Link to this code: https://cses.fi/paste/56357dd86f619a1eda2deb/
// #pragma GCC optimize("O3")

#include <algorithm>
#include <array>
#include <bitset>
#include <cassert>
#include <chrono>
#include <climits>
#include <cmath>
#include <complex>
#include <cstring>
#include <functional>
#include <iomanip>
#include <iostream>
#include <map>
#include <numeric>
#include <queue>
#include <random>
#include <set>
#include <vector>
using namespace std;

using ll = long long;
using db = long double;  // or double, if TL is tight
using str = string;      // yay python!

// pairs
using pi = pair<int, int>;
using pl = pair<ll, ll>;
using pd = pair<db, db>;
#define mp make_pair
#define f first
#define s second

#define tcT template <class T
#define tcTU tcT, class U
// ^ lol this makes everything look weird but I'll try it
tcT > using V = vector<T>;
tcT, size_t SZ > using AR = array<T, SZ>;
using vi = V<int>;
using vb = V<bool>;
using vl = V<ll>;
using vd = V<db>;
using vs = V<str>;
using vpi = V<pi>;
using vpl = V<pl>;
using vpd = V<pd>;

// vectors
#define sz(x) int(size(x))
#define bg(x) begin(x)
#define all(x) bg(x), end(x)
#define rall(x) rbegin(x), rend(x)
#define sor(x) sort(all(x))
#define rsz resize
#define ins insert
#define pb push_back
#define eb emplace_back
#define ft front()
#define bk back()

#define lb lower_bound
#define ub upper_bound
tcT > int lwb(const V<T> &a, const T &b) { return int(lb(all(a), b) - bg(a)); }
tcT > int upb(const V<T> &a, const T &b) { return int(ub(all(a), b) - bg(a)); }

// loops
#define FOR(i, a, b) for (int i = (a); i < (b); ++i)
#define F0R(i, a) FOR(i, 0, a)
#define ROF(i, a, b) for (int i = (b) - 1; i >= (a); --i)
#define R0F(i, a) ROF(i, 0, a)
#define rep(a) F0R(_, a)
#define each(a, x) for (auto &a : x)

const int MOD = 998244353;  // 1e9+7;
const int MX = (int)2e5 + 5;
const ll BIG = 1e18;  // not too close to LLONG_MAX
const db PI = acos((db)-1);
const int dx[4]{1, 0, -1, 0}, dy[4]{0, 1, 0, -1};  // for every grid problem!!
mt19937 rng((uint32_t)chrono::steady_clock::now().time_since_epoch().count());
template <class T> using pqg = priority_queue<T, vector<T>, greater<T>>;

// bitwise ops
// also see https://gcc.gnu.org/onlinedocs/gcc/Other-Builtins.html
constexpr int pct(int x) { return __builtin_popcount(x); }  // # of bits set
constexpr int bits(int x) {  // assert(x >= 0); // make C++11 compatible until
	                         // USACO updates ...
	return x == 0 ? 0 : 31 - __builtin_clz(x);
}  // floor(log2(x))
constexpr int p2(int x) { return 1 << x; }
constexpr int msk2(int x) { return p2(x) - 1; }

ll cdiv(ll a, ll b) {
	return a / b + ((a ^ b) > 0 && a % b);
}  // divide a by b rounded up
ll fdiv(ll a, ll b) {
	return a / b - ((a ^ b) < 0 && a % b);
}  // divide a by b rounded down

tcT > bool ckmin(T &a, const T &b) {
	return b < a ? a = b, 1 : 0;
}  // set a = min(a,b)
tcT > bool ckmax(T &a, const T &b) {
	return a < b ? a = b, 1 : 0;
}  // set a = max(a,b)

tcTU > T fstTrue(T lo, T hi, U f) {
	++hi;
	assert(lo <= hi);  // assuming f is increasing
	while (lo < hi) {  // find first index such that f is true
		T mid = lo + (hi - lo) / 2;
		f(mid) ? hi = mid : lo = mid + 1;
	}
	return lo;
}
tcTU > T lstTrue(T lo, T hi, U f) {
	--lo;
	assert(lo <= hi);  // assuming f is decreasing
	while (lo < hi) {  // find first index such that f is true
		T mid = lo + (hi - lo + 1) / 2;
		f(mid) ? lo = mid : hi = mid - 1;
	}
	return lo;
}
tcT > void remDup(vector<T> &v) {  // sort and remove duplicates
	sort(all(v));
	v.erase(unique(all(v)), end(v));
}
tcTU > void safeErase(T &t, const U &u) {
	auto it = t.find(u);
	assert(it != end(t));
	t.erase(it);
}

inline namespace IO {
#define SFINAE(x, ...)                                                         \
	template <class, class = void> struct x : std::false_type {};              \
	template <class T> struct x<T, std::void_t<__VA_ARGS__>> : std::true_type {}

SFINAE(DefaultI, decltype(std::cin >> std::declval<T &>()));
SFINAE(DefaultO, decltype(std::cout << std::declval<T &>()));
SFINAE(IsTuple, typename std::tuple_size<T>::type);
SFINAE(Iterable, decltype(std::begin(std::declval<T>())));

template <auto &is> struct Reader {
	template <class T> void Impl(T &t) {
		if constexpr (DefaultI<T>::value) is >> t;
		else if constexpr (Iterable<T>::value) {
			for (auto &x : t) Impl(x);
		} else if constexpr (IsTuple<T>::value) {
			std::apply([this](auto &...args) { (Impl(args), ...); }, t);
		} else static_assert(IsTuple<T>::value, "No matching type for read");
	}
	template <class... Ts> void read(Ts &...ts) { ((Impl(ts)), ...); }
};

template <class... Ts> void re(Ts &...ts) { Reader<cin>{}.read(ts...); }
#define def(t, args...)                                                        \
	t args;                                                                    \
	re(args);

template <auto &os, bool debug, bool print_nd> struct Writer {
	string comma() const { return debug ? "," : ""; }
	template <class T> constexpr char Space(const T &) const {
		return print_nd && (Iterable<T>::value or IsTuple<T>::value) ? '\n'
		                                                             : ' ';
	}
	template <class T> void Impl(T const &t) const {
		if constexpr (DefaultO<T>::value) os << t;
		else if constexpr (Iterable<T>::value) {
			if (debug) os << '{';
			int i = 0;
			for (auto &&x : t)
				((i++) ? (os << comma() << Space(x), Impl(x)) : Impl(x));
			if (debug) os << '}';
		} else if constexpr (IsTuple<T>::value) {
			if (debug) os << '(';
			std::apply(
			    [this](auto const &...args) {
				    int i = 0;
				    (((i++) ? (os << comma() << " ", Impl(args)) : Impl(args)),
				     ...);
			    },
			    t);
			if (debug) os << ')';
		} else static_assert(IsTuple<T>::value, "No matching type for print");
	}
	template <class T> void ImplWrapper(T const &t) const {
		if (debug) os << "\033[0;31m";
		Impl(t);
		if (debug) os << "\033[0m";
	}
	template <class... Ts> void print(Ts const &...ts) const {
		((Impl(ts)), ...);
	}
	template <class F, class... Ts>
	void print_with_sep(const std::string &sep, F const &f,
	                    Ts const &...ts) const {
		ImplWrapper(f), ((os << sep, ImplWrapper(ts)), ...), os << '\n';
	}
	void print_with_sep(const std::string &) const { os << '\n'; }
};

template <class... Ts> void pr(Ts const &...ts) {
	Writer<cout, false, true>{}.print(ts...);
}
template <class... Ts> void ps(Ts const &...ts) {
	Writer<cout, false, true>{}.print_with_sep(" ", ts...);
}
}  // namespace IO

inline namespace Debug {
template <typename... Args> void err(Args... args) {
	Writer<cerr, true, false>{}.print_with_sep(" | ", args...);
}
template <typename... Args> void errn(Args... args) {
	Writer<cerr, true, true>{}.print_with_sep(" | ", args...);
}

void err_prefix(str func, int line, string args) {
	cerr << "\033[0;31m\u001b[1mDEBUG\033[0m"
	     << " | "
	     << "\u001b[34m" << func << "\033[0m"
	     << ":"
	     << "\u001b[34m" << line << "\033[0m"
	     << " - "
	     << "[" << args << "] = ";
}

#ifdef LOCAL
#define dbg(args...) err_prefix(__FUNCTION__, __LINE__, #args), err(args)
#define dbgn(args...) err_prefix(__FUNCTION__, __LINE__, #args), errn(args)
#else
#define dbg(...)
#define dbgn(args...)
#endif

const auto beg_time = std::chrono::high_resolution_clock::now();
// https://stackoverflow.com/questions/47980498/accurate-c-c-clock-on-a-multi-core-processor-with-auto-overclock?noredirect=1&lq=1
double time_elapsed() {
	return chrono::duration<double>(std::chrono::high_resolution_clock::now() -
	                                beg_time)
	    .count();
}
}  // namespace Debug

inline namespace FileIO {
void setIn(str s) { freopen(s.c_str(), "r", stdin); }
void setOut(str s) { freopen(s.c_str(), "w", stdout); }
void setIO(str s = "") {
	cin.tie(0)->sync_with_stdio(0);  // unsync C / C++ I/O streams
	cout << fixed << setprecision(12);
	// cin.exceptions(cin.failbit);
	// throws exception when do smth illegal
	// ex. try to read letter into int
	if (sz(s)) setIn(s + ".in"), setOut(s + ".out");  // for old USACO
}
}  // namespace FileIO

#include <cassert>
#include <vector>

#include <cassert>
#include <numeric>
#include <type_traits>

namespace atcoder {

namespace internal {

#ifndef _MSC_VER
template <class T>
using is_signed_int128 =
    typename std::conditional<std::is_same<T, __int128_t>::value ||
                                  std::is_same<T, __int128>::value,
                              std::true_type, std::false_type>::type;

template <class T>
using is_unsigned_int128 =
    typename std::conditional<std::is_same<T, __uint128_t>::value ||
                                  std::is_same<T, unsigned __int128>::value,
                              std::true_type, std::false_type>::type;

template <class T>
using make_unsigned_int128 =
    typename std::conditional<std::is_same<T, __int128_t>::value, __uint128_t,
                              unsigned __int128>;

template <class T>
using is_integral =
    typename std::conditional<std::is_integral<T>::value ||
                                  is_signed_int128<T>::value ||
                                  is_unsigned_int128<T>::value,
                              std::true_type, std::false_type>::type;

template <class T>
using is_signed_int =
    typename std::conditional<(is_integral<T>::value &&
                               std::is_signed<T>::value) ||
                                  is_signed_int128<T>::value,
                              std::true_type, std::false_type>::type;

template <class T>
using is_unsigned_int =
    typename std::conditional<(is_integral<T>::value &&
                               std::is_unsigned<T>::value) ||
                                  is_unsigned_int128<T>::value,
                              std::true_type, std::false_type>::type;

template <class T>
using to_unsigned = typename std::conditional<
    is_signed_int128<T>::value, make_unsigned_int128<T>,
    typename std::conditional<std::is_signed<T>::value, std::make_unsigned<T>,
                              std::common_type<T>>::type>::type;

#else

template <class T> using is_integral = typename std::is_integral<T>;

template <class T>
using is_signed_int =
    typename std::conditional<is_integral<T>::value && std::is_signed<T>::value,
                              std::true_type, std::false_type>::type;

template <class T>
using is_unsigned_int =
    typename std::conditional<is_integral<T>::value &&
                                  std::is_unsigned<T>::value,
                              std::true_type, std::false_type>::type;

template <class T>
using to_unsigned =
    typename std::conditional<is_signed_int<T>::value, std::make_unsigned<T>,
                              std::common_type<T>>::type;

#endif

template <class T>
using is_signed_int_t = std::enable_if_t<is_signed_int<T>::value>;

template <class T>
using is_unsigned_int_t = std::enable_if_t<is_unsigned_int<T>::value>;

template <class T> using to_unsigned_t = typename to_unsigned<T>::type;

}  // namespace internal

}  // namespace atcoder

namespace atcoder {

template <class T> struct fenwick_tree {
	using U = internal::to_unsigned_t<T>;

  public:
	fenwick_tree() : _n(0) {}
	explicit fenwick_tree(int n) : _n(n), data(n) {}

	void add(int p) {
		assert(0 <= p && p < _n);
		p++;
		while (p <= _n) {
			++data[p - 1];
			p += p & -p;
		}
	}

	T sum(int l, int r) {
		assert(0 <= l && l <= r && r <= _n);
		return sum(r) - sum(l);
	}

	// private:
	int _n;
	std::vector<U> data;

	U sum(int r) {
		U s = 0;
		while (r > 0) {
			s += data[r - 1];
			r -= r & -r;
		}
		return s;
	}
};

}  // namespace atcoder

using namespace atcoder;

struct buffed {
	int B;
	fenwick_tree<int> fw;
	V<uint64_t> mask;
	buffed(int N) : B(N / 64 + 1), fw(B), mask(B) {}
	void add(int x) {
		fw.add(x / 64);
		mask[x / 64] ^= 1ULL << (x & 63);
	}
	int sum(int x) {
		return fw.sum(x / 64) +
		       __builtin_popcountll(mask[x / 64] & ((1ULL << (x & 63)) - 1));
	}
	void clear() {
		fill(all(fw.data), 0);
		fill(all(mask), 0);
	}
};

int main() {
	// read read read
	setIO();

	int N = 3000;
	int K = 1;

	re(N, K);

	vs grid(N);
	// each(t, grid) t = str(N, 'A');
	re(grid);

	V<vi> u(N, vi(N)), r(N, vi(N)), l(N, vi(N)), d(N, vi(N));
	R0F(i, N) R0F(j, N) {
		u[i][j] = r[i][j] = 1;
		if (j + 1 < N) u[i][j] += (grid[i][j] == grid[i][j + 1]) * u[i][j + 1];
		if (i + 1 < N) r[i][j] += (grid[i][j] == grid[i + 1][j]) * r[i + 1][j];
	}
	F0R(i, N) F0R(j, N) {
		l[i][j] = d[i][j] = 1;
		if (j) d[i][j] += (grid[i][j] == grid[i][j - 1]) * d[i][j - 1];
		if (i) l[i][j] += (grid[i][j] == grid[i - 1][j]) * l[i - 1][j];
	}

	vl ans(K);
	V<vi> upds(N + 1);

	buffed fw(N + 1);

	vi tmp_u(N), tmp_d(N), tmp_c(N);

	vi head(N + 1), val(N + 1), lnk(N + 1);

	FOR(diff, -N + 1, N) {
		fw.clear();

		int L = max(-diff, 0);
		int R = min(N - 1 - diff, N - 1);

		FOR(x, L, R + 1) {
			int y = x + diff;
			tmp_c[x] = grid[x][y] - 'A';
			tmp_u[x] = min(u[x][y], r[x][y]);
			tmp_d[x] = min(d[x][y], l[x][y]);
		}

		int tot_ones = 0;
		auto del_all = [&](int x) {
			for (int ptr = head[x]; ptr; ptr = lnk[ptr]) {
				int del_x = val[ptr];
				++tot_ones;
				fw.add(del_x);
			}
			head[x] = 0;
		};

		int ctr = 0;
		FOR(x, L, R + 1) {
			del_all(x);

			int y = x + diff;
			int c = tmp_c[x];

			int nx = x + tmp_u[x];
			++ctr;
			val[ctr] = x;
			lnk[ctr] = head[nx];
			head[nx] = ctr;

			ans[c] += tmp_d[x];
			ans[c] -= (tot_ones - fw.sum(x + 1 - tmp_d[x]));
		}
		del_all(R + 1);
		// dbg(ans);
	}
	F0R(i, K) ps(ans.at(i));
	dbg(time_elapsed());

	// you should actually read the stuff at the bottom
}

/* stuff you should look for
 * int overflow, array bounds
 * special cases (n=1?)
 * do smth instead of nothing and stay organized
 * WRITE STUFF DOWN
 * DON'T GET STUCK ON ONE APPROACH
 */