@Article{JICS-14-195, author = {Jingji Zhao}, title = {Image Retrieval Method based on Integration of Principal Component Analysis and Multiple Features}, journal = {Journal of Information and Computing Science}, year = {2019}, volume = {14}, number = {3}, pages = {195--202}, abstract = {Jingji Zhao School of Mathematics and Statistics, Nanjing University of Information Science & Technology,   Nanjing, 210044, China (Received May 11 2019, accepted July 20 2019) Existing  content-based  image  retrieval  methods  exist  some  drawbacks,  such  as  low  retrieval precision,  unstable  performance.  To  address  these  drawbacks,  in  this  paper  a  content-based  image  retrieval method  is  presented  based  on  multi-feature  fusion  of  principal  component,  oriented-gradient  and  color histogram.  The  idea  for  the  proposed  method  is:  firstly,  input  image  is  grayscale  and  flattened  into  a  one- dimensional vector, and the first n principal  components from the vector yielded by the  PCA algorithm  are extracted, in other word, input image is represented as  a n×1 dimensional PCA  feature  vector. Secondly, to remedy color and orientation information missed by PCA, oriented-gradient and color histograms are used to extract orientation and  color features  respectively. Thirdly, extracted oriented-gradient  and color histograms are  merged  with  PCA  features  to  generate  the  multi-feature  representation  of  the  input  image.  This  paper confirms that the proposed multi-feature method can better represent an input image and can easily measure the similarity between images. The experiments are carried out and evaluated based on Corel-1000 , the target method is significantly better than the four popular methods. }, issn = {3080-180X}, doi = {https://doi.org/}, url = {http://global-sci.org/intro/article_detail/jics/22413.html} }