Volume 3 No 4 Oct 2009
|
| 1. |
Content Based Image Retrieval for Cervical Cytology Images Download PDF |
|
|
Authors |
S. Allwin |
|
|
Abstract |
An image is a collection of objects which can be visually characterized by its shape, color, size and spatial location. These characteristics can be used to define what the image means. Content Based Image retrieval indexes these features in a database and retrieves then when a query image with matching features is presented. This concept when implemented to the medical field is Medical CBIR which is governed by the DICOM standard. The DICOM standard exists for various medical images but not for cervical cyto images. In this paper we have tried to create a standard for cervical cyto images by indexing certain features which would help detect the cancer easily |
|
|
Keywords |
CBIR, Medical CBIR, Cervical Cytology, Cervical Cancer, DICOM Standards |
|
| 2. |
Mitochondrial Genome Classification from Chaos Game Representation Images Download PDF |
|
|
Authors |
Vrinda V. Nair and Achuthsankar S. Nair |
|
|
Abstract |
Species identification using biological sequence classification finds wide theoretical as well as practical applications. Techniques exist for biological sequence classification based mostly on machine learning approaches. Here a probabilistic sequence classification algorithm is presented, based on Bayesian theory, known as naïve Bayesian classification which in turn identifies the species to which a genome fragment belongs. The efficiency of classification is enhanced using a Difference Boosting Algorithm (DBA). Mitochondrial genomes from seven categories of phyla from NCBI were used for the study. The method constructs a Chaos Game Representation (CGR) image of the sequence data, the dimension of which is reduced by constructing a Frequency CGR (FCGR). The resulting lower-dimensional matrix serves as input to the naive Bayesian classifier. Classification accuracy of 90% and above is obtained for all the seven categories. It is also shown that the method gives surprisingly good results, even on short sequences (300 base pairs). |
|
|
Keywords |
chaos game representation images, genomic signature, mitochondrial genomes, difference boosting, naïve Bayesian |
|
| 3. |
Writer-Independent Off-Line Hand Drawn Arabic Numerals Recognition Download PDF |
|
|
Authors |
D.V. Rojatkar, Prof. G.G.Sarate, S.R.Hirekhan |
|
|
Abstract |
Recognition of handwritten, hand drawn characters is an archetypal recognition problem. Input data to the computer can be acquired by variety of resources. Admired methods are scanning the printed document, with digital camera, writing pad etc. In presented experimentation, data is acquired directly through the mouse dragging. The number of sub-images of Arabic Numerals in the inputted image are alienated using 8 neighborhood connectivity. The correlation of the segmented sub-images is computed with the every image in the data base and then the highest correlation value is the indication of the recognised numeric character. The declared result is based on the best match; the correlation value shall have some value above predefined threshold. The programming is done using MATLAB[1]. The character images are acquired through Microsoft Paint 2000 tool. |
|
|
Keywords |
Correlation, Handwritten Character Recognition, OCR, Pattern Recognition |
|
| 4. |
Application of Multi-Resolution Techniques for Medical Image Compression Download PDF |
|
|
Authors |
S. S. Gornale*, Vikas Humbe, Ganesh Magar, Ramesh Manza, K.V. Kale |
|
|
Abstract |
Telemedicine consists of Medical image acquisition, storage, display and processing and image transfer. This acquisition, storage requires an enormous amount of data. To save the storage space and to reduce the bandwidth for communication; the image has to be compressed. Recently the Multi-resolution technique by wavelet transform has emerged as a cutting edge technology within the field of image analysis and compression. Relatively new class of wavelets called Multiwavelets; are a new addition to the body of wavelet theory. Realizable as matrix –valued filter banks leading to wavelet basis, were introduced, which are able to posses all desirable properties like Orthogonality, Symmetry, and Short support etc. simultaneously, which are needed for better performance compression. In this paper we have carried out an experimental work on medical images mri.jpg, bonescan.jpg, follicle.jpg and mammogram.jpg by proposing the new algorithm. The performance of the proposed method is compared and analyzed in detail and the promising results and findings are presented and the predicted results satisfy the existing necessities to a greater extent, to suit the competitive era. |
|
|
Keywords |
Medical Image compression, Wavelet Filters, Retain Energy (RE) and Number of Zeros (NZ), Multi-Resolution Techniques |
|
| 5. |
Secure a Biometric Traits using Nonlinear Tent Map Download PDF |
|
|
Authors |
N.Viswanath,A.Senthil Arumugam |
|
|
Abstract |
Tent Map is a discrete-time piecewise-affine I/O characteristic curve, which is used for chaos-based Applications, Such as True random Number generation. In this Paper ,we propose a Modified tent map, Which is interchangeable with the tent map for Biometric Applications. In the proposed modified tent map, the confident problem is solved while maintaining the functionality of the tent map. |
|
|
Keywords |
Biometric Traits,M-Tent Map,Biometric Encryption |
