Maybe it can work as follows: Support a concept of "reviewer group" that can be created by anyone but can only be joined on invitation. When you post a review, the groups you belong to are shown; you can also filter articles that are approved by a certain group. Eventually, some groups will gain reputation by producing quality reviews on certain fields and people will treat papers approved by them as high-quality.
Heck, you can even create groups like "former reviewers of XXX journal" and as soon as you convince other people, gain reputation very quickly.
Heck, you can even create groups like "former reviewers of XXX journal" and as soon as you convince other people, gain reputation very quickly.