In the dental implant operation, the quality of bone mineral density (BMD) is required since if a patient is low BMD, the implant operation may be failed and lead to complication. Thus, the dentist should have a BMD examination before the implant operation. In order to explore the BMD, the patient need to take extra time and cost to take extra efforts to take a BMD examination and get the t-score by Dual-energy X-ray absorptiometry (DXA). In this paper, image processing technique is used to getting the gray level of trabecular bone from cone beam computed tomography (CBCT) images. And then we explore the relationship between the gray level and t-score by linear regression. That is, the gray level of CBCT images can be transform into the t-score by the regression equation and discriminate whether the patient is normal or low BMD. The experiment results show that the CBCT image can be used to explore the BMD of the implant area and provide bone information before the implant operation.