Dissertation > Excellent graduate degree dissertation topics show

Obstacle detection based on semi - supervised learning

Author: WangPing
Tutor: LuJianFeng
School: Nanjing University of Technology and Engineering
Course: Applied Computer Technology
Keywords: Obstacle detection Semi-supervised learning Self-training Coordinated training k-nearest neighbor
CLC: TP242.6
Type: Master's thesis
Year: 2009
Downloads: 155
Quote: 2
Read: Download Dissertation

Abstract


Obstacle detection is an important content of the mobile robot environment sensing technology . The conventional obstacle detection method requires manual adjustment of parameters. The introduction of machine learning for obstacle detection can reduce labor participation and improve the degree of intelligence of the mobile robot can also better handling high-dimensional input information . Traditional supervised learning algorithms only use has been labeled samples to learn complex and changing environment of the mobile robot , road conditions and obstacles vagaries artificial marker to detect obstacles as accurately as possible , the need for a large number of samples to establish a complete set of training samples , and very time-consuming and labor-intensive . Above , the introduction of semi-supervised learning method , the application of self- training and collaborative training , the first use of a small amount of labeled samples to establish the initial classification , then the use of a large number of unlabeled samples constantly update the classifier, and to improve the performance of the classifier . In this paper, context-aware image data of a large number of experiments : the use of partitioning strategies , respectively color feature extraction and texture feature extraction, a small amount of labeled samples and unlabeled samples , and then in the k-nearest neighbor algorithm based on the introduction of semi- supervision of learning self - training and collaborative training for obstacle detection . Experimental results show that the introduction of the semi-supervised learning , in only a small amount of labeled samples circumstances , can achieve a higher rate of obstacle detection and correct classification rate .

Related Dissertations

  1. Research on Improved K Neighbor Support Vector Machine Algorithm Faced Text Classification,TP391.1
  2. Semi-supervised Learning and Active Learning of Sentiment Classification Coupled with Domain Knowledge,TP181
  3. Membrane protein transmembrane helix prediction,Q51
  4. Semi-supervised hashing algorithm based Image Retrieval Methods,TP391.41
  5. Based on Information Fusion roads and obstacle detection method,TP242
  6. Rural road environment obstacle detection,TP391.41
  7. Research and Implementation of Time Series Classification Based on Semi-supervised Learning,TP181
  8. Research and Application on Design Resource Dynamic Scheduling Faced to Design Reuse,TB47
  9. K nearest neighbor classification based on improved IT asset management system development,TP315
  10. SVM-based target tracking algorithm,TP391.41
  11. Research on the Application of Data Mining Technology in the Regulating Matriculation for Postgraduate,TP311.13
  12. Comparison and Improvement of Two Methods Based on Semi-Supervised Learning,TP18
  13. Obstacle Detection Method Based on Single Camera,TP242
  14. Research on Intrusion Detection Based on Semi-Supervised SVM,TP393.08
  15. Research of Face Recognition Technology,TP391.41
  16. The LVCSR system based on adaptive methods of semi-supervised learning,TN912.34
  17. Design and Implementation of Hydropower Fault Classifer Based on Support Vector Machine,TV738
  18. A Study on Some Problems of Semi-supervised Learning,TP181
  19. Research on the Technology of Face Recognition Based-on Locality Preserving Projection,TP391.41
  20. Based semi - supervised learning Chinese Question Classification,TP391.1
  21. The Design and Implementation of an Image Segmentation System Based on Semi-Supervised Learning,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Robotics > Robot > Intelligent robots
© 2012 www.DissertationTopic.Net  Mobile