This invention discloses an intelligent identification method for hepatobiliary and renal stones based on
ultrasound images, belonging to the field of stone
identification technology. The method includes the following steps: acquiring an
ultrasound image to be detected and marking it as a real-time detection image; obtaining real-time detection
grayscale values based on the real-time detection image; obtaining organ
grayscale thresholds based on
ultrasound images in a
database; obtaining real-time organ regions based on the real-time detection
grayscale values and organ grayscale thresholds; constructing real-time stone detection values based on the real-time organ regions; obtaining historical normal
boundary values based on ultrasound images in a
database; obtaining stone boundary thresholds based on the historical normal
boundary values; and determining whether
gallstones are present based on the real-time stone detection values and stone boundary thresholds. This invention addresses the problem that existing stone identification technologies suffer from poor accuracy due to significant
noise in ultrasound images.