies (on an average > 99%). Experimental results and performance comparisons with state-of-the-art techniques, show that the proposed scheme is efficient in brain MR image classification.
Uploads
Books by MALAY KUMAR KUNDU
Papers by MALAY KUMAR KUNDU
There is indeed a growing availability of images, but unfortunately the traditional metadata-based search
systems are unable to properly exploit their visual information content. In this article we introduce a
novel CBIR scheme that abstracts each image in the database in terms of statistical features computed
using the Multi-scale Geometric Analysis (MGA) of Non-subsampled Contourlet Transform (NSCT). Noise
resilience is one of the main advantages of this feature representation. To improve the retrieval
performance and reduce the semantic gap, our system incorporates a Relevance Feedback (RF) mechanism
that uses a graph-theoretic approach to rank the images in accordance with the user’s feedback.
First, a graph of images is constructed with edges reflecting the similarity of pairs of images with respect
to the proposed feature representation. Then, images are ranked at each feedback round in terms of the
probability that a random walk on this graph reaches an image tagged as relevant by the user before
hitting a non-relevant one. Experimental analyses on three different databases show the effectiveness
of our algorithm compared to state-of-the-art approaches in particular when the images are corrupted
with different types of noise
There is indeed a growing availability of images, but unfortunately the traditional metadata-based search
systems are unable to properly exploit their visual information content. In this article we introduce a
novel CBIR scheme that abstracts each image in the database in terms of statistical features computed
using the Multi-scale Geometric Analysis (MGA) of Non-subsampled Contourlet Transform (NSCT). Noise
resilience is one of the main advantages of this feature representation. To improve the retrieval
performance and reduce the semantic gap, our system incorporates a Relevance Feedback (RF) mechanism
that uses a graph-theoretic approach to rank the images in accordance with the user’s feedback.
First, a graph of images is constructed with edges reflecting the similarity of pairs of images with respect
to the proposed feature representation. Then, images are ranked at each feedback round in terms of the
probability that a random walk on this graph reaches an image tagged as relevant by the user before
hitting a non-relevant one. Experimental analyses on three different databases show the effectiveness
of our algorithm compared to state-of-the-art approaches in particular when the images are corrupted
with different types of noise