Inspection method using area of interest (AOI) analysis

• US 5,369,713 A
• Filed: 06/28/1993
• Issued: 11/29/1994
• Est. Priority Date: 07/09/1992
• Status: Expired due to Fees

• Current Assignee: Arie Shahar, Nira Schwartz, Richard Woods
• Original Assignee: Arie Shahar, Nira Schwartz, Richard Woods
• Inventors: Woods, Richard, Shahar, Arie, Schwartz, Nira
• Primary Examiner(s): Couso, Jose L.
• Assistant Examiner(s): ANDERSON, DAVID
• Application Number:
  US08/082,437
• Time in Patent Office: 519 Days
• Field of Search: 382/8, 356/240, 425/172, 425/140, 250/223 B, 358/101
• US Class Current: 382/142
• CPC Class Codes: B07C 5/3408 for bottles, jars or other glassware according to size B07C5/122
• CPC Class Codes: G01B 11/024 by means of diode-array scanning
• CPC Class Codes: G01N 21/9027 in containers after filling
• CPC Class Codes: G01N 21/9054 Inspection of sealing surface and container finish
• CPC Class Codes: G01N 21/9081 Inspection especially designed for plastic containers, e.g. preforms

Abstract

A technique for detecting defects in stationary products or in products moving on a production line (102, FIG. 1) by analyzing area of interest (AOls) of their modified images uses a matrix or linescan camera (104, FIG. 1) for taking images of products (102). The product'"'"'s dimensions are measured with accuracy, and the existence and alignment of caps and seals is determined. The technique is much faster and more accurate than current techniques and is based up on an analysis of the AOls and their discontinuities. (209 FIG. 1). Carefully selected AOls of the modified image (2 to 8, FIG. 2) are saved in the memory of a computer (106, FIG. 1). The method also includes loading look-up tables to modify the gray levels of the products; saving AOls in memory to be analyzed (FIG. 1); analyzing AOl data, counting pixel discontinuities, etc. The results can be used to measure product ovality, check caps and seals on products, and check changes of fluid or content levels in containers.

Claims

1. A method for inspecting a container that is partially filled with a fluid having a surface level, which may have defects, and which moves on a production line in a predetermined direction, comprising:

squeezing said container with a predetermined force to change the surface level of said fluid,

providing a light source which is arranged to shine light upon a finish of said container, including the surface level of said fluid, so that said light is reflected from said finish of said container along a return path,

providing a sensor in said return path to collect light reflected from said finish of said container,

providing a processing unit with a memory,

creating a container image of said finish by using said sensor to sense reflected light from the finish of said container, said sensor being arranged to produce an electrical signal representative of a reflected product image of said finish, said container image comprising a multiplicity of pixels with intensity levels expressed as respective gray levels,

modifying said container image in said processing unit to produce a modified container image by modifying said gray levels of said container image to highlight transition points of said container image,

providing and saving in said memory of said processing unit a plurality of computer-generated artificial areas of interest of said modified container image, said plurality of computer generated artificial areas of interest representing separate areas of said container which are separated by a plurality of pixels, all of said pixels of said plurality of computer generated artificial areas of interest together being less than the number of pixels of which represent said container image, at least one of said plurality of computer generated artificial areas of interest including the gray levels of only some of said container'"'"'s pixels, regardless of any uncertainty in said container'"'"'s position in said given direction,

analyzing said transition points in said plurality of computer generated artificial areas of interest, and

utilizing said plurality of computer generated artificial areas of interest of said modified container image for product inspection.

2. The method of claim 1 further including making container images of said finish of said container, including the surface level of said fluid, before and after said squeezing.

3. A method for inspecting a sealed container that has been partially filled with a fluid having a surface level, which may have defects, and which moves on a production line in a predetermined direction, comprising:

applying a force to said container to cause the level of said fluid to rise inside said container, providing a light source which is arranged to shine light upon a finish of said container, including the surface level of said fluid, so that said light is reflected from said finish of said container along a return path,

providing a sensor in said return path to collect light reflected from said finish of said container,

providing a processing unit with a memory,

creating a container image of said finish by using said sensor to sense reflected light from the finish of said container, said sensor being arranged to produce an electrical signal representative of a reflected product image of said finish, said container image comprising a multiplicity of pixels with intensity levels expressed as respective gray levels,

modifying said container image in said processing unit to produce a modified container image by modifying said gray levels of said container image to highlight transition points of said container image,

providing and saving in said memory of said processing unit a plurality of computer-generated artificial areas of interest of said modified container image, said plurality of computer generated artificial areas of interest representing separate areas of said container which are separated by a plurality of pixels, all of said pixels of said plurality of computer generated artificial areas of interest together being less than the number of pixels of which represent said container image, at least one of said plurality of computer generated artificial areas of interest including the gray levels of only some of said container'"'"'s pixels, regardless of any uncertainty in said container'"'"'s position in said given direction,

analyzing said transition points in said plurality of computer generated artificial areas of interest, and

utilizing said plurality of computer generated artificial areas of interest of said modified container image for product inspection.

4. The method of claim 3 wherein said container is squeezed to cause the level of said fluid therein to rise.