Identification and segmentation of mass are critical for medical image processing. In this paper a combined approach for image segmentation based on watershed transform and k-means clustering is proposed. A preprocessing step is applied to get an initial region of interest which is enhanced using adaptive histogram equalization. Watershed transform is applied to obtain an initial segmentation of the mammograms. Statistical texture features are also computed for the identified regions. K-Means clustering is then applied to produce foreground markers. These markers are given as input to a second phase of marker controlled watershed segmentation. With this, the unwanted regions are greatly reduced giving the suspicious mass region. The proposed approach is validated on a set of 50 mammograms from DDSM database. These mammograms are selected randomly from malign mass classified images. The identified regions are compared with the ground truth values marked in the database. Results show that the algorithm is more effective for mammogram image segmentation as compared to direct application of watershed segmentation approach.

Leukemia, a subdivision of cancer, develops in human blood and the bone marrow. The reason behind is the expeditious and sudden formation and accumulation of WBCs in blood. Identification & diagnosis of these types of abnormalities by humans is difficult and may lead to misidentification. Therefore an automatic system for the identification and classification would be of great help. This paper aims at proposing a technique for correct and quick classification of leukemia images and categorizing them into their respective types. For this, different features are extracted from the input images and then based on these features a data set for the input images is created. This data set is then utilized as input data to a neural network for training purposes. This neural network is designed and created to categorize the images according to their corresponding leukemia type.

A great deal of computer vision research and study is dedicated to the systems designed to detect and analyze computer printed documents and human written text. Optical Character Recognition (OCR) refers to the process of converting images of hand-written, typewritten, or printed text into a format understood by machines for the purpose of editing, indexing/searching, and a reduction in storage size. In this paper we have combined the functionality of Optical Character Recognition and have focused on its applications like Image Sudoku Solver, Car License Plate Detection and Recognition, Handwritten and Computer Printed Documents Recognition. This paper develops a user friendly application for performing image to text conversion. The developed system is organized as a set of modules, each dedicated to a specific application. Car License Plate Detection and Recognition system extracts out the License plate accurately and produce an effective recognition of the characters in the License Plate. With the proposed methodology, we have been able to achieve results with 96% accuracy for the tested images. Image Sudoku Solver is intended to work with Sudoku Puzzle images by extracting out the numbers, boundaries from them and then solving the puzzle. In this we are able to extract, recognize and solve around 98% of Sudoku Puzzles being tested for the purpose. Online Handwriting Recognition accomplishes the real time recognition of user's Handwriting from a Mouse or a Laptop Touchpad.

During the unfolding measures that are taken for the purpose of leukemia detection, segmentation of blood cells is a vital step.In this paper two approaches of such segmentation technique is proposed.While one uses K-means clustering, other uses color image based segmentation method.Both the processes segment the image into two regions, blasts & backgrounds.These blasts are our area of interest.The performance measure is based on the comparison of the two proposed techniques tends to find the more suitable approach for correct leukemia image segmentation.The results show that the segmentation based on K-means clustering gives better results preserving important information and removing background noise.

In human-computer interaction facial expression is the characteristic and proficient method for correspondence, and has been acknowledged as essential input of such interface.In this paper, we present an enhancement in facial expression recognition for image sequence.The most important step is to extract essential features from face to efficiently determine facial expression.Experimentation shows that LBP method performs well while extracting facial features.We further found that Boosted-LBP extracts most distinct features and the best recognition is calculated by SVM classifier.,

Image segmentation commonly known as partitioning of an image is one of the intrinsic parts of any image processing technique. In this image processing step, the digital image of choice is segregated into sets of pixels on the basis of some predefined and preselected measures or standards. There have been presented many algorithms for segmenting a digital image. This paper presents a general review of algorithms that have been presented for the purpose of image segmentation.

This paper presents a texture based approach for distinguishing mass from normal breast tissue in a mammogram. Identification of high probability area as mass is done on the basis of statistical features obtained from Gray-Level-Co-occurrence Matrix (GLCM) of mammogram image. The input mammogram is first pre-processed to remove the labeling artifacts and enhanced using adaptive histogram equalization. Unwanted details from the mammogram are excluded on the basis of block processing and histogram based features are extracted. Features based on GLCM are computed and analyzed to distinguish a suspicious mass from a non-mass region. Obtained results are promising in terms of correct classification. Contrast and energy measure from GLCM and mean, standard deviation and entropy helps to appropriately differentiate malign mass and normal tissue area.

This paper presents an analysis and identification of image pre-processing techniques suitable for mammogram images. Pre-processing is one of the preliminary stages used for mammogram image enhancement to aid an early identification of suspicious lesions and micro calcifications. Nine image preprocessing techniques are considered here for mammogram images. These techniques are implemented using Matlab. The results obtained are compared on the basis of Peak Signal to Noise Ratio (PSNR) for a set of 30 mammogram images. A high value of PSNR indicates better suitability of the pre-processing technique for further image processing. Anisotropic diffusion and Median filtering are suitable for noise removal. Power-law transformation, morphological processing, and un-sharp masking are giving better enhancement results in terms of achieved PSNR values as compared to other pre-processing techniques analyzed here. These identified pre-processing techniques, chosen carefully may give better results for further identification of masses, calcification, architectural distortion, bilateral asymmetry and aid in early detection of breast cancer.

In the perspective of growing demand for research accountability, every university is making it an objective to achieve optimum performance in terms of research and development. Successful completion of research depends on the kind of supervision received by students from their supervisors. As students' satisfaction directly affects the quality of research at the university level, there is an urgent need of an explicit system that evaluates students' satisfaction with their supervisors. This paper aims to examine the experiences and challenges of post graduate research students with their supervisors. For this, data were collected through face-to-face interviews and a questionnaire aimed at recording response from 100 post graduate students of a private university in north India. From the study we identified relevant parameters like supervisor's expertise, active listening, availability and self seeking behavior etc. As these parameters are imprecise and difficult to measure directly, these can be expressed in fuzzy context. A fuzzy based approach is used to propose a system that measures the overall student's satisfaction.

Internet and Network applications have seen a tremendous growth in the last decade. As a result incidents of cyber attacks and compromised security are increasing. This requires more focus on strengthening and securing our communication. One way to achieve this is cryptography. Although a lot of work has been done in this area but this problem still has scope of improvement. In this paper we have focused on asymmetric cryptography and proposed a novel method by combining the two most popular algorithms RSA and Diffie-Hellman in order to achieve more security.