Feature selection using cuckoo search algorithm for object classification


  • Nawal Murad Khatur
  • Abdulameer Abdullah Karim


In any object  classification system, there is a need to extract features and use them  to classify the objects,  most of the extracted features have   succeed  to classify some object but failed to classify others.  Feature selection is a general problem used   for dimensionality reduction purposes. Feature selection  aims to select important   features. The major objective of the this paper is to use the Binary Cuckoo Search(BCS) to select important features from the set of the extracted features. The feature extracted were grey level features ( texture features ), the size features, shape features  and the optimal  feature has been selected by BCS. Support vector machine (SVM) classifier  used in BCS as a fitness function. The cuckoo search algorithm select only seven feature from (25) features  where performed average accuracy 92% and improve the classification time  from ( 1.55  ) second at average to ( 57 ) millisecond at average.




How to Cite

Murad Khatur, N. ., & Abdullah Karim, A. . (2019). Feature selection using cuckoo search algorithm for object classification. Mustansiriyah Journal for Sciences and Education, 20(5), 35-46. Retrieved from https://edumag.uomustansiriyah.edu.iq/index.php/mjse/article/view/671



Research Article