Retrieval performance of a content-based image retrieval system is affected by similarity measures used in the development of the system. Similarity measures indicate that how two images are matching to each other. Several similarity... more
Retrieval performance of a content-based image retrieval system is affected by similarity measures used in the development of the system. Similarity measures indicate that how two images are matching to each other. Several similarity measures for retrieval have been developed by various researchers. In this chapter, some commonly used similarity measures are described. After development of a retrieval system, it is necessary to check performance of the system in terms of output generated in response to a query, in comparison to other state-of-the-art systems. This chapter also describes some common measures that are used to evaluate the performance on CBIR systems.
Research Interests:
In this chapter, a content-based image retrieval technique based on the concept of region-based image retrieval has been described. This technique integrates color, texture, and shape features using local binary patterns (LBPs). In this... more
In this chapter, a content-based image retrieval technique based on the concept of region-based image retrieval has been described. This technique integrates color, texture, and shape features using local binary patterns (LBPs). In this technique, the image is divided into a fixed number of blocks and from each block LBP-based color, texture, and shape features are computed. The color and texture features are extracted using LBP histograms of quantized color image and gray-level images, respectively. Shape features are computed using the binary edge map obtained using Sobel edge detector from each block. All three features are combined to make a single completed binary region descriptor (CBRD) represented in the LBP way. To support region-based retrieval, an effective region code-based scheme is employed. In this technique, the spatial relative locations of objects are also considered to increase the retrieval accuracy.
Research Interests:
Online auction, shopping, electronic billing etc. all such types of application involves problems of fraudulent transactions. Online fraud occurrence and its detection is one of the challenging fields for web development and online... more
Online auction, shopping, electronic billing etc. all such types of application involves problems of fraudulent transactions. Online fraud occurrence and its detection is one of the challenging fields for web development and online phantom transaction. As no-secure specification of online frauds is in research database, so the techniques to evaluate and stop them are also in study. We are providing an approach with Hidden Markov Model (HMM) and mobile implicit authentication to find whether the user interacting online is a fraud or not. We propose a model based on these approaches to counter the occurred fraud and prevent the loss of the customer. Our technique is more parameterized than traditional approaches and so,chances of detecting legitimate user as a fraud will reduce.
Research Interests:
Research Interests:
This chapter provides an introduction to information retrieval and image retrieval. Types of image retrieval techniques, i.e., text-based image retrieval and content-based image retrieval techniques are introduced. A brief introduction to... more
This chapter provides an introduction to information retrieval and image retrieval. Types of image retrieval techniques, i.e., text-based image retrieval and content-based image retrieval techniques are introduced. A brief introduction to visual features like color, texture, and shape is provided. Similarity measures used in content-based image retrieval and performance evaluation of content-based image retrieval techniques are also given. Importance of user interaction in retrieval systems is also discussed.
Research Interests:
Uses of digital images have increased multifold in last few years in various important fields such as virtual reality, gaming, social media, magazine, news papers, medical, legal issues, law, academics etc. At the same time, image editing... more
Uses of digital images have increased multifold in last few years in various important fields such as virtual reality, gaming, social media, magazine, news papers, medical, legal issues, law, academics etc. At the same time, image editing and rendering tools have also evolved significantly. With the help of computers and such advanced image rendering tools it is possible to create photorealistic computer graphics images effortlessly. It is very difficult to discriminate such photorealistic computer graphics images from actual photographic images taken from digital cameras by human visual system. If computer generated images are used with malicious intentions it creates negative impact on the society. Therefore, several methods have been proposed in last few years to distinguish computer generated images from photographic images. This paper presents a comprehensive review of the existing methods. A classification of all existing methods is also provided based on the use of feature extraction techniques and classifier used. Accordingly, all the existing methods are grouped into four categories: statistical feature based, acquisition process based, visual feature based, and hybrid feature based methods. This paper also reviews publically available related image datasets and suggests the future directions.
Research Interests:
Abstract Today’s advanced multimedia tools allow us to create photorealistic computer graphic images, effortlessly. There are various fields such as the film industry, virtual reality, video games where computer-generated (CG) images are... more
Abstract Today’s advanced multimedia tools allow us to create photorealistic computer graphic images, effortlessly. There are various fields such as the film industry, virtual reality, video games where computer-generated (CG) images are used widely. CG images can also be misused in many ways. Therefore, there is a need of distinguishing CG images from real photographic (PG) images. This paper proposes a method to distinguish CG images from PG images using a two-stream convolutional neural network (CNN). In the proposed method, the first stream takes the advantage of the knowledge learned by the pre-trained VGG-19 network, and then this knowledge is transferred to learn the distinct features of CG and PG images. Here, we propose a second stream, that preprocesses the images using three high-pass filters which aim to help the network to focus on noise-based distinct features of CG and PG images. Finally, we propose an ensemble model to merge the outcomes of the proposed two streams. Comparative and self-analysis experiments demonstrates that the proposed method outperforms the state-of-the-art methods in terms of classification accuracy. The experimental results also show that the proposed method performs satisfactorily under the additive white Gaussian noise postprocessing operation in the images.
Research Interests:
This paper introduces a novel texture image retrieval technique based on block level processing using Tetrolet and optimized directional local extrema patterns. Texture image categorization is performed for uniform and non-uniform... more
This paper introduces a novel texture image retrieval technique based on block level processing using Tetrolet and optimized directional local extrema patterns. Texture image categorization is performed for uniform and non-uniform distribution of the intensities within the image. Texture features are extracted by using Tetrolet transform and directional local extrema pattern. Image is processed at block level for extracting these features. The main concept of this approach is to analyze the image at block level to get better results in retrieval process. During image search, each block is compared with the corresponding block of another image. Categorization of the images reduces the search space. Proposed approach uses spatial and spectral domain analysis of the image. Performance of proposed retrieval system is tested on the Brodatz and VisTex benchmark databases. Retrieval results show that the proposed technique performs better in terms of average retrieval rate in comparison to other state-of-the-art techniques.
Research Interests:
Image manipulation has become an easy task due to the availability of user-friendly multimedia tools. The images can be manipulated in several ways. Image splicing is one of such image manipulation methods in which two or more images are... more
Image manipulation has become an easy task due to the availability of user-friendly multimedia tools. The images can be manipulated in several ways. Image splicing is one of such image manipulation methods in which two or more images are merged to obtain a single composite image. These manipulated images can be misused to cheat others. This paper proposes a deep learning-based method to detect image splicing in the images. First, the input image is preprocessed using a technique called ‘Noiseprint’ to get the noise residual by suppressing the image content. Second, the popular ResNet-50 network is used as a feature extractor. Finally, the obtained features are classified as spliced or authentic using the SVM classifier. The experiments performed on the CUISDE dataset show that the proposed method outperforms other existing methods. The proposed method achieves an average classification accuracy of 97.24%.
Research Interests:
In digital images, the most common forgery is copy-move image forgery in which some region(s) of an image is replicated within the image. The copy-move forgery detection (CMFD) techniques fall under two categories; keypoint-based and... more
In digital images, the most common forgery is copy-move image forgery in which some region(s) of an image is replicated within the image. The copy-move forgery detection (CMFD) techniques fall under two categories; keypoint-based and block-based. The keypoint-based techniques perform well under rotation and scaling but show very poor performance in the case of smooth images. On the contrary, the block-based techniques perform better in smooth images but are comparatively more time demanding. In this paper, a hybrid technique has been proposed by combining the block-based technique using Fourier-Mellin Transform (FMT) and a keypoint-based technique using Scale Invariant Feature Transform (SIFT). In this technique, the input image to be checked for forgery is first divided into texture and smooth regions. Then the keypoints are extracted from the texture part of the image using the SIFT descriptor, and the FMT is applied on the smooth part of the image. Extracted features are then matched to detect the duplicated regions of the image. The experimental results illustrate that the proposed technique performs better in comparison to other state-of-the-art CMFD techniques under various geometric transformations and post-processing operations in reasonable time.
Research Interests:
Abstract Copy-move forgery is a common type of forgery in digital images. In copy-move forgery, one part of the image is replicated within the same image, generally at different location. For revival of trustworthiness of images, there is... more
Abstract Copy-move forgery is a common type of forgery in digital images. In copy-move forgery, one part of the image is replicated within the same image, generally at different location. For revival of trustworthiness of images, there is a need to develop an efficient and robust technique to detect such forgeries. This paper proposes a new copy-move image forgery detection technique based on Tetrolet transform. In this technique, initially the input image is divided into overlapping blocks, then four low-pass coefficients and twelve high-pass coefficients are extracted from each block by applying Tetrolet transform. Feature vectors are then sorted lexicographically, and similar blocks are identified by matching the extracted Tetrolet features. Experimental results show that the proposed technique can detect and locate the duplicated regions in the images very accurately, even when the copied regions have undergone some post-processing operations blurring, color reduction, adjustment of brightness and contrast, rotation, scaling, JPEG compression. In addition, it is also observed that the proposed technique is able to detect very small duplicated regions and multiple forgery cases, even when image is smooth.
Research Interests:
This paper presents a novel technique for content based image retrieval (CBIR) that selects and assigns weights to the regions of the image on the basis of their contribution to image contents, using a new region-weight assignment scheme.... more
This paper presents a novel technique for content based image retrieval (CBIR) that selects and assigns weights to the regions of the image on the basis of their contribution to image contents, using a new region-weight assignment scheme. Assigning the weight to each region ignores the irrelevant regions of the image during retrieval and thus maximizes the retrieval accuracy. The proposed approach performs the feature extraction at both region-level and image-level. Texture and edge features are extracted at region-level whereas shape feature is extracted at image-level. At region-level, the image is divided into non-overlapping regions and texture and edge features are calculated for each region separately. Curvelet transform is used for extracting the texture feature using the curve continuity as well as line continuity in the feature extraction process. Moment invariant is used for extracting the shape features. Integrated Region Matching (IRM) technique is used for retrieving the relevant images. The proposed approach does the best use of the features by balancing the regions and features in the similarity matching of the regions. The performance of the proposed technique is tested on COREL and CIFAR databases. Experimental results show the effectiveness of proposed region weight assignment scheme over the feature weight assignment scheme in image retrieval in comparison to other state-of-the-art techniques.
Research Interests:
This paper proposes a new approach for content based image retrieval based on feed-forward architecture and Tetrolet transforms. The proposed method addresses the problems of accuracy and retrieval time of the retrieval system. The... more
This paper proposes a new approach for content based image retrieval based on feed-forward architecture and Tetrolet transforms. The proposed method addresses the problems of accuracy and retrieval time of the retrieval system. The proposed retrieval system works in two phases: feature extraction and retrieval. The feature extraction phase extracts the texture, edge and color features in a sequence. The texture features are extracted using Tetrolet transform. This transform provides better texture analysis by considering the local geometry of the image. Edge orientation histogram is used for retrieving the edge feature while color histogram is used for extracting the color features. Further retrieval phase retrieves the images in the feed-forward manner. At each stage, the number of images for next stage is reduced by filtering out irrelevant images. The Euclidean distance is used to measure the distance between the query and database images at each stage. The experimental results on COREL- 1 K and CIFAR - 10 benchmark databases show that the proposed system performs better in terms of the accuracy and retrieval time in comparison to the state-of-the-art methods.
Research Interests:
In recent years, a rapid increase in the size of digital image databases has been observed. Everyday gigabytes of images are generated. Consequently, the search for the relevant information from image and video databases has become more... more
In recent years, a rapid increase in the size of digital image databases has been observed. Everyday gigabytes of images are generated. Consequently, the search for the relevant information from image and video databases has become more challenging. To get accurate retrieval results is still an unsolved problem and an active research area. Content-based image retrieval (CBIR) is a process in which for a given query image, similar images are retrieved from a large image database based on their content similarity. A number of techniques have been suggested by researchers for content-based image retrieval. In this chapter, a review of some state-of-the-art retrieval techniques is provided.
Research Interests:
Research Interests:
Abstract Dynamic texture has been found as a powerful cue for modeling natural scenes such as fire, waves and smoke, etc. It combines appearance with motion to characterize the moving scene that exhibits certain spatially repetitive and... more
Abstract Dynamic texture has been found as a powerful cue for modeling natural scenes such as fire, waves and smoke, etc. It combines appearance with motion to characterize the moving scene that exhibits certain spatially repetitive and time-varying visual patterns. This paper proposes a new method of recognizing dynamic texture using the well-known texture descriptor, local binary pattern. The new variant differentiates different structural patterns more efficiently using the additional information from the local patch. This pattern information is further combined with shape information to improve the discriminative power of texture descriptor. The proposed method is extended to multiscale using classifier fusion scheme to capture the spatio-temporal content of a moving scene at multiple scales, thus improves representation capability of the new descriptor. Proposed descriptor is tested on three dynamic texture databases: UCLA, Dyntex and Dyntex++. Results demonstrate that the proposed feature descriptor outperforms various state-of-the-art approaches on all representative databases in terms of classification accuracy.
Research Interests:
This paper presents a scheme for the representation and recognition of the dynamic texture in the noisy environment. Dynamic texture is the sequence of images of a moving scene that shows some form of temporal regularity. Though, the... more
This paper presents a scheme for the representation and recognition of the dynamic texture in the noisy environment. Dynamic texture is the sequence of images of a moving scene that shows some form of temporal regularity. Though, the dynamic texture is the spatiotemporal extension of the conventional texture, its analysis requires additional attention, since the motion causes continuous changes in the geometry of these textures. Hence, to recognize the noisy dynamic texture, the noise should be estimated not only at the spatial level; it must also be estimated at the temporal level. To this end, an auto tuned noise resistance descriptor, based on the Local Binary Pattern approach, is proposed for the modeling and classification of the dynamic texture. Our approach based on the fact that, uniform local binary patterns are the fundamental units of image texture and occur more frequently in an image in comparison to non-uniform patterns. Noise affects these uniform local binary patterns and more likely fall into non-uniform codes. To counter this, we have extended conventional local binary pattern descriptor from a 2-value code to a 3-value code to include an additional state (called indecisive state) to represent the noise affected pixels. However, the estimation of the indecisive state is of the primary concern of this letter due to the inherently varying nature of the dynamic texture. The proposed technique devised a new scheme to estimate the noisy pixels in a dynamic texture. Eventually, the indecisive state is corrected back to non-noisy natural states using a mapping function so as to form a uniform LBP code. Experimental results on the UCLA and Dyntex++ databases demonstrate high classification efficiency of the proposed approach in the noisy environment.
Research Interests:
This paper proposes an approach of object based image retrieval to retrieve the images based on location independent region of interest (ROI). In this approach, instead of extracting the features of the whole query image, features of the... more
This paper proposes an approach of object based image retrieval to retrieve the images based on location independent region of interest (ROI). In this approach, instead of extracting the features of the whole query image, features of the objects of interest are extracted. For this, some morphological operations are performed on the image. First, background subtraction is performed to reduce the effect of background intensities, then segmentation is performed and the regions are extracted. To minimize the number of comparisons in image retrieval process, the image is categorized into texture and non texture regions. This reduces the retrieval time by comparing the ROI on the basis of its category. During the feature extraction process, a flag is set to indicate the category of the image i.e. texture image or non-texture (natural) image. Feature vector of an image is stored along with respective objects within the image. Tetrolet transform is used to retrieve the texture features for the texture regions while moment invariants and edge features are used for non-texture regions. The performance and efficiency of the proposed system is tested on COREL and CIFAR databases. Experimental results show that the retrieval performance of the proposed algorithm is better in comparison to other state-of-the-art methods.
Research Interests:
Research Interests:
Image denoising has always been one of the standard problems in image processing and computer vision. It is always recommendable for a denoising method to preserve important image features, such as edges, corners, etc., during its... more
Image denoising has always been one of the standard problems in image processing and computer vision. It is always recommendable for a denoising method to preserve important image features, such as edges, corners, etc., during its execution. Image denoising methods based on wavelet transforms have been shown their excellence in providing an efficient edge-preserving image denoising, because they provide a suitable basis for separating noisy signal from the image signal. This paper presents a novel edge-preserving image denoising technique based on wavelet transforms. The wavelet domain representation of the noisy image is obtained through its multi-level decomposition into wavelet coefficients by applying a discrete wavelet transform. A patch-based weighted-SVD filtering technique is used to effectively reduce noise while preserving important features of the original image. Experimental results, compared to other approaches, demonstrate that the proposed method achieves very impressive gain in denoising performance.
Research Interests:
This paper proposes a novel technique for texture image retrieval based on tetrolet transforms. Tetrolets provide fine texture information due to its different way of analysis. Tetrominoes are applied at each decomposition level of an... more
This paper proposes a novel technique for texture image retrieval based on tetrolet transforms. Tetrolets provide fine texture information due to its different way of analysis. Tetrominoes are applied at each decomposition level of an image and best combination of tetrominoes is selected, which better shows the geometry of an image at each level. All three high pass components of the decomposed image at each level are used as input values for feature extraction. A feature vector is created by taking standard deviation in combination with energy at each subband. Retrieval performance in terms of accuracy is tested on group of texture images taken from benchmark databases: Brodatz and VisTex. Experimental results indicate that the proposed method achieves 78.80% retrieval accuracy on group of texture images D1 (taken from Brodatz), 84.41% on group D2 (taken from VisTex) and 77.41% on rotated texture image group D3 (rotated images from Brodatz).
Research Interests:
A local region in an image can be defined using centre pixel and its differences with neighboring pixels. In order to characterize different texture structure in a discriminating manner, this paper proposes twoframeworks of Noise... more
A local region in an image can be defined using centre pixel and its differences with neighboring pixels. In order to characterize different texture structure in a discriminating manner, this paper proposes twoframeworks of Noise Invariant Structure Pattern (NISP) which utilizes both the centre pixel and local and global information of an image. To replace the centre pixel, a threshold computed from adding centre pixel and intensity averages is used in the LBP code computation. For adding the magnitude information, binary patterns generated by taking thresholds involving centre pixel and local and global average contrast are adopted. Also for adding the information of individual neighborhood of a given pixel, the binary patterns generated from global thresholding of local averages are used. Based on the use of local and global information, this paper suggests two noise invariant models that are CNLP and CNGP (i.e. Completed Noise-invariant Local-structure Pattern and Global–structure Pattern). The proposed NISPs are also insensitive to noise as the centre pixel is not directly used as threshold. The proposed texture descriptors are tested on some of the representative texture databases like Outex, Curet, UIUC, Brodatz and XU –HR. The experimental results have shown that the proposed schemes can achieve higher classification and retrieval rates while being more robust to noise.
Research Interests:
In modern scenario we need to have mechanisms which can provide better interaction with physical world by an efficient and more effective communication and computation approach for multiple heterogeneous sensor networks. Previous work... more
In modern scenario we need to have mechanisms which can provide better interaction with physical world by an efficient and more effective communication and computation approach for multiple heterogeneous sensor networks. Previous work provides efficient communication approach between sensor nodes and a query centric approach for multiple collaborative heterogeneous sensor networks. Even there is energy issues involved in wireless sensor network operation. In this paper we have proposed Query centric Cyber Physical System (QCPS)model to implement query centric user request using Cyber Physical System (CPS). CPS takes both communication and computation in parallel to provide better interaction with physical world. This feature of CPS reduces system cost and makes it more energy efficient. This paper provides an efficient query processing approach for multiple heterogeneous sensor networks using cyber physical system.This approach results in reduction of communication and computation c...
Research Interests:
We introduce linear network coding on parallel architecture for multi-source finite acyclic network. In this problem, different messages in diverse time periods are broadcast and every nonsource node in the network decodes and encodes the... more
We introduce linear network coding on parallel architecture for multi-source finite acyclic network. In this problem, different messages in diverse time periods are broadcast and every nonsource node in the network decodes and encodes the message based on further communication.We wish to minimize the communication steps and time complexity involved in transfer of data from node-to-node during parallel communication.We have used Multi-Mesh of Trees (MMT) topology for implementing network coding. To envisage our result, we use all-to-all broadcast as communication algorithm.
Research Interests:
In this paper an integrated approach for image retrieval has been proposed that uses the concept of local binary pattern. The image is divided into a fixed number of blocks and from each block, LBP based color, texture and shape features... more
In this paper an integrated approach for image retrieval has been proposed that uses the concept of local binary pattern. The image is divided into a fixed number of blocks and from each block, LBP based color, texture and shape features are computed. LBP histogram is used for the extraction of color and texture features. Region code based scheme is used to support region based retrieval. Center pixel and its neighbors are used to improve the discrimination power of Local Binary Patterns. Shape feature computed using the binary edge map obtained using Sobel edge detector is combined with color and texture features to make a single completed binary region descriptor. To support region based retrieval, a more effective region code based scheme is employed. The approach is tested on different benchmark databases like COREL, CIFAR-10 and MPEG-7 CCD database. The experimental results have verified that the proposed scheme has impressive retrieval performance in comparison to state-of-the-art techniques.
Research Interests:
Abstract Image restoration is the operation of taking a degraded image and estimating the clean image. Degradation in an image occurs primarily due to blur and noise. Restoration attempts to reconstruct or recover an image that has been... more
Abstract Image restoration is the operation of taking a degraded image and estimating the clean image. Degradation in an image occurs primarily due to blur and noise. Restoration attempts to reconstruct or recover an image that has been degraded by using apriori knowledge of the degradation phenomenon. Thus restoration techniques are oriented toward modeling the degradation and applying the inverse process in order to recover the original image. This paper starts with basic model of image degradation/restoration process and proceeds towards enlisting commonly occurring blurs and noises. Thereafter, it reviews some spatial and frequency domain filters for restoration of noisy images.
ABSTRACT Denoising has always been one of the standard problems in image processing. It is always desirable to preserve important features, such as edges, corners and other sharp structures, during denoising. Wavelet transforms have been... more
ABSTRACT Denoising has always been one of the standard problems in image processing. It is always desirable to preserve important features, such as edges, corners and other sharp structures, during denoising. Wavelet transforms have been widely used in edge-preserving image denoising since these provide a suitable basis for suppressing noisy signals from the image. This paper presents a novel edge-preserving image denoising technique based on tetrolet transform (a Haar-type wavelet transform) and a locally adaptive thresholding method. The noisy image is decomposed into tetrolet (wavelets) coefficients through a tetrolet transform. A locally adaptive thresholding method exploiting interscale statistical dependency and based on computation of noise level is used to threshold the tetrolet coefficients to effectively reduce noise while preserving relevant features of the original image. Experimental results, compared to other approaches, demonstrate that the proposed method is suitable especially for the natural images contaminated by Gaussian noise.
Research Interests:
Research Interests:
ABSTRACT Dynamic textures are the sequences of images of moving scenes having some stationary properties in time; these include sea-waves, smoke, foliage, whirlwind etc. In recent years, dynamic texture description and recognition has... more
ABSTRACT Dynamic textures are the sequences of images of moving scenes having some stationary properties in time; these include sea-waves, smoke, foliage, whirlwind etc. In recent years, dynamic texture description and recognition has attracted growing attention. This paper presents a more effective completed modeling of volume local binary pattern (VLBP) to recognize dynamic textures. Due to the dependency on local binary pattern (LBP), traditional VLBP also suffers with noise sensitivity and may give the same LBP code to different structural patterns; thus limiting the discriminating power of a texture descriptor. To represent temporal textures we have used VLBP. Local region of a volume is represented using its center frame and local sign magnitude difference with the circularly symmetric neighborhood. We have proposed a new contrast operator to complement the sign information of the temporal texture. To add additional discriminative information, volume center pixel information is also fused with the sign magnitude difference of texture. By combining these features into hybrid distributions we get higher classification accuracy for rotation invariant texture classification. Experimental results on UCLA and Dyntex databases show that the proposed approach provides better performance in comparison to the existing approaches.
Research Interests:
Research Interests:
Research Interests:
Content based image retrieval involves extraction of global and region features of images for improving their retrieval performance in large image databases. Region based feature have shown to be more effective than global features as... more
Content based image retrieval involves extraction of global and region features of images for improving their retrieval performance in large image databases. Region based feature have shown to be more effective than global features as they are capable of reflecting users specific interest with greater accuracy. However success of region based methods largely depends on the segmentation technique used to automatically specify the region of interest (ROI) in the query. Apart from this user can also specify ROI’s in an image. The ROI image retrieval involves the task of formulation of region based query, feature extraction, indexing and retrieval of images containing similar region as specified in the query. In this paper state-of-the-art techniques for ROI image retrieval are discussed. Comparative study of each of these techniques together with pros and cons of each technique are listed. The paper is concluded with our views on challenges faced by researchers and further scope of research in the area. The major goal of the paper is to provide a comprehensive reference source for the researchers involved in image retrieval based on ROI.
Research Interests:
In this paper a region based image retrieval scheme has been proposed based on integration of color, texture and shape features using local binary patterns (LBP). The color and texture features are extracted using LBP histograms of... more
In this paper a region based image retrieval scheme has been proposed based on integration of color, texture and shape features using local binary patterns (LBP). The color and texture features are extracted using LBP histograms of quantized color image and gray level images respectively. For improving the discrimination power of LBP, threshold computed using both centre pixel and its neighbors is used. Finally, shape features are computed using the binary edge map obtained using Sobel edge detector from each block. All three features are combined to make a single completed binary region descriptor (CBRD) represented in the LBP way. To support region based retrieval a more effective region code based scheme is employed. The spatial relative locations of objects are also considered to increase the retrieval accuracy.
Traditional image retrieval systems match the input image by searching the whole database repeatedly for various image features. Intermediate results produced for these features are merged using data fusion techniques to produce one... more
Traditional image retrieval systems match the input image by searching the whole database repeatedly for various image features. Intermediate results produced for these features are merged using data fusion techniques to produce one common output. In this paper, a new image retrieval technique is presented, which retrieves similar images in three stages. A fixed number of images is first retrieved based on their color feature similarity. The relevance of the retrieved images is further improved by matching their texture and shape features respectively. This eliminates the need of fusion and normalization techniques, which are commonly used to calculate final similarity scores. This reduces the computation time and increases the overall accuracy of the system. Moreover, in this technique, global and region features are combined to obtain better retrieval accuracy. Experimental results on two databases (COREL and CIFAR) have shown that the proposed technique produces better results while consuming less computation time for large image databases.
Research Interests:
Multistage interconnection networks (MINs) consist of more than one stage of small interconnection elements called switching elements and links interconnecting them. A MIN connects N inputs to N outputs and is referred as an N × N MIN,... more
Multistage interconnection networks (MINs) consist of more than one stage of small interconnection elements called switching elements and links interconnecting them. A MIN connects N inputs to N outputs and is referred as an N × N MIN, having size N. An Optical MIN (OMIN) is an important class of Interconnection networks. The problem of crosstalk is caused by coupling two signals within switching elements. A number of techniques like Optical window, Heuristic, Genetic, and Zero have been proposed earlier in this regard. In this paper, we have proposed an Address Selection Algorithm (ASA) and we have applied to existing Omega network, having shuffle-exchange connection pattern. The aim of this algorithm is to minimize the number of switch conflicts in the network and to provide conflict free routes.
Research Interests:
Research Interests:
With the increase in the complexity of task, complex architectures such as grid systems and cluster computing are employed to process huge amount of data. The major problem issues of such task processing systems include heterogeneity,... more
With the increase in the complexity of task, complex architectures such as grid systems and cluster computing are employed to process huge amount of data. The major problem issues of such task processing systems include heterogeneity, load balancing, synchronization ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
ABSTRACT Reducing noise has always been one of the standard problems of the image analysis and processing community. Often though, at the same time as reducing the noise in a signal, it is important to preserve the edges. Edges are of... more
ABSTRACT Reducing noise has always been one of the standard problems of the image analysis and processing community. Often though, at the same time as reducing the noise in a signal, it is important to preserve the edges. Edges are of critical importance to the visual appearance of images. So, it is desirable to preserve important features, such as edges, corners and other sharp structures, during the denoising process. This paper presents a review of some significant work in the area of image denoising. It provides a brief general classification of image denoising methods. The main aim of this survey is to provide evolution of research in the direction of edge-preserving image denoising. It characterizes some of the well known edge-preserving denoising methods, elaborating each of them, and discusses the advantages and drawbacks of each. Basic ideas and improvement of the denoising methods are also comprehensively summarized and analyzed in depth. Often, researchers face difficulty in selecting an appropriate denoising method that is specific to their purpose. We have classified and systemized these denoising methods. The key goal of this paper is to provide researchers with background on a progress of denoising methods so as to make it easier for researchers to choose the method best suited to their aims.
Research Interests:
Effectiveness of local binary pattern (LBP) features is well proven in the field of texture image classification and retrieval. This paper presents a more effective completed modeling of the LBP. The traditional LBP has a shortcoming that... more
Effectiveness of local binary pattern (LBP) features is well proven in the field of texture image classification and retrieval. This paper presents a more effective completed modeling of the LBP. The traditional LBP has a shortcoming that sometimes it may represent different structural patterns with same LBP code. In addition, LBP also lacks global information and is sensitive to noise. In this paper, the binary patterns generated using threshold as a summation of center pixel value and average local differences are proposed. The proposed local structure patterns (LSP) can more accurately classify different textural structures as they utilize both local and global information. The LSP can be combined with a simple LBP and center pixel pattern to give a completed local structure pattern (CLSP) to achieve higher classification accuracy. In order to make CLSP insensitive to noise, a robust local structure pattern (RLSP) is also proposed. The proposed scheme is tested over three representative texture databases viz. Outex, Curet, and UIUC. The experimental results indicate that the proposed method can achieve higher classification accuracy while being more robust to noise.
Research Interests:
Research Interests:
In this study, a novel technique for image retrieval based on selective regions matching using region codes is presented. All images in the database are uniformly divided into multiple regions and each region is assigned a 4-bit region... more
In this study, a novel technique for image retrieval based on selective regions matching using region codes is presented. All images in the database are uniformly divided into multiple regions and each region is assigned a 4-bit region code based upon its location relative to the central region. Dominant color and Local Binary Pattern (LBP) based texture features are extracted from these regions. Feature vectors together with their region codes are stored and indexed in the database. During retrieval, feature vectors of regions having region codes similar to the query image region are used for comparison. To reflect the user's intent in query formulation in a better way, an effective technique for Region of Interest (ROI) overlapping block selection is also proposed. Region codes are further used to find relative locations of multiple ROIs in query and target images. The performance of the proposed approach is tested on the MPEG-7 CCD database and Corel image database. Experimental results show that the proposed approach increases the accuracy and reduces image retrieval time.