This document provides a review of graph-based image classification and frequent subgraph mining algorithms. It first introduces graph-based image representation and the need for approximate subgraph matching to account for noise and distortions. It then surveys several existing frequent subgraph mining algorithms, including CSMiner, SUBDUE, gdFil, FSG, and PrefixSpan. These algorithms are categorized as Apriori-based approaches or pattern-growth approaches. The document also discusses applications of graph mining in other domains such as chemistry and biology.