Informatics Publishing Limited

P. R. Shahane ¹, S. B. Mule ², S. R. Ganorkar ²

Affiliations
1 Sinhgad college of Engineering, Vadagaon (BK), Pune, IN
2 Department of E&TC, IN

Abstract

Noise filters tend to blur image detail, while filters for image sharpening tend to increase noise. So, cascading the two filters does not always give the best performance. We present an integrated filter that reduces noise or sharpens details in a noisy signal, depending on local image statistics. Most traditional noise reduction methods tend to oversuppress high-frequency details. For overcoming this problem we first decompose the input image into flat and edge regions,and remove noise using the alpha map computed from wavelet transform coefficients of LH, HL, and HH bands. After removing noise in the flat region, we further remove noise in edge regions by adaptively shrinking wavelet coefficients based on the entropy. Moreover, we present a new directional transform using wavelet basis and Gaussian low pass filters. The wavelet coefficients of edge regions are inverse transformed by using the filtered wavelet bases. Experimental results show the proposed algorithm can reduce noise without losing sharp details and is suitable for commercial low-cost imaging systems, such as digital cameras and surveillance system.

Keywords

Entropy Analysis, Image Denoising, Noise Reduction, Wavelet Transform.

Full Text

     

M. H. Deokate ¹, S. R. Ganorkar ²

Affiliations
1 Venutai Chavan Polytechnic, Pune, IN
2 Sinhgad College of Engineering, Pune, IN

Abstract

Digital watermark technology is now drawing attention as a new method of protecting digital content from unauthorized copying. Digital watermarking is the process of embedding sideband data within the samples of a digital audio, image, or video signal. A novel audio watermarking algorithm is used to protect against unauthorized copying of digital audio. This paper presents robust and efficient approach of audio watermarking for copyright protection. This watermarking scheme includes a psychoacoustics model to ensure that the watermarking does not affect the quality of the original sound. The Direct Sequence Spread spectrum (DSSS) technique is taken to spread the watermark bits over the entire spectrum of audio signals without the violation of psycho acoustic properties of Human Auditory System (HAS) in frequency domain. The size of watermark depends upon the size and numbers of audio file samples. After embedding the watermark, our scheme extracts copyright information without access to the original signal .The watermarking scheme is robust against common signal processing attacks and it introduces no audible distortion after watermark insertion. The watermark must be imperceptible to the intended audience of the host signal, and must withstand distortion from common signal processing operations.

Keywords

Audio Watermarking, Digital Watermark, Copyright Protection, Psychoacoustics Model.

Full Text

     

S. R. Ganorkar ¹, B. L. Gupata ¹

Affiliations
1 National Institute of Technical Teachers Training and Research, Bhopal, IN

Abstract

The quality of products of an institution and services of the institution is achieved through scientifically designed academic systems. Quality is never achieved as an accident. Sincere efforts, dedication and good intentions of the persons behind it, make quality. Quality concept is not new in India for education sector. It rests great responsibilities on the shoulders of educational institutions to demonstrate publicly that they are geared up to provide quality education and quality services to the society consistently for a long duration. It is, therefore, essential to develop the system that ensures the quality of the educational processes and consequently the quality of graduates. In the last two decades, it has been observed that the mushrooming growth of the large number of professional and technical institutions in the country with limited resources has affected the academic standards. The Quality can only be achieved by the education sector, when all the institutions come together and jointly design a mechanism, which assures the quality of graduates at par with to international standards.

The quality of graduates can be assured through scientifically designed and implemented academic systems. The quality of processes can be assured using academic audit approach.

In this paper an attempt is made to introduce Academic Audit as a system to assure and ensure the Quality of academic processes.

Full Text

     

Dipali A. Sable , S. R. Ganorkar

Abstract

Breast cancer is one of the most common cancer among women in India. Early detection of microcalcification cells is very important stage for the further treatment. Calcification is the deposit of calcium in breast tissues. It is helpful to classify benign or malignant. Early detection of cancer depends on the quality of images and the ability of radiologists to read the mammogram images.

In this proposed work, the mammogram images are initially preprocessed using different methods. In this, noise in the background will be removed using median filter, artifacts will be removed using thresholding method and contrast enhancement will be done using contrast limited adaptive histogram equalization techniques.

Then the region of interest will be determined using segmentation by Otsu’s thresholding algorithm. Features of the mammogram images will be extracted using wavelet transform and to determine the information from the images Support Vector Machine classifier will be used.

Keywords

Mammogram Images, Otsu’s Thresholding, Support Vector Machine, Wavelet Transform

Full Text