This paper presents the design and implementation of a tourist landmarks recommending system, which adopts the content-based image retrieval technology. The traditional text-based image retrieval technology has been hardly satisfying the users' requirements. To lower the cost, this system uses B/S architecture, distributing the work of logical transaction operation to the server side. Meanwhile, for promoting the maintainability and reducing the coupling of code, this system uses MVC model to organize code. Also, the users upload the data as input. The refined multiple-feature fusion algorithm and bidirectional matching strategy enable the functions like landmarks recognition or searching similar landmarks, which satisfy the customers' need preferably.

No abstract