CODEWITHSUNDEEP
pythonalgorithmpython-2.7convex-hull
sid597
sid597

Reputation: 1019

Maximum distance between points in a convex hull

I am solving a problem in which I need to find the maximum distance between two points on a plane (2D) .So there is an O(n^2) approach in which I calculate distance between every point in the graph . I also implemented a convex hull algorithm now my approach is I compute convex hull in O(nlogn) and then use the O(n^2) algorithm to compute maximum distance between points in the convex hull. Is there a better approach than this to compute the max distance in convex hull

Here are my algorithm :

O(n^2)

 def d(l1,l2):
    return ((l2[0]-l1[0])**2+(l2[1]-l1[1])**2)
def find_max_dist(L):
    max_dist = d(L[0], L[1])
    for i in range(0, len(L)-1):
        for j in range(i+1, len(L)):
            max_dist = max(d(L[i], L[j]), max_dist)
    return max_dist

convex hull

def convex_hull(points):
    """Computes the convex hull of a set of 2D points.

       Input: an iterable sequence of (x, y) pairs representing the points.
       Output: a list of vertices of the convex hull in counter-clockwise order,
       starting from the vertex with the lexicographically smallest coordinates.
       Implements Andrew's monotone chain algorithm. O(n log n) complexity.
"""

      # Sort the points lexicographically (tuples are compared lexicographically).
      # Remove duplicates to detect the case we have just one unique point.
        points = sorted(set(points))

      # Boring case: no points or a single point, possibly repeated multiple times.
    if len(points) <= 1:
        return points

    # 2D cross product of OA and OB vectors, i.e. z-component of their 3D cross product.
    # Returns a positive value, if OAB makes a counter-clockwise turn,
    # negative for clockwise turn, and zero if the points are collinear.
    def cross(o, a, b):
        return (a[0] - o[0]) * (b[1] - o[1]) - (a[1] - o[1]) * (b[0] - o[0])

    # Build lower hull
    lower = []
    for p in points:
        while len(lower) >= 2 and cross(lower[-2], lower[-1], p) <= 0:
            lower.pop()
        lower.append(p)

    # Build upper hull
    upper = []
    for p in reversed(points):
        while len(upper) >= 2 and cross(upper[-2], upper[-1], p) <= 0:
            upper.pop()
        upper.append(p)

    # Concatenation of the lower and upper hulls gives the convex hull.
    # Last point of each list is omitted because it is repeated at the beginning of the other list.
    return lower[:-1] + upper[:-1]

overall algorithm

 l=[]
 for i in xrange(int(raw_input())):   # takes input denoting number  of points in the plane
     n=tuple(int(i) for i in raw_input().split())  #takes each point and makes a tuple
     l.append(n)                                # appends to n

 if len(l)>=10:
        print find_max_dist(convex_hull(l))
 else:
        print find_max_dist(l)

Now how do I improve the running time of my approach and is there a better way to compute this ?

Upvotes: 4

Views: 4004

Answers (1)

kraskevich
kraskevich

Reputation: 18546

Once you have a convex hull, you can find two furthest points in linear time.

The idea is to keep two pointers: one of them points to the current edge (and is always incremented by one) and the other one points to a vertex.

The answer is the maximum distance between end points of an edge and the vertex for all edges.

It is possible to show (the proof is neither short nor trivial, so I will not post it here) that if we keep incrementing the second pointer every time after moving the first one as long as it increases the distance between the line that goes through the edge and a vertex, we will find the optimal answer.

Upvotes: 1

Related Questions