As the basic material of light emitting diode (LED) substrate, the quality of sapphire directly affects the production yield of high-brightness LED substrate chips. The detection and positioning of internal defects in sapphire crystal bar can greatly reduce the production cost and improve the yield. Based on machine vision technique, an internal defect detection and positioning system are bulit for sapphire on the basis of laser light scattering tomography taking the processing requirements on sapphire substrate into consideration. LabVIEW and image acquisition card and data acquisition card from NI Company are adopted in this system. The detection and positioning of internal defects in sapphire crystal is realized by scanning the sapphire crystal layer by layer through the light scattering effect generated by the irradiation of high-intensity linear laser on defects. A lot of image processing methods such as image enhancement, image segmentation and image extraction are adopted to extract and analyze the defects, enhancing the real-time performance and visuality of this system. The experimental results show that this system can identify the scattering particle defects in sapphire crystal effectively, and position them in depth direction accurately.