A face recognition system based on Five Transform Domains – Two Dimensional – Principal and Minor Component Analysis; TDs – 2D – Mu(P+m)CA is proposed. Each face in the spatial domain is first preprocessed then divided into horizontal, vertical halves and diagonal format. Second, these partitions are compressed using the Discrete Wavelet Transform (DWT). Third they are further transformed using the Discrete Cosine Transform (DCT). In addition, starting from the DWT, each partition underwent three processes, namely Horizontal and Vertical Edge Detection and High pass Filtered. Third, each partition is further compressed using 2DPCA. Finally, after computing and saving best principal components, minor components are investigated if they can further boost the recognition rate. A voting scheme is used to define ground truth identity. The performance of the proposed system is evaluated using k-fold cross validation of ORL, Yale and FERET databases. Sample results are presented. Compared with the state of art techniques, the proposed technique achieves higher recognition rates while retaining low savings in storage and computation requirements.