Machine vision system

• US 4,995,087 A
• Filed: 08/11/1989
• Issued: 02/19/1991
• Est. Priority Date: 05/15/1989
• Status: Expired due to Term

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Bottoms, John J., Wolff, Steven M., Burkel, Richard H., Gantner, Randall C., Rathi, Rajendra P., Tashjian, David G.
• Primary Examiner(s): Moore, David K.
• Assistant Examiner(s): Couso, Jose L.
• Application Number:
  US07/392,517
• Time in Patent Office: 557 Days
• Field of Search: 382/8, 382/22, 382/25, 382/48, 358/101, 358/107, 364/167.01, 364/474.28, 364/474.29, 219/121.7
• US Class Current: 382/152
• CPC Class Codes: B23K 26/032 using optical means
• CPC Class Codes: B23P 6/007 using only additive methods, e.g. build-up welding
• CPC Class Codes: G05B 19/4207 in which a model is traced or scanned and corresponding data recorded
• CPC Class Codes: G05B 2219/36495 Recording position and other parameters, current, tool diameter, voltage
• CPC Class Codes: G05B 2219/37572 Camera, tv, vision
• CPC Class Codes: G05B 2219/45165 Laser machining
• CPC Class Codes: G05B 2219/50214 Refurbish, refinish, reprofile, recondition, restore, rebuild profile

Abstract

A vision system for automating a machine process. The system includes a conventional CCD camera and a vision processor which automatically determines process parameters from a workpiece surface by generating an image of the workpiece surface, transducing the workpiece image into electrical signals, and electronically determining the process parameters from the electrical signals. Applied to a laser welding process, the vision system locates workpiece edges, determines part centerline and thickness, and calculates process parameters such as weld location, the rate of delivery of weld powder, laser intensity, and the speed at which the weld is applied.

Claims

1. A method for guilding a tool along a workpiece surface, the steps comprising:

generating an image of said workpiece surface;

dissecting said image into a plurality of picture elements;

identifying the coordinates of a set of picture elements associated with a boundary of said workpiece surface;

generating a mathematical equation from said picture element coordinates, said equation defining a curve formed by the boundary of said workpiece surface;

selecting a first plurality of points along said curve;

electronically determining a slope for said curve at each of said first plurality of points ;

generating a set of mathematical equations, each of said equations defining a line normal to said curve and each normal line intersecting said curve at a different one of said first plurality of points;

identifying a second plurality of points, each one of said second plurality of points being located along a different one of said normal lines at a predetermined distance from an associated one of said first plurality of points; and

guiding said tool along a path including said second plurality of points.

4. The method of claim 1, wherein:

said step of generating a mathematical equation from said picture element coordinates includes the step of determining a set of fourier Coefficients corresponding to said picture element coordinates and substituting said Fourier Coefficients into a Fourier Series equation; and

said step of electronically determining the slope of said curve at selected points along said curve includes the step of calculating the derivative of said Fourier Series equation at said selected points.

2. A method for guiding a tool along a workpiece surface, comprising the steps of:

generating an image of said workpiece surface, said step of generating an image including the steps of illuminating said surface, and focusing electromagnetic energy reflected from said surface onto a sensor;

dissecting said image into a plurality of picture elements;

identifying the coordinates of a set of picture elements associated with a boundary of said workpiece surface, said step of identifying the coordinates of the set of picture elements including the steps of;

(a) assigning a luminance value to each picture element, the magnitude of said luminance value being proportional to the intensity of electromagnetic energy associated with the corresponding picture element, and

(b) comparing said luminance values to identify picture elements associated with said workpiece surface and picture elements associated with background areas;

generating a mathematical equation from said picture element coordinates, said equation defining a curve formed by the boundary of said workpiece surface;

selecting a first plurality of points along said curve;

electronically determining a slope for said curve at each of said first selected points;

generating a set of mathematical equations, each of said set of equations defining a line normal to said curve and each normal line intersection said curve at a different one of said first plurality of points;

identifying a second plurality of points, each one of said second plurality of points being located along a different one of said normal lines at a predetermined distance from an associated one of said first plurality of points; and

guiding the tool along a path including said second plurality of points.

3. The method of claim 2, wherein said step of comparing luminance values to identify picture elements associated with said surface and picture elements associated with background areas includes the steps of:

determining a threshold luminance value;

comparing said picture element luminance values to said threshold luminance value;

identifying picture elements associated with said surface as those having luminance values greater than said threshold value; and

identifying picture elements associated with said background as those having luminance values less than said threshold value.