An　automatic　video　object　segmentation　algorithm　based　on　background　reconstruction　is　proposed　in　this　paper.　First,　HOS　(Higher　Order　Statistics)　is　used　to　automatically　separate　the　moving　areas　from　background.　The　initial　object　mask　is　extracted,　and　the　corresponding　initial　background　is　also　derived.　Second,　the　background　is　reconstructed　using　forecasting　theory.　Finally,　the　video　object　is　extracted　by　combining　the　frame　difference　mask　and　background　subtraction　mask.　Furthermore,　morphological　reconstruction　algorithm　is　also　used　to　eliminate　shadow.　©　2011　IEEE.