Research On Tube-type Bottle Inspecting System Based On Machine Vision

Currently, machine version technology has widely used in medical packaging and food processing industry due to its huge advantages such as out of contact inspection, no time limitation, high accuracy and speed and so on, having broad market prospects. In this paper, in order to realize automatic inspection towards bottle size and default, it describes the relevant researches based on machine vision. It focuses on the follow work:1.Introduce machine version technology and bottle inspection technique, set up hardware testing platform including designs of lighting system and layout and selections of PC, image acquisition card and camera.2.Carry out various image filtering , enhancement and segmentation algorithms' experiments with analysis and comparisons. select an feasible and advanced image processing operation flow embracing mathematical morphology filtering, OSTU segmentation, improved morphological edge detection, etc. And calibrate the circle center of top bottle image by Hough transform, calculate the bottles'size and faults' geometric parameters.3.Inspect bottles by taking pattern recognition measures, of which, using BP netral network to identify fault locations. And analyze system error, propose instruction of decreasing error for future research. At last, design this system's GUI.