INFUSION BAG FOREIGN MATTER IDENTIFICATION METHOD AND DETECTION METHOD BASED ON ARTIFICIAL INTELLIGENCE VISUAL TECHNOLOGY

摘要

An infusion bag foreign matter identification method and detection method based on artificial intelligence visual technology, relating to the technical field of infusion bag detection. The identification method comprises the following steps: step 1: preprocessing an infusion bag; step 2: performing image preprocessing on the infusion bag; and step 3: performing image post-processing on the infusion bag, and identifying the infusion bag containing foreign matters. In the method, continuous image frames of a liquid medicine are obtained by means of a high-speed CCD industrial camera, then, related invariant characteristics of particle insoluble foreign matters in the images are mined in combination with motion features and morphological features of the particle insoluble foreign matters in the liquid, and an image processing algorithm is used to determine whether the particle insoluble foreign matters are foreign matters or an external interference, so that visual online automatic identification of the visible foreign matters in the infusion soft bag is realized.