Dissertation > Excellent graduate degree dissertation topics show

Monocular visual servo gesture recognition and human-computer interaction research

Author: ZhangBaoYun
Tutor: ZuoRuWei
School: Nanjing Normal University
Course: Educational Technology
Keywords: Computer Vision Human-Computer Interaction Gesture recognition Skin color segmentation Distance distribution characteristics
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 95
Quote: 0
Read: Download Dissertation

Abstract


As a new generation of human-computer interaction, monocular gesture recognition based on computer vision (Computer Vision) has important theoretical value and practical application prospects. In this paper, the main flow of monocular vision, gesture recognition gesture image preprocessing, feature extraction, gesture recognition, and Microsoft Visual Studio 2008 environment input from ordinary camera 5 common simple gesture identified and developed as the basis for a simple gesture interaction system to verify the feasibility of the algorithm. Finally gesture interaction system developed on the basis of a practical application of the value of teaching prototype system, monocular gesture recognition has practical significance. First, collected from the camera to obtain the video stream gesture image to build gesture model library, after pretreatment. Gesture on artificial markers in the the gesture image preprocessing stage this article, which can get good segmentation results in a more complex context. In the the gesture image feature extraction stage, gesture binary image from the distribution characteristics as the gesture recognition robustness characterized by the distance distribution characteristics to describe the shape characteristics of the image of the gesture may be described in the general shape of the gesture information. At the same time using this method for gesture recognition also has translation invariance and scale invariance, select the image center of gravity as a reference point, and all the images of the target area are divided into the same number from the area, so the gesture target in the field of vision minimal impact on the position and scale the size of the distance distribution characteristics. Using this method to identify, the gesture feature does not change with rotation and translation, the variation of the size varies, the gesture recognition robustness. This study mainly around based on the monocular visual robust gesture recognition method and its application in the human-computer interaction, originality reflected in three aspects: First, the proposed pixel distance distribution characteristics extraction gesture area Lu stick characteristics of the the rotation invariant natural state gesture recognition. Compared with the general method, the use of this technique to identify the gesture is characterized by rotational invariance, real-time gesture recognition can get rid of the interference in the rotated image, enhance the robust recognition. Second, this paper, using the method of adding artificial markers, remove the skin color regions interference regional issues, with originality. Pixel distance distribution characteristics gesture recognition, color segmentation accuracy requirements. In the case of complex background, if there is a certain number of skin color region interference, gestures from the distribution characteristics will change, which will directly affect the results of the identification, this article by adding artificial markers method succeeded in solving this problem. Third, the development of gesture recognition-based human-computer interaction system prototype has a simple teaching function, to expand the field of application of the visual gesture-based human-computer interaction.

Related Dissertations

  1. Research on Learning-Based Low-Level Vision Problem,TP391.41
  2. Research on Nondestructive Detection Technology for External Qualities of Papayas Based-on Vision,S667.9
  3. Study on Detection and Grading of ’Jiro’ Persimmon’s External Quality Based on Computer Vision,S665.2
  4. Research on Hand Tracking and Application Platform for Hand Gesture Recognition,TP391.4
  5. Prediction of Pork Processing Functionalities,TS251.1
  6. Research on Inspection Technology of Dehydrated Garlic Slice Based on Computer Vision,TP391.41
  7. Moving target trajectory analysis based Intelligent Traffic Monitoring System,TP277
  8. Study and Implementation on Human-computer Interaction Based on EoC System of NGB,TP11
  9. Depth map and color images based on treadmill game interaction system,TP391.41
  10. Based on mobile interactive touch screen display technology research and implementation,TP334.3
  11. Images and image sequences on interactive matting technology research,TP391.41
  12. Image restoration of key technologies,TP391.41
  13. Research on Key Technologies of Monocular Visual Odometry Based on DSP,TP391.41
  14. Research to Line Feature Based Relational Matching of Organized Object’s Imagery,TP391.41
  15. Research on the Interactive Design of City Tourist Electronic Map,P208
  16. Study on Maintenance Training Technology of Hydro Turbine-Generator Sets Based on Virtual Reality,TV738
  17. Car dashboard computer vision detection system,TP391.41
  18. Research on the Technology of Moving Vehicle Detecting and Tracking in Video Image,TP391.41
  19. Study of Precision Measurement Technology Based on Optical Images,TP391.41
  20. The Design and Implementation of Golf Swing Coach System Based on OpenCV,TP391.41
  21. Research on Self-Occlusion Detection and Next Best View of Vision Target,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile