Sara Mohammed Osman Saleh Bilal

Hidden Markov model for human to computer interaction: a study on human hand gesture recognition

Human hand recognition plays an important role in a wide range of applications ranging from sign language translators, gesture recognition, augmented reality, surveillance and medical image processing to various Human Computer Interaction (HCI) domains. Human hand is a complex articulated object consisting of many connected parts and joints. Therefore, for applications that involve HCI one can find many challenges to establish a system with high detection and recognition accuracy for hand posture and/or gesture. Hand posture is defined as a static hand configuration without any movement involved. Meanwhile, hand gesture is a sequence of hand postures connected by continuous motions. During the past decades, many approaches have been presented for hand posture and/or gesture recognition. In this paper, we provide a survey on approaches which are based on Hidden Markov Models (HMM) for hand posture and gesture recognition for HCI applications.

Vision-based hand posture detection and recognition for Sign Language — A study

Unlike general gestures, Sign Languages (SLs) are highly structured so that it provides an appealing test bed for understanding more general principles for hand shape, location and motion trajectory. Hand posture shape in other words static gestures detection and recognition is crucial in SLs and plays an important role within the duration of the motion trajectory. Vision-based hand shape recognition can be accomplished using three approaches 3D hand modelling, appearance-based methods and hand shape analysis. In this survey paper, we show that extracting features from hand shape is so essential during recognition stage for applications such as SL translators.

A hybrid method using haar-like and skin-color algorithm for hand posture detection, recognition and tracking

Human hand posture detection and recognition is a challenging problem in computer vision. We introduce an algorithm that is capable to recognize hand posture in a sophisticated background. The system combines two algorithms to achieve better detection rate for hand. Recently Viola et al. in [10] have introduced a rapid object detection scheme; we use this approach to detect the hand posture in the first set of consecutive frames. The chromatic color distribution of skin can be found within this cluster. As the shape of hand posture keep changing in the subsequent frames, the skin regions updated dynamically. The classification of hand posture makes use of static feature for locating and counting hand fingers. Kalman Filter is used to track the face and hand blobs based on their position. In the experiments, we have tested our system in various environments, and results showed effectiveness of the approach.

Dynamic approach for real-time skin detection

Human face and hand detection, recognition and tracking are important research areas for many computer interaction applications. Face and hand are considered as human skin blobs, which fall in a compact region of colour spaces. Limitations arise from the fact that human skin has common properties and can be defined in various colour spaces after applying colour normalization. The model therefore, has to accept a wide range of colours, making it more susceptible to noise. We have addressed this problem and propose that the skin colour could be defined separately for every person. This is expected to reduce the errors. To detect human skin colour pixels and to decrease the number of false alarms, a prior face or hand detection model has been developed using Haar-like and AdaBoost technique. To decrease the cost of computational time, a fast search algorithm for skin detection is proposed. The level of performance reached in terms of detection accuracy and processing time allows this approach to be an adequate choice for real-time skin blob tracking.

Malaysian Sign Language database for research

Malaysian Sign Language (MSL) is the main language that is commonly used by the hearing and speech impaired person in Malaysian. The SL (SL) involves hand movement, and hand gestures. In order to help people who are not familiar, but need to understand a particular SL, an automatic SL recognition system is highly required. The research in this area, especially for MSL, has been conducted by many researchers, but one of the main challenges in this research is the availability of suitable sign database for the recognition. The existing databases, especially which of MSL database, are provided often without a proper standard of image resolution, structure and compression that are sufficiently good for research purpose. To provide comprehensive information for the research on MSL, the MSL database is highly required. In this project, a MSL database is developed. The database is the first of the kind and developed for research purpose. In general, the structure of the MSL database is classified into groups that deal with the hand movement, hand gestures, and hand location. For the classification in our proposed database, the MSL is classified into One Hand, Two Hands, Static, and Dynamic. This classification is made to ease researchers in defining the research method for each type offhand signing.

Modeling of Human Upper Body for Sign Language Recognition

Sign Language Recognition systems require not only the hand motion trajectory to be classified but also facial features, Human Upper Body (HUB) and hand position with respect to other HUB parts. Head, face, forehead, shoulders and chest are very crucial parts that can carry a lot of positioning information of hand gestures in gesture classification. In this paper as the main contribution, a fast and robust search algorithm for HUB parts based on head size has been introduced for real time implementations. Scaling the extracted parts during body orientation was attained using partial estimation of face size. Tracking the extracted parts for front and side view was achieved using CAMSHIFT [24]. The outcome of the system makes it applicable for real-time applications such as Sign Languages Recognition (SLR) systems.

VMware as an Intermediate Platform between Windows 7 and Linux Fedora 15 for Real-Time Malaysian Sign Language Translator

This research introduces a virtual interface between Windows 7 and LINUX Fedora 16 based on Virtual Machine ware (VMware) for real-time Malaysian Sign Language (MSL) translation into text and/or voice (in English). The developed method is based on HTK, Gt2k under LINUX Fedora 16 and VC++ 2010 and OpenCV pre 1.1 library under Windows7. The communication between client (Windows7) and server (LINUX Fedora 16) has been established using VMware. The main significance of this approach is that the best characteristics of both operating systems LINUX Fedora 16 and Windows7 have been utilized. Under Windows7, Visual C++ 2010 combined with OpenCV pre 1.1 library supports video processing algorithms and has a power graphical user interface. Meanwhile, the Gt2k for gesture recognition is fully supported under LINUX. Therefore, a client/server technology has secured much time during the MSL recognition system development and helped in terms of algorithms enhancement.

Human Upper Body Pose Region Estimation

The objective of this chapter is to estimate 2D human pose for action recognition and especially for sign language recognition systems which require not only the hand motion trajectory to be classified but also facial features, Human Upper Body (HUB) and hand position with respect to other HUB parts. We propose an approach that progressively reduces the search space for body parts and can greatly improve chance to estimate the HUB pose. This involves two contributions: (a) a fast and robust search algorithm for HUB parts based on head size has been introduced for real time implementations. (b) Scaling the extracted parts during body orientation was attained using partial estimation of face size. The outcome of the system makes it applicable for real-time applications such as sign languages recognition systems. The method is fully automatic and self-initializing using a Haar-like face region. The tracking the HUB pose is based on the face detection algorithm. Our evaluation was done mainly using 50 images from INRIA Person Dataset.