A method called MACA (Modified Active Contour Algorithm) was 
suggested. In the standard snake model, there is no requirement  
on the matching between the orientation of the snake contour 
and the orientation of image edge. This way, the standard snake 
model usually fails to segment objects from images with complicate background. 
To avoid it, two energy functions are proposed to 
improve the snake model. To reduce the time cost in the iteration
of MACA, a FGA (Fast Greedy Algorithm) was proposed. Also, the 
first-order and second-order derivatives, the domain of searching 
area, contour resampling, and conditions of corner locking 
were redefined. Experimental results show that the performance of the proposed 
method is much better than that of the standard snake model, 
particularly in extracting concave objects.