Consider the graph of y = f(x). We consider to be the area between the graph and the x-axis between the lines x = a and x = b to be S(b,a) y dx.
Volume = S(b,a) y dx
Let us try to find find the value of this integral using rectangles of equal width. As we understand from the trapezium rule, the sum of such rectangles results in an approximation of the area under the curve. The larger the number of rectangles, the more accurate the approximation. If we were to find the sum of an infinite number of rectangles of infinitesimal width, this would give us the value of the integral.
The area of each rectangle is found as y.(delta)x, where each (delta)x is an increment of the value of x and is the width of each triangle. Thus, the exact area under the curve can be taken as
lim((delta)x tends to 0) ( (sum) y.(delta)x )
Integration is defined as the limit of this summation, therefore
lim((delta)x tends to 0) ( (sum) y.(delta)x ) = S(b,a) y dx
Each of these rectangles is going to be rotated by 2(pi) radians about the y-axis. This shape formed by the rotation of each rect will be approximately cylindrical. The volume of each cylinder-like shape will be
(pi).y^2.(delta)x
This is the area of the circle formed by the rotation of a point, (pi).y^2, multiplied by the width of the rectangle.
As (delta)x tends to 0 in our summation,
(sum)( (pi).y^2.(delta)x ) tends to S (pi).y^2.(delta)x
In other words,
lim((delta)x tends to 0) (sum)( (pi).y^2.(delta)x ) = S (pi).y^2.(delta)x
I can't use LaTeX, but I hope that helps to make things a bit clearer for you. It's just summation of infinitely-thin cylinders, as opposed to standard integration which gives you a summation of infinitely thin strips.