The vexing difficulty in distinguishing glioma from normal tissues is a major obstacle to prognosis. In an attempt to solve this problem, we used a joint strategy that combined targeted-cancer stem cells nanoparticles with precise photoacoustic and fluorescence navigation. We showed that traditional magnetic resonance imaging (MRI) did not represent the true morphology of tumors. Targeted nanoparticles specifically accumulated in the tumor area. Glioma was precisely revealed at the cellular level. Tumors could be non-invasively detected through the intact skull by fluorescence molecular imaging (FMI) and photoacoustic tomography (PAT). Moreover, PAT can be used to excise deep gliomas. Histological correlation confirmed that FMI imaging accurately delineated scattered tumor cells. The combination of optical PAT and FMI navigation fulfilled the promise of precise visual imaging in glioma detection and resection. This detection method was deeper and more intuitive than the current intraoperative pathology.