Ramesh R. Manza

Personal Identification Algorithm Based on Retinal Blood Vessels Bifurcation

Biometric identifiers are the unique, measurable characteristics used to tag and describe individuals. Physiological characteristics are related to the shape of the body. Examples of biometric identifications are, fingerprint, face, DNA, Palm print, hand geometry, iris recognition, and retina. Human retina is another source of biometric system which provides the most reliable and stable means of authentication. We propose a new algorithm for the detection and measurement of blood vessels of the retina and finding the bifurcation points of blood vessels for personal identification. A minutiae technique for finding bifurcation points of the extracted blood vessels and according to bifurcation points identifies the individual person. Performance of these techniques is tested using the database from Dr. Manoj Saswade and Dr. Neha Deshpande (300 Images). This algorithm achieves a true positive rate of 98%, false positive rate of 20%, and accuracy score of 0.9702 and also classification down through Statistical Techniques.

Leaf shape extraction for plant classification

This research paper presents the leaf shape extraction for plant classification. Leaves are very important component of the plant which actually identifies and classify the plants. Classification of the plant by their leaf biometric features is commonly performed task of trained botanist and taxonomist. To perform this task they need to perform various set of operations. Because of this the task of classification of plants manually is time consuming. There are many biometric features of leaves of the plants for classification. Here the shape of leaves of the plant species are extracted for plant classification. In this paper, various operators are studied for the leaf extraction from images by using the image processing techniques.

Development of primary glaucoma classification technique using optic cup & disc ratio

Glaucoma is Eye dieses & one of the leading causes of blindness worldwide. It is due to the increase in intra ocular pressure within the eyes. The detection and diagnosis of glaucoma is very important. Here we present an algorithm which works on two different data base for the same purpose to calculate optic cup to disc ratio. For this purpose we use DRIONS-DB high resolution fundus images. From these high resolution RGB images we are going from preprocessing to ROI extraction steps i.e. our optic disc & cup detection. For that purpose we use K-means clustering for detection & measure area of disc & cup then, calculate ratio. After calculating ratio we apply that same method on another database i.e. RIM-I 64 healthy images & we got healthy CDR between 0.2 to 0.6. By the combination of two data base DRIONS-DB & RIM-Il, overall 97% result is achieved.

Detection of non-proliferative diabetic retinopathy lesions using wavelet and classification using K-means clustering

WHO predicts that in year 2012 there are about 347 million people worldwide have diabetes, more than 80% of diabetes deaths occur in different countries. WHO projects that diabetes will be the 7th major cause leading death in 2030. Diabetic Retinopathy caused by leakage of blood or fluid from the retinal blood vessels and it will damage the retina. Non-proliferative diabetic retinopathy (NPDR) is an early stage of diabetic retinopathy and it is categorized into three stages they are mild, moderate and sever NPDR. The characteristic of the Mild; is specified by the presence of minimum microaneurysm, Moderate; specifies the presence of hemorrhages, microaneurysms, and hard exudates where as Severe; determine on the blockage of vessels, depriving several areas of the retina. With their blood supply. These areas of the retina send signals to the body to grow new blood vessels for nourishment. The proposed algorithm tested on online databases like STARE, DRIVE, DiarectDB0, DiarectDB1 and SASWADE (the database collected during the research work). The statistical techniques were applied on NPDR lesion and calculate the mean, variance, standard deviation, & correlation for classification. K-means clustering have been applied on the dataset with extracted features 95% of correct classification have been achieved.

Secondary glaucoma diagnosis technique using retinal nerve fiber layer arteries

Glaucoma is an eye disease. In glaucoma retinal nerve fiber layers are damaged and if it is not treated earlier then it can cause permanent vision loss. This paper represents algorithm for detection of glaucoma using retinal nerve fiber layers. For this work we have used 2D median filter and HAAR wavelet transform methods. For this work we have also used Drishti-GS dataset which contains 101 glaucomatous images and HRF (High Resolution Fundus image) database. We have extracted the retinal nerve fiber layer Arteries. Then we have calculated its area and diameter. On the normal database we got the 100% result. We got 71.28% accuracy on glaucomatous images and when we have combined the normal and glaucomatous images then we got the 62.06% accuracy.

Automatic Video Scene Segmentation to Separate Script and Recognition

Text or character detection in images or videos is a challenging problem to achieve video contents retrieval. In this paper work we propose to improved VTDAR (Video Text Detection and Recognition) Template Matching algorithm that applied for the automatic extraction of text from image and video frames. Video Optical Character Recognition using template matching is a system model that is useful to recognize the character, upper, lower alphabet, digits& special character by comparing two images of the alphabet. The objectives of this system model are to develop a model for the Video Text Detection and Recognition system and to implement the template matching algorithm in developing the system model. The template matching techniques are more sensitive to font and size variations of the characters than the feature classification methods. This system tested the 50 videos with 1250 video key-frames and text line 1530. In this system 92.15% of the Character gets recognized successfully using Texture-based approaches to automatic detection, segmentation and recognition of visual text occurrences in images and video frames.

A study of eye tracking technology and its applications

We can measure the eye movement activity using eye tracking technology. Eye tracking gives us information about where do we look? What is ignored and how the pupil reacts to different stimuli. The eye tracking concept is basic but its process and interpretation can be very diverse and complex. ET measures the gaze points generated by our eye relative to the head. Eye trackers are availbble in either remote or mobile forms. It tracks and records where do we look and how we move the gaze. One can analyze, visualize and interpret this information with the help of software. We have gone through the common use of fingerprint analysis and applications, eye tracking also would be a great biometric tool for analysis in various applications. In this paper we discuss eye tracking technology and its various applications. Now days, ET is being employed in almost all field including psychology, human computer interaction, marketers, designers, academics, medical, research and many more.

Design New Biorthogonal Wavelet Filter for Extraction of Blood Vessels and Calculate the Statistical Features

World health organization predicts that in year 2012 there are about 347 million people worldwide have diabetes, more than 80 % of diabetes deaths occur in different countries. WHO projects that diabetes will be the 7th major cause leading death in 2030. Diabetic Retinopathy caused by leakage of blood or fluid from the retinal blood vessels and it will damage the retina. For extraction of retinal blood vessels we have invent new wavelet filter. The proposed filter gives the good extraction result as compare to exiting wavelet filter. In proposed algorithm, we have extract the retinal blood vessels features like area, diameter, length, thickness, mean, tortuosity, and bifurcations. The proposed algorithm is tested on 1191 fundus images and achieves sensitivity of 98 %, specificity of 92 % and accuracy of 95 %.