Volume 1 No 3 July 2007
|
| 1. |
A Novel Extension to Chain Code for Reversible Contour Representation Download PDF |
|
|
Authors |
Krishnan Nallaperumal , Justin Varghese, S.Saudia, K. Krishnaveni, S. Allwin, P. Kumar |
|
|
Abstract |
A novel contour representation algorithm for binary images is proposed in this paper. It is an extension of the conventional chain code. The precise, efficient and simple algorithm exploits the features of the conventional chain code when advantageous. More robust to noise, the algorithm also uses very low bit-rate. The corresponding reconstruction algorithm is completely reversible to give the lossless reconstruction of the contour. Experimental results on various binary images add to the improved efficiency claim of the proposed contour representation algorithm in terms of visual fidelity and bit-rate |
|
|
Keywords |
|
|
| 2. |
New effective Iterative Demosaicing Download PDF |
|
|
Authors |
N.Krishnan , C.Seldev Christopher, S.S.Vinsley,R.K.Selvakumar, C. Nelson Kennedy Babu, Subban Ravi |
|
|
Abstract |
Demosaicing is a process of obtaining a full color image by interpolating the missing colors of an image captured from a digital still and video cameras that use a single-sensor array. This paper provides an effective iterative demosaicing algorithm applying a weighted-edge interpolation to handle green pixels followed by a series of color difference interpolation to update red, blue, and green pixels. Then border of the image is enhanced using some adaptive techniques. Experimental results show that the proposed method performs much better than other latest demosaicing techniques in terms of image quality. In comparison to the other algorithms, this algorithm proposed here results in increased PSNR value. |
|
|
Keywords |
Bayers sampling, Iteractive demosaicing, color interpolation, color difference |
|
| 3. |
Face Detection using Multi-Scale Morphological Segmentation Download PDF |
|
|
Authors |
Nallaperumal Krishnan, S. Ravi , Krishnaveni. K, Member, Justin Varghese |
|
|
Abstract |
A novel face detection technique is proposed in this paper, which uses multi-scale morphological segmentation. This technique works satisfactorily on gray scale face images containing bright and dark features of various scales even in an impulse-corrupted environment. The face detection algorithm involves four stages. In the first stage, using an Iterative Adaptive Switching Median Filter, which reduces the impact of impulse noise that cause over segmentation, preprocesses the face image. In the second stage, the multiple scales of bright and dark features of different objects are extracted by the respective opening and closing of the preprocessed image. The resultant image is binarized using an optimum threshold. In the third stage, valid segments of the bright top-hat and dark bottom-hat images are detected and the contours of these face images are combined to give the final skin segmented face image. In the last stage, the facial region is detected from the skin-segmented image using template matching technique. The scheme is implemented on a set of test face images and the performance of the algorithm is proved better both objectively and subjectively than the standard methods. |
|
|
Keywords |
Face detection, Multi-scale morphology, Impulse noise, Median filters, Template matching |
|
| 4. |
Semantic-Based Indexing and Retrieval of CT - Brain Images using Asymmetric Features Download PDF |
|
|
Authors |
Dr.B.G. Prasad , Krishna. A N ,M Dhananjay |
|
|
Abstract |
Medical images play a central role in patient diagnosis, therapy, surgical planning, medical reference, and training. It is therefore natural to use these images as a front-end index to retrieve medically relevant cases from digital patient databases. To speed up the search process, selected features are extracted from each image when they are stored in the database, so each one is represented by a feature vector. Subsequent image searching operations are performed using the feature vectors in place of images. In this paper, we present a way to approach semantic analysis of an image using asymmetry features. Asymmetry features are obtained by voxelwise comparison of corresponding left and right half brains by extracting manual bilateral symmetry (Midsagittal) plane and calculating the differences between the left and right side pixels of the brain. Then a specific treatment is given to the difference to identify the presence of possible lesions |
|
|
Keywords |
|
|
| 5. |
Modified Standard Backpropagation Algorithm with Optimum Initialization for Feedforward Neural Networks Download PDF |
|
|
Authors |
V.V.Joseph Rajapandian, N.Gunaseeli |
|
|
Abstract |
Modified Backpropagation algorithm (MBP), an approach for the learning process of multilayer feedforward neural network with optimum initialization is proposed. One of the common complaint about the Standard Backpropagation algorithm (SBP) is that it is very slow and even simple problems may take hundreds of iterations to converge. SBP algorithm reduces only non-linear errors. Much work, therefore has been done in search of faster methods. One of such approach is modified form of the Standard Backpropagation algorithm. Modified Backpropagation algorithm consists of minimizing the sum of the squares of linear and non-linear errors for all output units. This leads to an efficient process in the network. Proper initialization always plays a key role in the robust neural networks. Therefore, the optimum initialization method is used for weight initialization which ensures that the outputs neurons are in the active region and the range of activation function is fully utilized. The proposed method is implemented on 2 bit parity problem, 4 bit parity checker and encoder problem and produced good results |
|
|
Keywords |
Linear error, modified standard backpropagation algorithm (MBP), neural network(NN), nonlinear error, standard backpropagation(SBP),optimum initialization. |
|
| 6. |
A Method of Shape Recognition Using the Smoothed Group Delay Function Download PDF |
|
|
Authors |
Sreyasee Das Bhattacharjee , Dr. Sukhendu Das ,Dr.Amitabha Datta |
|
|
Abstract |
Fourier descriptors have been used in the past as a feature for describing the shape through its boundary contour. The contour based Fourier transform and the corresponding phase information have been used implicitly, to improve the results for image matching and retrieval. The phase of a signal stores the higher order moments and hence can be used to discriminate among various shapes. This paper exploits phase information for signal explicitly by taking the derivative of the phase, termed as the smoothed Group Delay, and then uses it for shape matching. We try to extract the features of a shape from the Group Delay function (of the boundary) and try to classify them. Results are shown on a standard database of object shapes. |
|
|
Keywords |
Phase, SGD, Correlation, and Sim |
|
| 7. |
Optimal Discrete Wavelet Design for Cardiac Signal Download PDF |
|
|
Authors |
Shantha Selva Kumari, S.Bharathi, V.Sadasivam |
|
|
Abstract |
Wavelets have emerged as powerful tool for signal processing. Wavelet transform is used to represent the signal to some other time-frequency representation better suited for detection and removing redundancies. Electrocardiogram is extensively used as a low cost diagnostic tool to provide information concerning the heart performance. The design of good wavelet for cardiac signal is discussed from the perspective of orthogonal filter banks. In this work, two wavelets are developed and evaluated based on perfect reconstruction conditions. ECG signal records from MIT-BIH Arrhythmia database are chosen for processing. In the first step, the filters are designed by reparametrization of filter coefficients by θ. The designed filters are perfectly matched. This method of design leads to constraints like orthogonality, normality on filter coefficients. ECG signal is decomposed to three levels and then reconstructed for the same number of levels. Error signal from the reconstructed signal is compared between db4, bior4.4 and the proposed wavelets W1 and W2. The reconstructed results show the potential of the method. The wavelet W2 gives error in the range of 1.8*10-11 to 2.73*10-11 that is better than all other wavelets already exists in the literature. |
|
|
Keywords |
Cardiac signal, orthogonality, reparameterization, perfect reconstruction, wavelets |
|
| 8. |
Image Retrieval using Shape Feature Download PDF |
|
|
Authors |
S.Arivazhagan, L.Ganesan, S.Selvanidhyananthan |
|
|
Abstract |
While most traditional image retrieval systems perform searches using comparisons of text based strings, content based systems extract features from the content of an image to judge its similarity with another. There are three main types of features that are extracted from images: color, texture and shape. The first two approaches have been explored more thoroughly than shape based approaches. The focus of this paper is on shape based image retrieval. Like any other features based on human perception, the major problem in the use of shape is how to represent shape information, and how to describe the shape of an object. A new and easy to use technique for representing shapes for the purpose of image retrieval is proposed. The proposed representation is a contour based approach. Canny operator is used to detect the edge points of the image. The contour of the image is traced by scanning the edge image and re-sampling is done to avoid the discontinuities in the contour representation. The resulting image is swept line by line and the neighbor of every pixel is explored to detect the number of surrounding points and to derive the shape features. The good result obtained by this method confirms its effectiveness |
|
|
Keywords |
|
|
| 9. |
Performance Analysis of Image Denoising System for different levels of Wavelet decomposition Download PDF |
|
|
Authors |
S.Arivazhagan, S.Deivalakshmi, K.Kannan |
|
|
Abstract |
In diverse fields from planetary science to molecular spectroscopy and medical imaging to satellite imaging, scientists are faced with the problem of recovering original images from incomplete, indirect and noisy images. The conventional Fast Fourier Transform (FFT) based image denoising method is essentially a low pass filtering technique in which edge is not as sharp in the reconstruction as it was in the original. The drawback of the FFT is the fact that the edge information is spread across frequencies because of the FFT basis functions, not being localized in time or space and hence low pass-filtering results in the smearing of the edges. But the localized nature of the wavelet transform both in time and space results in denoising with edge preservation. In this paper, the performance of an Image Denoising System using Discrete Wavelet Transform (DWT) is experimentally analyzed for four levels of DWT decomposition, for Speckle noise added two facial and two CT images. |
|
|
Keywords |
|
|
| 10. |
Intelligent Segmentation of Industrial Component Images Download PDF |
|
|
Authors |
Commander K.Velu and V.Selladurai |
|
|
Abstract |
An application of machine vision, incorporating neural networks, which aims to fully automate real-time inspection in component identification process, is described. The current methodology adopted comprises two distinct stages - the segmentation of the component from the background content of the image, and the segmentation of suspect defect areas inside the region itself. In the first stage, preprocessing and enhancement, segmentation techniques have been employed to adaptively and accurately segment the region from a given image. The second processing stage utilizes a feature extraction and classification further Back propagation network which is trained on a test set of image data previously segmented by a conventional adaptive threshold method. It is shown that the two techniques can be combined to fully segment images. |
|
|
Keywords |
Automated inspection, Back propagation neural networks, image segmentation |
|
| 11. |
A Non-Linear Iterative Impulse Noise Removal Download PDF |
|
|
Authors |
R.K.Selvakumar, C. Nelson Kennedy Babu |
|
|
Abstract |
A novel approach in Non Linear filtering is presented for the noise reduction of images corrupted with salt and pepper noise. This filtering approach consists of two steps. The first step computes the difference matrix for the corrupted image in eight different directions. The second step replaces the noisy pixel’s value with the suitable neighbor’s value whose difference is a minimum value. The filter is applied iteratively to effectively reduce noise. Experimental results are obtained to show the feasibility of the proposed approach. These results are also compared with other filters by numerical measures and visual inspection. |
|
|
Keywords |
|
