Given a m x n
binary matrix mat
. In one step, you can choose one cell and flip it and all the four neighbours of it if they exist (Flip is changing 1 to 0 and 0 to 1). A pair of cells are called neighboors if they share one edge.
Return the minimum number of steps required to convert mat
to a zero matrix or -1 if you cannot.
Binary matrix is a matrix with all cells equal to 0 or 1 only.
Zero matrix is a matrix with all cells equal to 0.
Example 1:
Input: mat = [[0,0],[0,1]] Output: 3 Explanation: One possible solution is to flip (1, 0) then (0, 1) and finally (1, 1) as shown.
Example 2:
Input: mat = [[0]] Output: 0 Explanation: Given matrix is a zero matrix. We don't need to change it.
Example 3:
Input: mat = [[1,1,1],[1,0,1],[0,0,0]] Output: 6
Example 4:
Input: mat = [[1,0,0],[1,0,0]] Output: -1 Explanation: Given matrix can't be a zero matrix
Constraints:
m == mat.length
n == mat[0].length
1 <= m <= 3
1 <= n <= 3
mat[i][j]
is 0 or 1.
Related Topics:
Breadth-first Search
// OJ: https://leetcode.com/problems/minimum-number-of-flips-to-convert-binary-matrix-to-zero-matrix/
// Author: github.com/lzl124631x
// Time: O(MN * 2^(MN))
// Space: O(2^(MN))
class Solution {
public:
int minFlips(vector<vector<int>>& A) {
int start = 0, M = A.size(), N = A[0].size(), step = 0, dirs[5] = {1,0,-1,0,1};
for (int i = 0; i < M; ++i) {
for (int j = 0; j < N; ++j) {
start |= (A[i][j] << (i * 3 + j));
}
}
queue<int> q;
unordered_set<int> s;
q.push(start);
s.insert(start);
while (q.size()) {
int cnt = q.size();
while (cnt--) {
int state = q.front();
q.pop();
if (state == 0) return step;
for (int i = 0; i < 9; ++i) {
int next = state, r = i / 3, c = i % 3;
next ^= (1 << (r * 3 + c));
for (int j = 0; j < 4; ++j) {
int x = r + dirs[j], y = c + dirs[j + 1];
if (x < 0 || x >= M || y < 0 || y >= N) continue;
next ^= (1 << (x * 3 + y));
}
if (s.count(next) == 0) {
q.push(next);
s.insert(next);
}
}
}
++step;
}
return -1;
}
};