The aim of this study was firstly to assess inter- and intra- examiner reproducibility and accuracy in the detection of occlusal caries in extracted human teeth using the newly developed fluorescence-based camera VistaCam iX and secondly to compare the performance to the established fluorescence device VistaProof. The occlusal surfaces of 101 teeth were assessed. The distribution of the lesions were characterized first visually using ICDAS-II (consensus score). The investigation sites were assessed by two examiners with different levels of experience in cariology (one experienced dentist, one final-year dental student) by both fluorescence-based cameras VistaCam iX (FC1) and VistaProof (FC2). The teeth were hemisectioned and assessed for lesion depth. Intra-class-correlation coefficients for inter- and intra-examiner reproducibility were 0.88-0.97 (FC1) and 0.82-0.98 (FC2), respectively. There was significant positive correlation (r (s), p < 0.01) between all methods (ICDAS-II, fluorescence, and histological examinations: 0.63-0.89) and between FC1 and FC2 (r (s) 0.85-0.90), respectively. Areas under the ROC curves (AUC) were 0.87-0.92 (D(1) and D(3) diagnostic threshold, FC1) and 0.91-0.96 (FC2). There were no significant differences between the AUC of both fluorescence cameras (p values > 0.05). Both fluorescence cameras demonstrated high reproducibility and good performance for the detection of occlusal caries at various stages of the disease process. The novice and the experienced examiner were able to apply both systems for detection of lesions. The in vitro performance of both devices was comparable to each other, although there was a tendency of a better performance for the FC2. Thus, within the limitations of an in vitro study, measurements with the FC2 can be continued by the new fluorescence camera (FC1) and data formerly assessed can be compared without significant loss of information.