U.S.N. Raju

Texture Segmentation Methods Based on Combinatorial of Morphological and Statistical Operations

In this paper we introduce a novel and simple image segmentation schemes that are based on combinations of morphological and statistical operations. Mathematical morphology is very attractive for this purpose because it efficiently deals with geometrical features like as size, shape, contrast or connectivity that can be considered as segmentation oriented features. The present paper derives equations on the basis of dilation, erosion and median or mean which finally results segmentation. The segmentation algorithms are divided into three groups based on number of operations and type of operations, used. Some of the proposed methods of segmentation are useful for edge based segmentation while the other is useful for region based segmentation. The segmentation quality is improved, by dynamically changing the combinatorial coefficients that are used in equations. The present combinatorial method is applied on Brodatz textures and a good segmentation is resulted.

Content Based Image Retrieval on Hadoop Framework

This paper presents an approach for implementing Content Based Image Retrieval (CBIR) on Hadoop MapReduce framework. Although any algorithm can be used for the actual retrieval (provided it can be parallelized), in this paper, we discuss the approach of Local Tetra Patterns (LTrPs). The MapReduce implementation details presented here can be used for most CBIR techniques that involve feature-vector computations and distance measures.

Weighted finite automata based on local patterns for image authentication

This paper presents an efficient image authentication system. The authentication signature is extracted from WFA encoding of the image. For noises that are more textural rather than color-based, we transform the image using a Local-Binary-Pattern filter, which is then converted to automata. We present a technique that incorporates the weights of the WFA, unlike previous works.

Query Object Detection in Big Video Data on Hadoop Framework

In the emerging era of technology filled with multimedia sources video lecturing is becoming increasingly significant. The detection of objects in such systems is crucial for application areas such as identification of person, gender, event and other non-living things. This paper throws some light on adopting the methods to handle such Big video data using Hadoop framework. Hadoop Map-Reduce technology in assistance with the video processing algorithms refined till a certain efficiency helps in detecting the objects in Big video data. The Big video data is initially distributed to various nodes of Hadoop environment where in each node functions in processing of videos. The cull map per distributes videos to all nodes in a cluster. In the first map phase original videos are converted from RGB to gray scale since processing on gray scale videos is three times faster than that of processing on RGB videos. Then in second map phase, the background subtraction method for object detection is carried out. The detected object can be then classified into its category.

Video Copy Detection in Distributed Environment

Video copy detection is a complementary approach to watermarking. As opposed to watermarking, which relies on inserting a distinct pattern into the video stream, video copy detection techniques match content-based signatures to detect copies of video. Existing typical content-based copy detection schemes have relied on image matching. In this we try to implement a technique to detect copying of videos in distributed environment using content based fingerprints.

Weighted Finite Automata Based Image Compression on Hadoop MapReduce Framework

In this paper, we explore the algorithm for WFA inference from images, and its possibilities. We also provide a brief introduction into Hadoop and its application in image processing. Hadoop MapReduce framework can be extended to work with WFA based compression of images, as the data set of images is very huge.

A new hybrid watermarking technique using DCT and DWT based on scaling factor

The Digital Watermarking can be used to hide an image over a cover image and also for copyright protection. In particular, digital image watermarking algorithms based on DCT and DWT are most preferable than others. In this paper a hybrid watermarking technique has been introduced which combines DCT and DWT to produce the watermarked image. The proposed method allows us to get visible as well as the invisible watermarked image by changing the value of the scaling factor. Here in this paper, the perceptibility is measured using statistical parameters such as PSNR and MSE on different values of the scaling factor. Also, the effectiveness of this technique is measured for different types of attacks on the basis of BCR.