Dissertation > Excellent graduate degree dissertation topics show

Research on Face Detection Based on Pulse Coupled Neural Network

Author: ZhangCuiCui
Tutor: LiWenXing
School: Harbin Engineering University
Course: Communication and Information System
Keywords: Improved pulse coupled neural network ( MPCNN ) Image entropy Face Detection Skin color segmentation
CLC: TP391.41
Type: Master's thesis
Year: 2008
Downloads: 70
Quote: 0
Read: Download Dissertation

Abstract


The pulse coupled neural network (Pulse Coupled Neural Network, PCNN) is based on the visual cortex of animals sync pulse issuance of the experimental results obtained . Sync output pulse PCNN network model has links to domain characteristics and dynamic threshold attenuation characteristics similar state neurons , which is fully simulate the biological characteristics of the mammalian visual cortex neurons , which in image segmentation , edge detection , target identification and other image processing a wide range of applications . The The standard PCNN model improved to some extent , reduce Neuron Model some parameters to the on a keep PCNN link domain characteristics and dynamic threshold value of the attenuation characteristics of the basis put forward that is more applicable to image processing improvements on PCNN ( the Modified on PCNN - MPCNN ) model and in-depth study of its basic principles and characteristics . On the basis of MPCNN introduced image entropy segmentation algorithm based MPCNN and maximum entropy , apply it to the ORL face database , the test results show that the effectiveness of the algorithm . ORL face database images in the target and background there are significant differences in brightness , color image , in addition to the brightness of information as well as chrominance information , and the background is not uniform , easily separated target and background detection based on maximum entropy the effect is not very good, so this proposed algorithm combining based on MPCNN and color to color images for face detection , and then based on this algorithm color lena Figure , interception of Purdue - AR color video stream dynamic image with different expressions different light color image and color image segmentation, from the Internet and do a simple post-processing image segmentation . In addition, this algorithm is also based on template matching algorithm made ??a simple comparison , the experimental results show that this method has good practicability .

Related Dissertations

  1. Research on the Vision-Based Driver Fatigue Real-Time Detection,TP391.41
  2. Face Detection in Color Image and Facial Feature Location,TP391.41
  3. Real-Time Face Pose Estimation in Video Sequence,TP391.41
  4. Video target detection and tracking,TP391.41
  5. Notation of Speaking Face Based on Video and Text Infomation,TP391.41
  6. Pulse coupled neural network for image segmentation and face detection,TP391.41
  7. Research and Implementation on Face Detectionalgorithm,TP391.41
  8. Implementation of Face Detection Algorithm for AdaBoost Base on DSP,TP391.41
  9. Feature points based virtual try positioning research,TP391.41
  10. Research on Face Detection Based on Skin Color Segmentation and AdaBoost Algorithm,TP391.41
  11. Continuous video stream human face detection and recognition research,TP391.41
  12. Design and implementation of face detection and recognition system based on a static image of the person,TP391.41
  13. Face recognition authentication technology,TP391.41
  14. Face Detection and Tracking Method Research Based on Video Sequences,TP391.41
  15. The Research of Pre-processing in Video Coding of Videophone,TN919.81
  16. Face Detection Algorithm moment feature extraction and support vector machine,TP391.41
  17. Eye State Tracking Under Driving Conditions,TP391.41
  18. Research and Application of Face Detect Technology in Smartphone Terminal,TP391.41
  19. Research on the Driver Fatigue Detection System,TP391.41
  20. Driver Fatigue Recognition Based on AdaBoost Algorithm,TP391.41
  21. Research and Implementation of Intelligent Recognition Robot Based on WiFi Wireless Network,TP242.6

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