原题下载
答案:
(Analysis by Nathan Pinsker)
This problem suggests a DP approach, and the bounds suggest that O(n3)O(n3) is a comfortable target to aim for. There are several ways to approach this problem, but all of these focus on fixing some property of the rectangle, iterating over all possible values of that property, and finding the best rectangle for each iteration. One of the simpler ways is to fix an edge of the rectangle, and calculate the maximum possible rectangle with that edge.
For simplicity, we will fix the left edge of the rectangle. There are O(n2⋅n)=O(n3)O(n2⋅n)=O(n3) possible left edges, so we will reach our desired runtime if we can compute the best possible rectangle with this edge in O(1)O(1) time. Luckily, we can use a sliding window to do so: if we process the left edges in order from left to right, we can note that the right edge of the rectangle must also move to the right over time. This is because a candidate rectangle will never become invalid as we move its left edge more to the right. We can use a simple data structure to query subrectangles of our given array in O(1)O(1) time (see the code below if you're not quite sure how).
If you're interested, another approach is described in the analysis of a previous problem, Figure Eight. (The problems are very similar, and that approach can be readily adapted to this problem as well.)
Below is Nick Wu's code:
import java.io.*; import java.util.*; public class fortmoo { public static void main(String[] args) throws IOException { BufferedReader br = new BufferedReader(new FileReader("fortmoo.in")); PrintWriter pw = new PrintWriter(new BufferedWriter(new FileWriter("fortmoo.out"))); StringTokenizer st = new StringTokenizer(br.readLine()); int n = Integer.parseInt(st.nextToken()); int m = Integer.parseInt(st.nextToken()); int[][] grid = new int[n+1][m+1]; for(int i = 0; i < n; i++) { String s = br.readLine(); for(int j = 0; j < m; j++) { grid[i+1][j+1] = grid[i][j+1] + grid[i+1][j] - grid[i][j]; if(s.charAt(j) == 'X') { grid[i+1][j+1]++; } } } long ret = 0; for(int j = 0; j < m; j++) { for(int k = j; k < m; k++) { int lastRow = -1; for(int i = 0; i < n; i++) { boolean validRow = num(grid, i, j, i, k) == 0; if(validRow) { ret = Math.max(ret, k-j+1); } if(validRow && lastRow >= 0) { ret = Math.max(ret, (i - lastRow+1) * (k-j + 1)); } if(num(grid, i, k, i, k) > 0 || num(grid, i, j, i, j) > 0) { lastRow = -1; } if(validRow && lastRow == -1) { lastRow = i; } } } } pw.println(ret); pw.close(); } public static int num(int[][] grid, int a, int b, int c, int d) { return grid[c+1][d+1] - grid[a][d+1] - grid[c+1][b] + grid[a][b]; } }
© 2024. All Rights Reserved. 沪ICP备2023009024号-1