The detection of small radar cross section (RCS) targets is one of the most critical requirements for modern radars deployed in coastal and maritime scenarios. This requirement reflects the need to have a rapid awareness about the presence of potential threats close to strategical areas like military harbours, power plants, offshore facilities for oil and gas extraction, preserved wilderness area, etc. The first part of this article overviews the state-of-the-art of high-resolution radars working in X-band and Ka-band. The higher resolution leads to better target separation that is a key benefit in dense or restricted areas where close moving objects can be merged together by traditional maritime radars. The detection performance of these sensors is analyzed considering worst-case scenario where the RCS of a sea wave can exceed the RCS of a real small target. If proper clutter filtering is not adopted, the principal consequence of a high presence of sea clutter is the masking of real target echoes and the increasing of the false alarm rate with subsequent detection performance degradation. Modern processing techniques for the detection of small targets in sea clutter are then presented; these include both non-coherent techniques based on constant false alarm rate (CFAR), time-frequency diversity, scan-to-scan correlation and coherent multichannel techniques.