Can Postgres beat Oracle for regexp_count?

However, it is not easy to replicate that in Postgres.  The following codes have been experimented but without any luck.

select regexp_matches('My High Street', '([A-Z][a-z]+[\s])', 'g')

select regexp_matches('My High Street', '([A-Z][a-z]+[\s]*)', 'g')

County occurrences of 'My High Street' in one of the following strings:

'My High Street'    1

''                             0

'My High Street My High Street'   2

Can anyone enlighten all of us?

Regards,

David