New Challenge of Protecting Privacy due to Stained Recognition

k-Anonymity is a good way to protect people's privacy. It is efficient to find which part should be covered up by using mosaic. Traditional OCR could lose efficiency when the part covered by mosaic. Considering that words in text can be connected by logical relationship, with the help of the idea that reducing the size of word stock which references from license plate recognition, this paper established N-GRAM model and get a wonderful result to recognize fuzzy parts. This may be a big challenge of k-Anonymity.