Rezumat articol ediţie STUDIA UNIVERSITATIS BABEŞ-BOLYAI

În partea de jos este prezentat rezumatul articolului selectat. Pentru revenire la cuprinsul ediţiei din care face parte acest articol, se accesează linkul din titlu. Pentru vizualizarea tuturor articolelor din arhivă la care este autor/coautor unul din autorii de mai jos, se accesează linkul din numele autorului.

 
       
         
    STUDIA INFORMATICA - Ediţia nr.2 din 2016  
         
  Articol:   SUPPORT VECTOR MACHINE AND BOOSTING BASED MULTICLASS CLASSIFICATION FOR TRAFFIC SCENE OBSTACLES.

Autori:  LAURA DIOŞAN.
 
       
         
  Rezumat:  
VIEW PDF: SUPPORT VECTOR MACHINE AND BOOSTING BASED MULTICLASS CLASSIFICATION FOR TRAFFIC SCENE OBSTACLES

Multiclass classification is an extensively researched topic due to its importance in making the binary classification problems a complex and well tuned system and minimising the running time for multiple classification problems. In the traffic scenes one can encounter several types of obstacles like cars, pedestrians, animals, low elevated objects, road signs that must be detected and categorised for safety reasons regarding the driver and traffic. The purpose of this paper is two-folds: to accurately classify four obstacle types (pedestrians, cars, animals and other types of objects) and to compare some multiclass classification methods based on Support Vector Machine and Boosting algorithms. The experiments showed that the method Fuzzy Clustering with improvements using Particle Swarm Optimisation achieves great results compared to the traditional hierarchical multiclass classification and the proposed hybrid approach that combines Boosting and Support Vector Machine increases the classification accuracy even further.

2010 Mathematics Subject Classification. 68T05, 91E45.

1998 CR Categories and Descriptors. I.2.6 [Learning]: {Induction, I.2.6 [Learning]: {Concept learning.

Key words and phrases. Multiclass classification, HOG, Decision Trees, Boosting.
 
         
     
         
         
      Revenire la pagina precedentă