Cities in Alberta tend to be laid out as rectangular grids of blocks. Blocks are labeled with coordinates to from north to south and to from west to east.

The quality of living in each particular block has been ranked by a distinct number, called quality rank, between and , where is the best and is the worst.

The city planning department wishes to identify a rectangular set of blocks with dimensions from north to south and from west to east, such that the median quality rank among all blocks in the rectangle is the best. and are *odd* numbers not exceeding and respectively. The *median quality rank* among an odd number of quality ranks is defined to be the quality rank in the set such that the number of quality ranks better than equals the number of quality ranks worse than .

#### Input Specification

The first line of input will contain the four space-separated integers , , , and , where and represent the total size of the city, and and represent the dimensions of the set of blocks.

The next lines each contain integers, denoting an array such that is the quality rank for the block labeled from north to south and from west to east.

#### Output Specification

A single integer, the best (numerically smallest) possible median quality rank of an by rectangle of blocks.

#### Sample Input 1

```
5 5 3 3
5 11 12 16 25
17 18 2 7 10
4 23 20 3 1
24 21 19 14 9
6 22 8 13 15
```

#### Sample Output 1

`9`

#### Explanation for Sample Output 1

R=5,C=5,H=3,W=3,Q= 5 11 12 16 25 17 182 7 104 2320 3 124 2119 146 22 8 13 159

For this example, the best (numerically smallest) median quality rank of is achieved by the middle-right rectangle of shown in bold.

#### Sample Input 2

```
2 6 1 5
6 1 2 11 7 5
9 3 4 10 12 8
```

#### Sample Output 2

`5`

##### Subtask 1 [20 points]

Assume and do not exceed .

##### Subtask 2 [20 points]

Assume and do not exceed .

##### Subtask 3 [20 points]

Assume and do not exceed .

##### Subtask 4 [20 points]

Assume and do not exceed .

##### Subtask 5 [20 points]

Assume and do not exceed .

## Comments