Feature Extraction Technique for Static Hand Gesture Recognition

Abstract

The goal of static hand gesture recognition is to classify the given hand gesture data represented by some features into some predened nite number of gesture classes.The main objective of this eort is to explore the utility of two feature extraction methods, namely, hand contour and complex moments to solve the hand gesture recognition problem by identifying the primary advantages and disadvantages of each method. Articial neural network is built for the purpose of classication by using the back-propagation learning algorithm. The proposed system presents a recognition algorithm to recognize a set of six specic static hand gestures, namely: Open, Close, Cut, Paste, Maximize, and Minimize. The hand gesture image is passed through three stages, namely, pre-processing, feature extraction, and classication. In the pre-processing stage some operations are applied to extract the hand gesture from its background and prepare the hand gesture image for the feature extraction stage.In the rst method, the hand contour is used as a feature which treats scaling and translation problems (in some cases). The complex moments algorithm is, however,

A probabilistic model for state sequence analysis in hidden Markov model for hand gesture recognition

“…After the preprocessing of the hand image, extract the discriminant feature vector points from the hand gestures stored in the database . Feature extraction is a predominant technique used to minimize the data size (information) in the data set .…”

Vision Based Classification of Hand Gestures towards Human Robot Interaction by Using Neural Network

“…The aim of using edge detection is to simplify the hand gesture image but to have more meaning full image since the edges are known as basic image features carrying useful information regarding the object boundaries. Edge detection algorithms are also used to find complicated boundaries belonging objects based on marking the potential edge points, which correspond to the points in the image where there are changes in brightness [16]. Among the large number of edge detection algorithms, Sobel algorithm working based gradient is used due to its popularity on computational simplicity [3].…”