#include <bits/stdc++.h>
#define va first
#define vb second
#define lb lower_bound
#define ub upper_bound
#define bs binary_search
#define pp push_back
#define ep emplace_back
#define all(v) (v).begin(),(v).end()
#define szz(v) ((int)(v).size())
#define bi_pc __builtin_popcount
#define bi_pcll __builtin_popcountll
#define bi_tz __builtin_ctz
#define bi_tzll __builtin_ctzll
#define fio ios_base::sync_with_stdio(0);cin.tie(0);
#ifdef TAMREF
#define debug(...) fprintf(stderr, __VA_ARGS__)
#else
#define debug(...) 42
#endif
using namespace std;
using ll = long long; using lf = long double;
using pii = pair<int,int>; using ppi = pair<int,pii>;
using pll = pair<ll,ll>; using pff = pair<lf,lf>;
using ti = tuple<int,int,int>;
using base = complex<double>;
const lf PI = 3.14159265358979323846264338L;
template <typename T>
inline T umax(T& u, T v){return u = max(u, v);}
template <typename T>
inline T umin(T& u, T v){return u = min(u, v);}
mt19937_64 rng(chrono::steady_clock::now().time_since_epoch().count());
template<class> struct is_container : false_type {};
template<class... Ts> struct is_container<vector<Ts...>> : true_type {};
template<class... Ts> struct is_container<map<Ts...>> : true_type {};
template<class... Ts> struct is_container<unordered_map<Ts...>> : true_type {};
template<class... Ts> struct is_container<set<Ts...>> : true_type {};
template<class... Ts> struct is_container<unordered_set<Ts...>> : true_type {};
template<class... Ts> struct is_container<multiset<Ts...>> : true_type {};
template<class... Ts> struct is_container<unordered_multiset<Ts...>> : true_type {};
template<class T, class = typename enable_if<is_container<T>::value>::type>
ostream& operator<<(ostream &o, T x) {
#ifndef TAMREF
return o;
#endif
int f = 1;
o << "{";
for (auto y : x) {
o << (f ? "" : ", ") << y;
f = 0;
}
return o << "}\n";
}
template<class T, class U>
ostream& operator<<(ostream &o, pair<T, U> x) {
#ifndef TAMREF
return o;
#endif
return o << "(" << x.first << ", " << x.second << ")";
}
template<ll mod>
struct NTT{
using ll = long long;
inline ll pw(ll x, ll n) {
ll ret = 1, pv = x;
for(; n; n >>= 1){
if(n & 1) ret = ret * pv % mod;
pv = pv * pv % mod;
}
return ret;
}
inline ll multInv(ll x){
return pw(x, mod - 2);
}
void ntt(vector<ll>& a, bool inv){
const int w = 3; //for mod = 998244353
int n = szz(a), j = 0;
vector<ll> roots(n/2);
for(int i = 1; i < n; i++){
int b = (n >> 1);
while(j >= b) {
j -= b; b >>= 1;
}
j += b;
if(i < j) swap(a[i], a[j]);
}
ll ang = pw(w, (mod - 1) / n);
if(inv) ang = pw(ang, mod - 2);
for(int i = 0; i < n / 2; i++){
roots[i] = (i ? roots[i-1] * ang % mod : 1);
}
for(int i = 2; i <= n; i <<= 1){
int step = n / i;
for(int j = 0; j < n; j += i){
for(int k = 0; k < i / 2; k++){
ll u = a[j + k], v = a[j + k + i/2] * roots[step * k] % mod;
a[j + k] = (u + v) % mod;
a[j + k + i/2] = (u + mod - v) % mod;
}
}
}
ll ninv = multInv(n);
if(inv) for(int i = 0; i < n; i++) a[i] = a[i] * ninv % mod;
}
void ntt_row(vector<vector<ll>> &a, bool inv) {
const int w = 3; //for mod = 998244353
int n = szz(a), j = 0, m = szz(a[0]);
vector<ll> roots(n/2);
for(int i = 1; i < n; i++){
int b = (n >> 1);
while(j >= b) {
j -= b; b >>= 1;
}
j += b;
if(i < j) a[i].swap(a[j]);
}
ll ang = pw(w, (mod - 1) / n);
if(inv) ang = pw(ang, mod - 2);
for(int i = 0; i < n / 2; i++){
roots[i] = (i ? roots[i-1] * ang % mod : 1);
}
for(int i = 2; i <= n; i <<= 1){
int step = n / i;
for(int j = 0; j < n; j += i){
for(int k = 0; k < i / 2; k++){
for(int h = 0; h < m; h++) {
ll u = a[j + k][h], v = a[j + k + i/2][h] * roots[step * k] % mod;
a[j + k][h] = (u + v) % mod;
a[j + k + i/2][h] = (u + mod - v) % mod;
}
}
}
}
ll ninv = multInv(m);
if(inv) for(int i = 0; i < n; i++) for(int j = 0; j < m; j++) a[i][j] = a[i][j] * ninv % mod;
}
void ntt2d(vector<vector<ll>> &a, bool inv) {
int n = szz(a);
for(int i = 0; i < n; i++) ntt(a[i], inv);
ntt_row(a, inv);
}
vector<vector<ll>> mult2d(vector<vector<ll>> a, vector<vector<ll>> b) {
int n = 2; while(n < szz(a) + szz(b)) n <<= 1;
int m = 2; while(m < szz(a[0]) + szz(b[0])) m <<= 1;
a.resize(n); b.resize(n);
for(int i = 0; i < n; i++) a[i].resize(m);
for(int i = 0; i < n; i++) b[i].resize(m);
ntt2d(a, false); ntt2d(b, false);
for(int i = 0; i < n; i++) for(int j = 0; j < m; j++) a[i][j] = a[i][j] * b[i][j] % mod;
ntt2d(a, true);
return a;
}
vector<ll> mult(vector<ll> &a, vector<ll> &b){
vector<ll> fa(all(a)), fb(all(b));
int n = 2; while(n < szz(a) + szz(b)) n <<= 1;
fa.resize(n); fb.resize(n);
ntt(fa, false); ntt(fb, false);
for(int i = 0; i < n; i++) fa[i] = fa[i] * fb[i] % mod;
ntt(fa, true);
return fa;
}
};
constexpr ll mod1 = 998244353, mod2 = 7 << 26 | 1;
NTT<mod1> n1;
NTT<mod2> n2;
vector<vector<ll>> a, b;
int n, m, r, c;
const ll tmods[4] = {1000000007, 1000000009, 998244353, 7 << 26 | 1};
using L = __int128_t;
void merge_by_crt(vector<vector<ll>> &a, vector<vector<ll>> &b) {
ll inv1 = n2.multInv(mod1), inv2 = n1.multInv(mod2), pmod = mod1 * mod2;
for(int i = 0; i < szz(a); i++) {
for(int j = 0; j < szz(b); j++) {
a[i][j] = (
__int128_t(a[i][j]) * mod2 * inv2 + __int128_t(b[i][j]) * mod1 * inv1
) % pmod;
}
}
}
int main(){
fio;
cin >> n >> m;
a = vector<vector<ll>>(n, vector<ll>(m));
for(int i = 0; i < n; i++) for(int j = 0; j < m; j++) cin >> a[i][j];
cin >> r >> c;
b = vector<vector<ll>>(r, vector<ll>(c));
for(int i = 0; i < r; i++) for(int j = 0; j < c; j++) cin >> b[r-1-i][c-1-j];
auto ab = n1.mult2d(a, b);
auto ab2 = n2.mult2d(a, b);
merge_by_crt(ab, ab2);
#ifdef TAMREF
cerr << ab << '\n';
#endif
__int128_t asum = 0, a2sum = 0, bsum = 0, b2sum = 0;
for(int i = 0; i < r; i++) {
for(int j = 0; j < c; j++) {
bsum += b[i][j];
b2sum += b[i][j] * b[i][j];
}
}
ll ans = 0;
auto compare = [&](int i, int j) {
for(const ll mod : tmods) {
auto pw = [&](ll x, ll n) -> ll {
ll r = 1, p = x;
for(;n;n>>=1) {
if(n&1) r = r * p % mod;
p = p * p % mod;
}
return r;
};
auto multInv = [&](ll x) -> ll {
return pw(x, mod - 2);
};
auto normalize = [&](ll x) -> ll {
x += mod;
return x >= mod ? x - mod : x;
};
ll _a = asum % mod, _b = bsum % mod, _a2 = a2sum % mod, _b2 = b2sum % mod;
ll rcinv = multInv(r * c), rcinv2 = rcinv * rcinv % mod;
ll bvar = normalize((_b2 * rcinv - _b * _b % mod * rcinv2) % mod);
ll cov = normalize((ab[i + r - 1][j + c - 1] % mod * rcinv - _a * _b % mod * rcinv2) % mod);
ll avar = normalize((_a2 * rcinv - _a * _a % mod * rcinv2) % mod);
ll diff = normalize((cov * cov - avar * bvar) % mod);
if(diff) {
debug("i = %d, j = %d, mod = %lld, diff = %lld\n", i, j, mod, diff);
}
if(diff) return;
}
++ans;
debug("ans = %lld\n", ans);
};
for(int k = 0; k + r - 1 < n; k++) {
asum = a2sum = 0;
for(int i = 0; i < r; i++) {
for(int j = 0; j < c; j++) {
asum += a[k+i][j];
a2sum += a[k+i][j] * a[k+i][j];
}
}
compare(k, 0);
for(int j = 1; j + c - 1 < m; j++) {
for(int i = 0; i < r; i++) {
asum += a[k+i][j+c-1] - a[k+i][j-1];
a2sum += a[k+i][j+c-1] * a[k+i][j+c-1] - a[k+i][j-1] * a[k+i][j-1];
}
compare(k, j);
}
}
cout << ans << '\n';
}
