The technology in present-day is very useful for human-physical recognition and the hand gesture recognition is one of them. Nevertheless, that technology still had various weakness, for example in the image brightness, contrast, recognition time, and accuracy rate. The objective of this paper was to construct hand gesture recognition system using RGB images. Pre-processing is wrought by resizing the image, separate the hand area, and pick the specific layer. This experiment used the YCbCr because its derived directly from RGB and had a higher contrast compared to other layers. The feature value was gathered from feature extraction on Discrete Wavelet Transform (DWT) using Low-Low sub-band and 2nd level decomposition. The current sub-band had smoothest contours in comparison with other sub-band. The final process was gesture classification using Hidden Markov Models (HMM) and K-Nearest Neighbor (KNN). The amount of training and testing data used were 150 and 100 images respectively, divided into five gestures with accuracy using HMM and KNN consecutively was 58% and 100%. The research novelty was that the classification impacted positively on accuracy level.

Gesture recognition based on computer-vision is an important part of human-computer interaction. But it lacks in several points, that was image brightness, recognition time, and accuracy. Because of that goal of this research was to create a hand gesture recognition system that had good performances using discrete wavelet transform and hidden Markov models. The first process was pre-processing, which done by resizing the image to 128x128 pixels and then segmented the skin color. The second process was feature extraction using the discrete wavelet transform. The result was the feature value in the form of a feature vector from the image. The last process was gesture classification using hidden Markov models to calculate the highest probability of feature matrix which had obtained from the feature extraction process. The result of the system had 72% of accuracy using 150 training and 100 test data images that consist five gestures. The newness thing found in this experiment were the effect of acquisition and pre-processing. The accuracy had been escalated by 14% compared to Sebastien's dataset at 58%. The increment effect propped by brightness and contrast value.

Public services are available to all communities including people with disabilities. One obstacle that impedes persons with disabilities from participating in various community activities and enjoying the various public services available to the community is information and communication barriers. One way to communicate with people with disabilities is with hand gestures. Therefore, the hand gesture technology is needed, in order to facilitate the public to interact with the disability. This study proposes a reliable hand gesture recognition system using the convolutional neural network method. The first step, carried out pre-processing, to separate the foreground and background. Then the foreground is transformed using the discrete wavelet transform (DWT) to take the most significant subband. The last step is image classification with convolutional neural network. The amount of training and test data used are 400 and 100 images repectively, containing five classes namely class A, B, C, # 5, and pointing. This study engendered a hand gesture recognition system that had an accuracy of 100% for dataset A and 90% for dataset B.