Highly Accurate Static Hand Gesture Recognition Model Using Deep Convolutional Neural Network for Human Machine Interaction

In this work, we propose a deep convolutional neural network (DCNN) based model for static hand gestures recognition. Static hand gesture images corresponding to five different classes are presented to DCNN model without any preprocessing. The model has achieved a train and test accuracy of 97.9% and 99.6% respectively which is one of the best ever reported accuracy in static hand gesture recognition applications. It is also found that the performance of the model is good even with complex backgrounds and poor lighting conditions. Due to its accuracy and robustness, this model can be implemented in applications such as human machine interaction and autonomous cars.