Investigates Image data fusion techniques that combine image and track data from multiple sensors to achieve improved accuracies and more specific inferences than could be achieved by the use of a single sensor alone. Our aim is to explore the state-of-the-art image processing algorithms for achieving effective data fusion as in:
Hand Mouse Based Video Wall Control Interface
Contents
1. Introduction
2. Main Algorithm and Principle
3. Demo
1. Introduction
The goal of Hand Mouse Based Video Wall Control Interface is to control video wall for surveillance system more efficiently and intuitively by replacing functionalities of mouse and parts of keyboard to some simple hand gestures. This process is divided into hand mouse mode and hand gesture mode. And the recognition algorithm follows rule based method.
2. Main Algorithm and Principle
i. Get palm and each finger’s coordinates and velocities using Leap motion sensor.
ii. Label the fingers according to their relative length and location.
iii. Make a rule based gesture recognizer which alternates mouse and some keyboard functions
iv. If the input hand movement satisfies certain rule, then recognize the gesture and implement corresponding function and offer visual feedback to user for the recognition result.
Fig 1. Diagram for hand mouse mode
Fig 2. Diagram for hand gesture mode
Fig 3. Click command recognition method
Fig 4. Command list
3. Demo
