Dissertation > Excellent graduate degree dissertation topics show

Research on Super Resolution Image Restoration for Face Recognition at a Distance

Author: LiuXiaoZuo
Tutor: WanBaiKun
School: Tianjin University
Course: Biomedical Engineering
Keywords: face recognition at a long distance high resolution image restoration image processing in frequency domain simplified AC coefficients inference model
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 71
Quote: 0
Read: Download Dissertation

Abstract


In Biometric Identification, face recognition has become one of the main technologies for the advantages of natural interaction and non-intrusive detection. However, due to the cluttered background and long distance between camera and faces, normally the size of the side face region is small, the resolution is low and the illumination changes unpredictably. To deal with the problems above, this thesis introduces a method which restores the original low resolution image (LRI) from video to produce the high resolution image (HRI) and use the HRI in the face recognition.First of all, the problem about modeling of degraded of image was analyzed and corresponding degradation model was built. All the works is solved in Discrete Cosine Transform domain, the desired DC coefficients of objective high resolution image are estimated by Cubic B-Spline. Relatively, a simplified inference model was introduced to infer the AC coefficients. The main work is as follows:1)Did DCT transform of the LRI, chose different numbers of DC coefficients and AC coefficients and did IDCT transform. Chose different numbers of AC coefficients and selected their Peak Signal to Noise Ratio(PSNR) to represent the results of restoration. found out that the first 15 AC coefficients induced the best restoration result;2) the Markov model was simplified, used Locally Linear Embedding(LLE) to reduce the coefficients’dimension and collected minimum of the energy function to calculate AC coefficients of the objective high resolution image (HRI).Plus the images based on the above AC and DC coefficients to receive image blocks. After post-filtering, the HRI was gained. Tested the face images captured from gait video at a distance, recognized the HRI of the faces using Adaboost arithmetic. After restoration of the LRI, the fall-out ratio and omission factor have remarkable decrease, the recognition results are improved correspondingly. Compared with C.Liu methods in restoration, this method has advantages on restoring detail information. The result will help to improve the face recognition with the cluttered background at a distance. Moreover, it gives support to the research of combination between face recognition and gait recognition in identification.

Related Dissertations

  1. Research on Basic Algorithms of Digital Image Processing and Implementation with FPGA,TP391.41
  2. Research on Facial Feature Extraction and Matching Algorithms for Image Retrieval,TP391.41
  3. Research of High Speed Image Pre-processing System Based on FPGA,TP391.41
  4. Research on Algorithms of 2D Face Template Protection,TP391.41
  5. The Research and Implemention of Image Retrieval Based on User Interested Feature,TP391.41
  6. Research of Image Mosaic Technology,TP391.41
  7. Research and Implementation of Exact String Matchiing Algorithms,TP391.41
  8. Research on the Classification Based on the Reconstruction of Solder Joint,TP391.41
  9. Tongue Feature Extraction and Research of Fusion Classification,TP391.41
  10. The Fatigue State Recognition of the Driver Based on Eye Detection,TP391.41
  11. Research on Infrared Image Simulation for Aerial Objects and Background,TP391.41
  12. Research on Intelligent Learning-Based Multi-Sensor Target Recognition and Tracking System,TP391.41
  13. Research on Image Compression and Implementation Using TMS320C6713 Based on SPIHT Algorithm,TP391.41
  14. Research on Joint Target Detection for Dual-Sensor Image and System Implementation,TP391.41
  15. Research of Images Enhancing Algorithms on Fog or Backlighting Conditions and Implementation with Hardwares,TP391.41
  16. Research on Feature Extraction and Classification of Pulse Waveform for Cholecystitis and Nephrotic Syndrome Diagnosis,TP391.41
  17. Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
  18. Research and Implementation on Content-Based Clothing Image Retrieval,TP391.41
  19. The Discovery of User Concept Region Based on Multiple Instance Learning,TP391.41
  20. Research on Sparse Image Representation and Coding Model,TP391.41
  21. Simulation and Experimental Study on the Scattering Properties of Two Dimensional Rough Surfaces,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