A quality control system ensures that lumber is manufactured to the correct size and highest grade. The system currently used by most sawmills is labour intensive and as a result, limits the frequency of quality monitoring. Automatic in-line systems have been developed but found to be inaccurate and produce false alarms. UBC researchers have identified the problem with automatic systems to be faulty statistical assumptions and poor design and is developing a new statistical method and scanning system that is more reliable and accurate. Forintek Canada Corp. is developling a defect recognition system based on the new UBC system. The UBC system utilizes single point laser sensors that scan the surface of lumber along its length. This provides the potential for detecting eleven common machining and machinery related lumber defects. Washboarding, saw marks, roll marks, spike marks and scallop are repetitive defects that can potentially be detected by the system. Wane, taper, snipe, tear out and wedge are non-repetitive defects that may also be identifiable. Fast Fourier Transform (FFT) and other algorithms are being developed to detect repetitive defects. Each defect produces a FFT frequency spectrum that can be used as a fingerprint for identification. Development of algorithms to diagnose non-repetitive defects will begin in the near future.