In this paper, we propose an automatic system for recognizing continuous gestures in real-time, including Arabic numbers (0-9) and alphabets (A-Z). We present an improved method for hand area detection and segmentation based on YCbCr and HSI mixed skin color space, the improved CAMSHIFT algorithm used for hand tracking. Orientation dynamic features are obtained from spatiotemporal trajectories and then quantized to generate its code words. An improved HMM-FNN model is proposed for gesture recognition based on the code words, which combines ability of HMM model for temporal data modeling with that of fuzzy neural network for fuzzy rule modeling and fuzzy inference. The algorithm we presented has better performance and achieves average recognition rate 95.76% and 93.64% for Arabic Numbers and Alphabets, respectively.

Continuous Gesture Trajectory Recognition System Based on Computer Vision

In this paper, we will present a method to detect human hand and recognize hand gesture. For detecting the hand region, we use the feature of human skin color and hand feature (with boundary complexity) to detect the hand region from the input image; and use algorithm of optical flow to track the hand movement. Hand gesture recognition is composed of two parts: 1. Posture recognition and 2. Motion recognition, for describing the hand posture feature, we employ the Fourier descriptor method because it's rotation in variant. And we employ PCA method to extract the feature among gesture frames sequences. The HMM method will finally be used to recognize these feature to make a final decision of a hand gesture. Through the experiment, we can see that our proposed method can achieve 99% recognition rate at environment with simple background and no face region together, and reduce to 89.5% at the environment with complex background and with face region. These results can illustrate that the proposed algorithm can be applied as a production.

/pdf/hand-gesture-recognition-using-optical-flow-field-2ixlqwn4k1.pdf

Hand Gesture Recognition using Optical Flow Field Segmentation and Boundary Complexity Comparison based on Hidden Markov Models

The present disclosure is directed to a display device and a method of controlling the display device. The display device includes a memory that stores at least one first gesture image corresponding to at least one function, a camera that obtains a basic image including user's second gesture image, and a controller that extracts the second gesture image from the basic image and displays at least one third gesture image similar to the extracted second gesture image, wherein the third gesture image is included in the first gesture image stored in the memory. The display device and method of controlling the display device may guide a user to make a correct gesture by displaying a gesture image similar to a gesture made by the user among gesture images stored in a memory.

Display device and method of controlling the same

The purpose of this study is to support the bedridden and physically handicapped person who live independently. In this study, we developed Electric appliances control system that can be used on the bed. The subject can control Electric appliances using hand motion. Infrared sensors of a Kinect are used for the hand motion detection. Our developed system was tested with some normal subjects and results of the experiment were evaluated. In this experiment, all subjects laid on the bed and tried to control our system. As results, most of subjects were able to control our developed system perfectly. However, motion tracking of some subject's hand was reset forcibly. It was difficult for these subjects to make the system recognize his opened hand. From these results, we think if this problem will be improved our support system will be useful for the bedridden and physically handicapped persons.

Development of the bedridden person support system using hand gesture

In this paper, an intelligent music player system based on computer vision is proposed. We present an improved method for skin color area detection and segmentation based on YCbCr and HSI mixed skin color space. Then we use the Adaboost learning algorithm to obtain the face detector based on the Haar wavelet transform with in skin color region and receive the location of eyes and mouth. The frame subtraction algorithm is used to achieve position of hand motion, we used CAMSHIFT algorithm to track the motion trajectory and estimate the center of hand, which used for controlling the Music Player. Afterwards, the system will recognize the hand motion by calculating the relationship between the mouth location and the position of hand area in real time. Therefore, experimental results illustrate this intelligent music controlling system has satisfying interactive input functions in real time, and it can be applied in different illumination condition and complex background well.

Intelligent Music Player Based on Human Motion Recognition

In this paper, we present a novel method of hand detection and improve the CAMSHIFT algorithm for hand tracking. We first detect the hand by using YCbCr hand-color likelihood. It is based the nature of skin color clustering in the distribution of CbCr and the maximum likelihood principle. Then, the detect result is applied to hand tracking by using CAMSHIFT algorithm. We have improved the formula of the search window to suitable for hand. Finally we carried out experiments on a Windows XP system and achieved good tracking results.

The Hand Mouse: Hand Detection and Hand Tracking

In this paper, we proposed a new multi-view human recognition method based on face and gait features detection algorithm. For getting the position of moving object, we used the different of two consecutive frames. And then, base on the extracted object, the first important characteristic, walking direction, will be determined by using the contour of head and shoulder region. If this individual appears in camera with frontal direction, we will use the face features for recognition. The face detection technique is based on the combination of skin color and Haar~like feature whereas eigen-images and PCA are used in the recognition stage. In the other case, if the walking direction is frontal view, gait features will be used. To evaluate the effect of this proposed and compare with another method, we also present some simulation results which are performed in indoor and outdoor environment. Experimental result shows that the proposed algorithm has better recognition efficiency than the conventional single view recognition method.

Multi-view Human Recognition based on Face and Gait Features Detection

In this paper, we will propose a new-style remote controller system. The system uses human hand as commander to control the IP-TV. In the system, we detect the hand based on human skin color feature, and describe the hand shape information using Fourier descriptor. In the recognition part, we will use SVM based recognizer to classify the Fourier descriptor of hand feature. In the experiment part, we define 10 kinds of hand posture which is come from Chinese traditional hand-gesture. The experiment results show that our hand gesture remote controller system can achieve 97% recognition rate in the natural light environment. In this case, the system can be applied to IP-TV system.

An IP-TV Remote Controller based on Hand Posture Recognition using Fourier Descriptor

In this paper, we propose a human activity recognition algorithm based on gait and arm swing feature information. For the recognition of human's activity, we first detect moving object and then extract human's biometric features such as joint angle, swing amplitude, cycle, and speed of leg and arm. In the experimental results, the proposed method shows better recognition performance than conventional method.

Jae-Ho Shin

Papers

The Hand Mouse: Hand Detection and Hand Tracking

Multi-view Human Recognition based on Face and Gait Features Detection

An IP-TV Remote Controller based on Hand Posture Recognition using Fourier Descriptor

Gait and Arm Swing Detecting for Human Activity Recognition