Identification of Screw threads parameters automatically by artificial intelligence, classifications, grouping, labeling and fuzzy expert systems.

The production of threaded parts in some industrial environments requires
high precision, measurement and reliable inspection methods.
In this process, we use a pin configurator to calibrate pixel-sized vision
systems and two standard steps (maximum and minimum) for image correction
and image distortion. These methods set the system to determine the maximum,
and prepare the average of the inner and outer diameters of each piece.
Measuring the step using the cross-section of each pair of threads in the
image is obtained. Dark areas and methods (crown and root) for screw diameter
screws are tested accurately. The measurement methods are for the length of the
thread and the thread length is useful in the thread inspection process.
The initial image of the screw thread can be obtained by CCD and
geometric parameters can be measured through filtering (image smoothing),
edge detection, generating banner images, and detecting contour.
A method for measuring the angle parameter of the threaded crowns by the
visual machine is discussed in this study and the measurement data is provided
at the end. The feasibility and feasibility of this method are discussed in general
and in theory.