Ashok T. Gaikwad

Fingerprint Recognition for Person Identification and Verification Based on Minutiae Matching

There are various types of applications for fingerprint recognition which is used for different purposes. Fingerprint is one of the challenging pattern Recognition problem. The Fingerprint Recognition system is divided into four stages. First is Acquisition stage to capture the fingerprint image, The second is Pre-processing stage to enhancement, binarization, thinning fingerprint image. The Third stage is Feature Extraction Stage to extract the feature from the thinning image by use minutiae extractor methods to extract ridge ending and ridge bifurcation from thinning. The fourth stage is matching(Identification, Verification) to match two minutiae points by using minutiae matcher method in which similarity and distance measure are used. The algorithm is tested accurately and reliably by using fingerprint images from different databases. In this paper the fingerprint databases used are FVC2000 and FVC2002 Databases, we see that, the FVC2002 database perform better results compare with FVC2000 database. The recognition system evaluate with two factor FAR and FRR, In this system the result of FAR is 0.0154 and FRR is 0.0137 with Accuracy equal to 98.55%.

Study of edge detection methods based on palmprint lines

The palm was used in fortune telling 3000 years ago. Thus, During this period, many different problems related to palmprint recognition have been addressed. In the recent years, the palm print has been used for biometric applications as human verification and identification. The palm print has many features comparing with a fingerprint, The palm print has number of lines. One group of these lines is known as the principle lines which contains three lines(head line, heart line and life line). The lines are extracted from palm print image by edge detection algorithm which is implementing on ROI of palm print. The main goal of edge detection algorithm is to produce a line and extract important features and reduce the amount of data in the image. This paper investigates the several edge detection methods such as Sobel, Prewitt, Roberts, LOG, and Canny. In addition, we used edge detection using local entropy information and local variance. The experiment is tested on samples taken from four palm print databases (CASIA, PolyU, IIT and database available online). The analysis work has been performed by using PSNR and MSE of resultant images on these popular edge detection methods which improve the palm print matching process. The Prewitt, Roberts and LOG edge detection methods ignore the small lines and identify only the main longer lines while the Sobel identifies the medium and longer lines. The canny edge detection algorithm identifies the complete set of edges of various sizes. From experiment it was seen that good result found with an online database and polyU database by classical edge detection methods.

Overview of fingerprint recognition system

This article is an overview of a current research based on fingerprint recognition system. In this paper we highlighted on the previous studies of fingerprint recognition system. This paper is a brief review in the conceptual and structure of fingerprint recognition. The basic fingerprint recognition system consists of four stages: firstly, the sensor which is used for enrolment & recognition to capture the biometric data. Secondly, the pre-processing stage which is used to remove unwanted data and increase the clarity of ridge structure by using enhancement technique. Thirdly, feature extraction stage which take the input from the output of the pre-processing stage to extract the fingerprint features. Fourthly, the matching stage is to compare the acquired feature with the template in the database. Finally, the database which stores the features for the matching stags. The aim of this paper is to review various recently work on fingerprint recognition system and explain fingerprint recognition stages step by step and give summaries of fingerprint databases with characteristics.

Plant Classification Using Image Processing and Neural Network

This paper is presenting here the plant classification method using imaging technology which is useful for classifying the plants by providing the leaf image as an input. The proposed method performs classification by automatically extracting shape patterns and features performing the image processing techniques and neural network model. This method gets the leaf image as an input, performs the leaf image processing tasks, and automatically extracts the leaf shape pattern and leaf shape features. It performs the classification with the help of leaf shape features using neural network techniques. This method presents plant classification using the leaf shape features and feed forward back propagation neural network model. This method results in up to 99% accuracy of classification. This method extracts the leaf shape features and patterns automatically using image processing techniques.

A Robust Voice-Based Color Object Detection System for Robot

The robot is an intelligent machine or intelligent agent. When the concept of artificial intelligence is applied to machines, it mimics the functions performed by human such as decision making, learning, recognition, problem solving. The robot exactly mimics the function of humans. Automatic color object detection and tracking of it by listening audio instruction is a function performed by every human. The proposed system receives acoustic instruction as an input and analyzes the video frames and output, the location of a moving object within the video frames. The aim is to implement this system for robots, so it will behave like the human.

Image inconsistency detection using histogram of orientated gradient (HOG)

Today there are various types of image editing tools which make totally changes in image with free of cost, Image has performed a significant role in Human life but image has easily fiddle using image processing software. Fiddle image has difficult to detect that it is original or not for this reasons the image forgery detection topic is active research work nowadays. The proposed of this paper to detect image inconsistency using Histogram of Orientated Gradient (HOG) method which help us to determining which block has manipulation of an images. The paper conducting with many stages namely acquisition, preprocessing, and feature extraction and matching the performance of this system are based on false accepted rate (FAR) and false reject rate (FRR)