Rapid influenza diagnostic test (RIDT) is one of the most common tools for screening patients suspected of influenza infection. The principle is to detect the surface antigen of influenza virus with known antibodies, and then to interpret it with the naked eye in the form of immune chromatographic as says. It has the advantage of obtaining speedy results (10-30 minutes) and ease of operation (which can be interpreted with the naked eye).There is a variety of rapid influenza diagnostic tests (RIDTs) available in the market, with different sensitivities and specificities depending on the design of the antibody location and reagent composition. Despite its advantages of speed and convenience, a high percentage of test results (20 to 50% or higher) do not correctly reflect the patient's status. In addition to possible misses in the specimen collection process that will affect the tests; the naked eye may not be able to distinguish the unapparent results and cause false negatives. At the same time, because a healthcare worker may not accurately grasp the time of interpretation, false positives can also occur due to excessive test times. To minimize incorrect diagnoses, we propose an interpretation system using machine vision. The system replaces the function of a healthcare worker by a camera and computer. The camera captures the image of the test piece then sent it to the computer for processing and identification; the result can provide the medical staff reference.