Search

Gait is one of the biological information which indicates how to walk. Gait is able to acquire from a distance and any views. It is, however, difficult to identify a person with belongings. The purpose of this paper is, therefore, to propose an effective identification method when the subject has a bag. First, GEI(Gait Energy Image) is generated with head position adjustment. Then, gradient of GEI for head, both arms and leg regions are obtained. Finally, features are obtained from each region except for the hand region with a bag and the bag itself in the leg region. The proposed method obtains 95.0% identification ratio for back view gait with belongings.

The paper proposes a novel gesture recognition and finger tip detection algorithm for Human Computer Interaction, in particular mouse control operations using real time camera. The hand gestures are captured using real time camera. First, the hand region alone is segmented using region growing algorithm followed by morphological operations. The centroid of the palm region is calculated and the finger tips are then detected using the convex hull algorithm. The proposed method is tested on five different gestures and the results prove that the gestures are able to be recognized and the finger tips detected. The method can be applied for hand gesture controlled mouse operations.

This paper proposes a sign language translation system for communication among deaf and common people. It uses an algorithm about hand detection and analysis from previous studies and project assignments. The algorithm performs five steps within the limitations, which is the key that it measures a center of gravity with 4 vertexes from sign language images and sifts the latent features from calculated distance, angle for 5 determined points of hand region through specific experiments. Applying the features, success rate is improved from 23% to 100%. The system can make use of development for human-robot interface.

In this paper, we introduce a two-stage Hand Posture Recognition system using a Kinect sensor. In the first stage, we propose an effective algorithm to segment hand region from complex background, which incorporates both color, depth and skeleton information, without specific requirements on uniform-colored or stable background. It is robust with skin-color noise like arms and face, and can handle both-hands gesture. In the second stage, we use a standard Nearest Neighbor to recognize the posture. Experiments shows that the proposed system achieves high recognition accuracy.

In this paper, we proposed a dynamic hand gesture recognition system by using only the depth information. The proposed system can recognize twelve different dynamic hand gestures, including swipes, scales, push, wave, rotates, circle, and drag. First, the background subtraction is used to remove the unnecessary information, and the depth information of main user can be obtained. Furthermore, hand position can be tracked, and the region of hand is extracted as an adaptive square. Once the region of hand is obtained, the hand parameters are obtained by calculating the depth information of hand region. The proposed system can recognize dynamic hand gesture by using the hand parameters. In the experiment, the performance of the proposed system is verified by two different people at 2 different depths, and both right and left hands are verified. The experimental result show that the proposed system can recognize the dynamic hand gestures with an average recognition rate of 90.08%.

First there was input of information or commands through keyboard, and then touch input was introduced. Now the big question is, What Next? The answer is Gesture Recognition. The best method for gesture recognition is through image processing techniques. In this paper a simple yet effective approach has been used to detect the number of fingers raised. The algorithm presented in the paper gives much better results that other algorithms used for the purpose. This paper used a 30 fps with up to 30 mega pixel camera. Furthermore, pre-processing is done to segment the hand region from its background. The proposed algorithm can be used for controlling basic applications such as image viewing etc.

This paper introduces a human computer interaction system by using hand gesture recognition to apply into game-based rehabilitation application. The basic concept of this project is to assist patient to carry out rehabilitation with hand gesture recognition, meanwhile therapist is able to follow recovery progress of patient through results. In this manner, a rehabilitation system created aimed to increase efficacy in patient treatment with interactive game. The system is applied in real time application to contribute as a function of a virtual mouse. Approach proposed including hand region segmentation, hand palm partition and finger counting. Computer mouse is configured to move according to the location of hand palm region, and perform function of ordinary mouse according to number of fingers detected. In the analysis, experiments is carried out to show how this system collaborate with the simple rehabilitation game in computer. Experiment evaluation as shown has successfully demonstrate the effectiveness and robustness in performing the mouse function and feasible to be implemented into rehabilitation games.

We present a vision-based approach for positioning and tracking fingertip in a video. Multiple cues are fused through defining the likelihood probability terms in particle filtering framework. Skin color has been proven its robustness toward hand region localization in complex background. Thus, we describe it by a Gaussian distribution and further define a skin-color likelihood term. For lighting invariance, we also incorporate the contour information to define two contour likelihood terms. They respectively model the fingertip contour and two-side boundaries of finger. However, the particle filtering generally has degradation problem. To overcome this, we embed the mean shift to the particle filtering framework for convergence consideration. Finally, we validate the proposed approach by providing some experimental results.

This paper presents a real-time 3D hand posture estimation using the data from a RGB-D camera. A hierarchical method is adopted to first estimate the pose of palm and then the pose of fingers. The particle filtering algorithm with efficient particle generation for tracking the palm and fingers is proposed by analyzing the depth gradient and the contour of hand region. The finger motion model with chain reaction is also constructed to predict the hand pose transition feasibly.

Hand detection is the first step of any hand biometric recognition process, which determines the outcome of the following treatments. In this paper, we propose a robust method for hand detection without contact and without constraints on the capture environment for hand biometric applications. This method is based on a color based approach adopting a non-parametric modeling. Our contribution consists mainly in the choice of the most relevant color axes and the choice of the decision rules automatically using a process of the data-mining as a new philosophy of data processing. To improve the achieved results of skin detection and to determine the hand region in the image, a succession of post-processing was proposed. Our hand detection method was evaluated experimentally on a real database; the outcomes of this evaluation show promising results and demonstrate the effectiveness of the proposed method.

This paper introduces a novel vision device for hand gesture-based human-robot interaction. Taking advantage of an eye-in-hand configuration, the proposed device guarantees that a target hand is projected always onto images captured from a camera. On the basis of the device, an adaptive histogram-based hand region segmentation algorithm is also proposed especially to enhance the performance against background and illumination changes. The proposed method is evaluated through several experiments using test images.

This paper proposes an automatic method to recognize the hand in complex background. With this method, contour of hand in pictures can be figured out and the fingertips can be marked out. Skin color detection is used to find out the hand region out of a complex background, then a serial of morphological operations is performed to get the contour of the hand, and the computation of Fingertip-Angle is for marking out the fingertips. The simulation result shows that out method has a good performance for images with low computation cost.

The aim of this paper is to present the methodology for hand detection and propose the finger detection method. The detected hand and finger can be used to implement the non-contact mouse. This technology can be used to control the home devices such as curtain and television. Skin color is used to segment the hand region from background and counter is extracted from the segmented hand. Analysis of counter gives us the location of finger tip in the hand. We have performed extensive experiment and achieve very encouraging result.

In order to realize high-precision hand gesture recognition, it is necessary that hand region is extracted correctly from moving image. In this paper, we propose a new hand region extraction method by using background subtraction algorithm. In the proposed method, we introduce the renewal of background image according to the situation of input image. In comparison with thc conventional method, our method is able to extract hand region more accurately and more efficiently. Moreover, the proposed method has been applied to hand region extraction for Japanese sign language, and its validity has been confirmed.