Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Human Interactive System
Transcript of Human Interactive System
your world.. Let your gestures
take you to the new world..! GESTURE -> "Motion of the body that contains any information." Our proposed approach towards this vision based hand interaction is based upon the Augmented Reality. Submitted by,
Sruthy M (62)
Vasudev (69) INTERFACE INTERACTION INTRODUCTION And hence provide you with high degrees of freedom. Algorithms Used: 1) Convex Hull Algorithm
2) Camshift Algorithm
Thank You..! LITERATURE SURVEY 2. Colour Marker Based HCI. 3. Using Haar Cascade Gestures in Colour Marker Using Haar Touch Screen Devices Advantage:Responsive Disadvantage: Requires a touch screen Based HCI Advantage:Easy to detect Disadvantage: Requires color markers Cascade :using shape models Advantage:Maximum detection Disadvantage: Difficult to train & limited hand gestures a) a camera; b) a projector; c) a mobile computing device like a laptop. Camera Computing Device Hand Detection Cam shift Tracking Fingertip Detection Gesture Recognition Pocket Projector Projected Information HAND SEGMENTATION -> Skin colour mainly used because of its unique features. -> Concentrated skin colour reduces skin colour overlap => easier detection of hand in complex background. -> YCrCb colour shows more concentrated skin colour distribution than RGB & YUV colour spaces and is modeled with Single Gaussian Model or Elliptical Boundary Models. Next step is
the Hand. HAND TRACKING Camshift algorithm is used to track the hand in future frames where the bounding box of the hand contour is selected as the Region Of Interest(ROI) The captured image is converted into its corresponding Hue Saturation Value(HSV) and a histogram is created for the The camshift calculates the angle & size of hand bounded in image. the rectangle to calculate the relative change in hand size to find the corresponding movement of hand to & fro from the camera. FINGER TIP DETECTION 1) Curvature Based: 2) Convexity Defect Based: a) finger tips are separated from the valleys.
b) center of mass of the hand is computed. a) convex hull of the contour of the hand is computed.
b) convexity defects are then refined to get one defect for each finger tip. GESTURE a) Uni-stroke gestures are easy to draw & hence are much simpler & faster. b) Capturing of the gesture starts only when the opens his thumb & fore finger. user c) Pointer position is captured after every 20ms & stored as array of pointers. d) The spacing between the captured points vary with the speed at which the gesture is drawn. e) Number of points in the captured gesture is calculated & this gesture length is used in the normalizing process of space. APPLICATIONS CONCLUSION a) Control the mouse. b) Generate mouse clicks. c) Enable drag & drop. d) Capture photos using the webcam attached. e) Zoom & pan photos using 'pinch' gesture. a) The project mainly serves as a helping hand for the disabled.
b) Scope of implementing additional applications is very high.
c) User can access & use his system even away from the limit of the wired devices, provided he has a good resolution camera & projector. What more..! Now that's what we call, "the world at your finger tips..!" RECOGNITION