Gesture recognition using field programmable gate arrays

Abstract

Gesture recognition is a topic of immense interest in the field of computing and image processing involving numerous factors and constraints nevertheless yielding remarkable simplification of various human affairs. In this work, the problem of gesture recognition is narrowed down to that of hand gesture recognition and specifically deals with finger count extraction to facilitate further processing using the control so effected. A hand gesture recognition system has been developed, wherein the finger count in a certain input hand image is computed in accordance with a simple yet effective procedure. The problem of hand gesture recognition is solved by means of adopting a lucid albeit efficient algorithm which has been implemented using the Xilinx System Generator software tool. The algorithm followed is invariant to rotation and scale of the hand. The approach involves segmenting the hand based on skin color statistics and applying certain constraints to classify pixels as skin and non skin regions.

Architecture Design and VLSI Implementation of 3D Hand Gesture Recognition System

“…Referring to previous works [ 20 , 21 ], both used the number of spreading fingers for recognition, which is not accurate for gesture recognition. Although dynamic gestures can be recognized (as in [ 22 ]), they can only operate in a simpler background. The work in [ 23 ] was designed using a CNN for hand recognition.…”

“…There are some related works in which the stereo vision algorithm is used in gesture recognition. Raj et al [ 22 ] used skin color detection to perform hand segmentation and convert the image into a binary signal and find the centroid of a hand. The hand recognition is done by counting the number of zero-to-one (black-to-white) transitions from left to right to determine the number of fingers.…”

An efficient static gesture recognizer embedded system based on ELM pattern recognition algorithm

“…Examples of vision-based gesture-recognition systems were encountered, in which whole systems were implemented in FPGA. A hand gesture recognition system was proposed in [32] that included skin color segmentation followed by the number of fingers being counted in the image. The system recognized a set of five hand gestures, namely human finger counts from one to five.…”

Segmentation Methods for Image Classification Using a Convolutional Neural Network on AR-Sandbox

“…The study starts from previous results of [7] that had the purpose of creating contributions towards image recognition based on immersive techniques, through deep learning using CNN, hyperparameters optimization and image processing. From this base model, CNN is trained, using a training dataset composed of a series of vowels (A, E, I, O, U), which will be previously put through image processing, trough Python and OpenCV, using three segmentation methods, Canny Edge Detector [6], color-space segmentation [7] and Threshold [8]. When obtaining the trained model, we proceed to make the model evaluation, making use of a test dataset, which was acquired in its entirety through the AR-Sandbox.…”

“…[5]. Authors like [6] use image segmentation in order to get information for underwater images, which are difficult to get and analyze due to environmental conditions, for these reasons, canny edge detector algorithm is used because it gave localization and response with extreme accuracy and it has a superior capability under noise conditions. Finally, in previous studies, [7] where the purpose of this research was to establish a model of a CNN for the classification of geometric figures by optimizing hyperparameters using random search, evaluating the impact of the implementation of a previous phase of colorspace segmentation to a set of tests captured from the AR-Sandbox, authors find that using the proposed method, an average decrease of 39.45% to a function of loss and an increase of 14.83% on average in the percentage of correct answers is presented.…”