Debasree Sarkar

A novel technique for image steganography based on Block-DCT and Huffman Encoding

Image steganography is the art of hiding information into a cover image. This paper presents a novel technique for Image steganography based on Block-DCT, where DCT is used to transform original image (cover image) blocks from spatial domain to frequency domain. Firstly a gray level image of size M x N is divided into no joint 8 x 8 blocks and a two dimensional Discrete Cosine Transform (2-d DCT) is performed on each of the P = MN / 64 blocks. Then Huffman encoding is also performed on the secret messages/images before embedding and each bit of Huffman code of secret message/image is embedded in the frequency domain by altering the least significant bit of each of the DCT coefficients of cover image blocks. The experimental results show that the algorithm has a high capacity and a good invisibility. Moreover PSNR of cover image with stego-image shows the better results in comparison with other existing steganography approaches. Furthermore, satisfactory security is maintained since the secret message/image cannot be extracted without knowing decoding rules and Huffman table.

Mining Frequent Itemsets Using Genetic Algorithm

In general frequent itemsets are generated from large data sets by applying association rule mining algorithms like Apriori, Partition, Pincer-Search, Incremental, Border algorithm etc., which take too much computer time to compute all the frequent itemsets. By using Genetic Algorithm (GA) we can improve the scenario. The major advantage of using GA in the discovery of frequent itemsets is that they perform global search and its time complexity is less compared to other algorithms as the genetic algorithm is based on the greedy approach. The main aim of this paper is to find all the frequent itemsets from given data sets using genetic algorithm.

Image encryption using affine transform and XOR operation

Image encryption is a suitable method to protect image data. Image and text data has their unique features. The available encryption algorithms are good for text data. They may not be suitable for multimedia data. In fact the pixels of natural images are highly correlated to their neighboring pixels. Due to this strong correlation any pixel can be practically predicted from the values of its neighbors. In this article, we propose a new location transformation based encryption technique. We redistribute the pixel values to different location using affine transform technique with four 8-bit keys. The transformed image then divided into 2 pixels × 2 pixels blocks and each block is encrypted using XOR operation by four 8-bit keys. The total key size used in our algorithm is 64 bit which proves to be strong enough. The experimental results proved that after the affine transform the correlation between pixel values was significantly decreased.

A dual tuned Complementary Structure Frequency Selective surface for WLAN applications

A single substrate layer double metallic plane frequency selective surface (FSS) is designed with miniaturized electrical elements and two tunable pass-bands enabling Wireless Local Area Network applications. Each FSS periodic cell consists of a metallic square loop on the top metal layer and its complementary etched on the bottom metal layer separated by a thin dielectric layer. This designed FSS provides two tunable transmission bands. For predicting the frequency selective characteristics of the designed FSS an equivalent electric circuit model is also presented. To satisfy the needed requirements for Wireless Local Area Network (WLAN) applications the designed structure is optimized. Due to symmetrical nature of the design, the FSS is insensitive to variations of RF incidence angle for 90° (degree) rotations of the FSS. Simulated theoretical investigations are done by Ansoft Designer® software. Experimental investigation is performed using standard microwave test bench.

A Weighted Location Based LSB Image Steganography Technique

Steganography is the art of hiding the presence of communication by embedding secret messages into innocent, innocuous looking cover documents, such as digital images, videos, sound files. We present here a novel steganographic method based on affine cipher encryption algorithm and the least significant bit (LSB) substitution in order to provide a strong security and imperceptible visual quality to secret message. We encrypt the 8 bit secret image by changing pixel values using affine cipher. After that each 8 bit pixel of encrypted secret image is divided into 4 groups of 2 bit each. Each part which have a decimal value between 0 to 3 determines the location in each pixel of cover image where to embed the message. We do not store the actual secret message instead we encode the secret message into cover image using the value of each group of secret message. Since, we have two layers of encoding: one using private keys of affine cipher and other for steganography, our methods proves to be more secure than others. Our experimental results also proves that the proposed method has got an acceptable image quality as supported by PSNR values.

Weather Data Mining using Artificial Neural Network

Weather Data Mining is a form of Data mining concerned with finding hidden patterns inside largely available meteorological data, so that the information retrieved can be transformed into usable knowledge. A variety of data mining tools and techniques are available in the industry, but they have been used in a very limited way for meteorological data. In this paper, a neural network-based algorithm for predicting the atmosphere for a future time and a given location is presented. We have used Back Propagation Neural (BPN) Network for initial modelling. The results obtained by BPN model are fed to a Hopfield Network. The performance of our proposed ANN-based method (BPN and Hopfield Network based combined approach) tested on 3 years weather data set comprising 15000 records containing attributes like temperature, humidity and wind speed. The prediction error is found to be very less and the learning converges very sharply. The main focus of this paper is based on predictive data mining by which we can extract interesting (non-trivial, implicit, previously unknown and potentially useful) patterns or knowledge from huge amount of meteorological data.

A Tutorial on Different Classification Techniques for Remotely Sensed Imagery Datasets

Classification techniques are used on large databases to develop models describing different data classes. Such analysis can provide deep insight for better understanding of different large-scale databases. Studies related to knowledge acquisition and effective knowledge development are also very popular in the remote sensing field with satellite imagery datasets. In any remote sensing research, the decision-making process mainly depends on the effectiveness of the classification process. Efficient classification techniques were developed and applied to the Statlog (Landsat Satellite) database at the University of California, Irvine Machine Learning Repository to identify six land type classes. We used three different classification algorithms on the large satellite imagery: multilayer perceptron backpropagation neural network (MLP BPNN), support vector machine (SVM), and k-nearest neighbor (k-NN). This research study aimed to evaluate the performance of these classification algorithms in the prediction of the classified lands from this large set of satellite imagery. We used different performance measures, such as classification accuracy, root-mean-square error, kappa statistic, true positive rate, false positive rate, precision, recall, and F-measure to evaluate the performance of each classifier. Among the three classification techniques applied, MLP BPNN turned out to be the best; next was k-NN, followed by SVM.

A new contrast enhancement method of retinal images in Diabetic Screening System

A new retinal image contrast enhancement method for Diabetic Screening System is presented here. The proposed method evaluated by comparing the retinal image quality with different contrast enhancement methods which are applied in numerous papers. The proposed method produces better image quality and also preserves the mean brightness of the input images which is very important for medical image analysis. Publicly available datasets like DRIVE, STARE, DIARETDB0, and DIARETDB1 are used for testing purpose. The low contrast noisy retinal images are collected from the Hospital and performance of the proposed contrast enhancement is also measured. The SSIM (Structure Similarity Index Measurement) and AAMBE (Average Absolute Mean Brightness Error) are the important performance measurement parameters applied here to examine the image quality and brightness preserving ability respectively. The average SSIM is reported as 0.82(Std. Dev=0.011) and AAMBE is 0.023(Std. Dev=0.030). The success rate on low contrast noisy retinal image analysis is showing the importance of the proposed method.

Secret Image Sharing Scheme Based on a Boolean Operation

Abstract Traditionally extensive researches have been done on secret image sharing which support the fault tolerance property. But their reconstruction complexity is high. Some research papers on secret image sharing are also available with smaller reconstruction complexity, due to the use of a Boolean operation. But these research works lack the fault tolerance property which is the heart of secret sharing. This paper deals with a general (k, n) secret image sharing scheme for gray scale images with both low reconstruction complexity and preservation of the fault tolerance property. Moreover, the proposed sharing generation technique can also be applied on colour images.

Semi Random Position Based Steganography for Resisting Statistical Steganalysis

Steganography is the branch of information hiding for secret communication. The simplest and widely used steganography is the LSB based approach due to its visual quality with high embedding capacity. However, LSB based steganography techniques are not secure against statistical steganalysis mainly χ2 attack and Regular Singular (RS) attack. These two staganalysis can easily estimate the hidden message length. This work propose a LSB based steganography technique where first a location is obtained randomly based on the bit pattern (except LSB) of a cover pixel using linear probing and embed a secret bit into LSB. This technique makes the stego-image completely secure against both χ2 attack and RS attack.