Tire cord zone impurity detection method, electronic device, readable storage medium and program product

Abstract

The application discloses a tire cord area impurity detection method, an electronic device, a readable storage medium and a program product, relates to the technical field of computer vision and industrial surface defect detection, and solves the problems that in the existing conventional detection method, dark impurities in an X-ray image are difficult to separate in a complex background, a weak target is prone to breakage after binarization, and long strip-shaped structural textures are prone to cause misjudgment. The method of the application highlights the cord and impurities through local adaptive threshold binarization, eliminates the vertical interference of cord textures by using a horizontal single-pixel inflation unit, thereby retaining the impurity form, then calculates the adaptive span lower limit by counting the length of continuous foreground pixel segments according to columns, dynamically selects the impurity suspected area, and removes abnormal data by using a column-level pixel total number threshold. Finally, the suspected area is subjected to joint geometric constraints of area and minimum bounding rectangle vertical height, real impurities are accurately identified, and the real impurities are mapped back to the X-ray image for color calibration. The application is mainly used for impurity detection on a tire cord area.

Images