Traversing the database to find the most similar palmprint is an extremely time-consuming operation.

Therefore, an algorithm needs to be designed to achieve similar palmprint search in less time while ensuring the accuracy.

There are N records in the database. If we traverse the database, time complexity is O(n).

Now let's group all data into L1 clusterings. For every clustering, if it's records more than x, group all data into L2 clusterings. Iterate this step until the minimum granularity of the clusterings data size is less than x. The time complexity is O(L1+L2+...+Ln+x).