With the widespread of the network and development of  Image technology,  the  older information  retrieval techniques  do  not  meet  today’s  demand.  Recently,  the content-based image retrieval has become the most developing  topic and the techniques of content-based image retrieval have acquired great attention.  In  this  paper,  the  basic  components  of content-based  image  retrieval  system  are discussed.  Image retrieval  methods  based  on  color,  texture,  shape  and  semantic image are studied discussed and compared. Other related techniques such as relevance feedback and performance evaluation are also discussed. In the end of paper the  problems  and  challenges  are  fetched out of each available technique. In  recent  areas  of commerce,  government,  academics,  and  hospitals,  large collections  of  digital  images  are  being  created and stored.  Large amount of the digital image collections  are  the  product  of  digitalizing  existing collections of hardcopy photographs, diagrams, drawings, paintings,  and  prints.  Usually,  the  only  way  of  searching  these  collections  was  by  keyword  indexing,  or  simply  by browsing related words. Digital image database however, open the way to content-based searching. In this paper we review some technical aspects of current content-based image retrieval systems.