System and method for counting parts in multiple fields of view using machine vision

• US 6,483,935 B1
• Filed: 10/29/1999
• Issued: 11/19/2002
• Est. Priority Date: 10/29/1999
• Status: Expired due to Term

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Dampier, Todd O., Mangalindan, Edwin C., Rostami, Fariborz
• Primary Examiner(s): CHANG, JON CARLTON
• Application Number:
  US09/430,107
• Time in Patent Office: 1,117 Days
• Field of Search: 382/141, 382/142, 382/143, 382/192, 382/194, 382/104, 382/133, 382/134, 209/509, 209/551, 348/86, 198/958, 340/933, 340/934, 340/937
• US Class Current: 382/141
• CPC Class Codes: G01N 21/8901 Optical details; Scanning details per se G02B
• CPC Class Codes: G06M 1/101 by electro-optical means
• CPC Class Codes: G06M 7/00 Counting of objects carried by a conveyor

Abstract

A system and method for counting objects or parts on a continuously moving conveyor, or other elongated surface, provides a camera and associated machine vision system that acquire/capture and analyze multiple, successive fields of view taken within an area of interest defined by the system. Each of the fields of view includes flanking right and left overlap zones with respect to adjacent fields of view. The overlap zones are defined to be at least as wide as the maximum width of a part being counted. The captured fields of view are stored within the system and analyzed in succession based upon a set of rules. These rules determine whether or not a particular part within the subject field of view is to be counted. More specifically, the rules are applied within the overlap zones for each field of view to determine whether or not to count the part. Certain refinements to basic rules are employed to establish uncertainty regions within the overlapping zones to account for possible errors in tracking objects between fields of view.

Claims

1. A system for counting objects on an elongated surface comprising:

a camera that defines an area of interest with respect to an elongated direction of the surface, the camera and the surface being in relative movement with respect to each other in an upstream-to-downstream direction;

a vision system for identifying locations of objects occurring within the area of interest within a predetermined interval;

a trigger mechanism that operates at predetermined times to enable the vision system to acquire an image within the area of interest each time the surface moves a predetermined distance along the elongated direction, the trigger mechanism being constructed and arranged so that each acquired image overlaps an adjacent acquired image by a predetermined overlap distance along the elongated direction, thereby defining, in each area of interest, a central section, an upstream overlapping section upstream of the central section, and a downstream overlapping section downstream of the central section; and

a counter constructed and arranged to count (a) objects identified that are located at least partially in the central section and (b) objects identified that are located fully within only one of either the upstream overlapping section and the downstream overlapping section in the area of interest for each acquired image.

2. The system as set forth in claim 1 wherein the surface is a moving conveyor and is operatively connected to a movement measuring device that provides a signal based upon the movement for the predetermined distance of the surface, and further comprising a controller responsive to the measuring device that operates the trigger mechanism at the predetermined times each time the predetermined movement of the surface is signaled.

3. The system as set forth in claim 2 further comprising a shutter assembly operatively connected to the camera constructed and arranged to selectively acquire an image at the predetermined times in which the trigger mechanism operates.

4. The system as set forth in claim 2 further comprising a strobe flash responsive to the controller, constructed and arranged to illuminate the area of interest to thereby acquire an image by the camera at the predetermined times in which the trigger mechanism operates.

5. The system as set forth in that claim 1 wherein the counter is constructed and arranged to define an upstream uncertainty region associated overlaid upon the upstream overlapping region and a downstream uncertainty region overlaid upon the downstream overlapping region, based upon a predetermined range of expected erroneous movement of the objects in the elongation direction between each acquired image in the area of interest.

6. The system as set forth in claim 5 wherein each of the upstream uncertainty region and the downstream uncertainty region is defined as a distance in the elongated direction that is a difference between a maximum expected erroneous movement of the objects and a minimum expected erroneous movement of the objects and wherein each of the upstream uncertainty region and the downstream uncertainty region is at a respective outer upstream boundary and outer downstream boundary of the area of interest and each of the upstream uncertainty region and downstream uncertainty region has a width from the respective outer upstream boundary and outer downstream boundary to a respective inner upstream boundary and inner downstream boundary, in the elongated direction, of at least one half the difference.

7. The system as set forth in claim 6 wherein the counter is further constructed and arranged (A) to count only the objects located fully in the uncertainty region and (B) to define a counting region between each of the inner upstream boundary of the upstream uncertainty region and the inner downstream boundary of the downstream uncertainty region, and further including:

(a) means for counting the objects located in the upstream uncertainty region when the objects therein are located only partially in the left uncertainty region and that have respective predetermined origin points thereof located in the counting region;

(b) means for defining a holdover set including the objects having the respective predetermined origin points thereof located in the upstream uncertainty region, the holdover set including, for each of the objects located therein, a respective location thereof, and determining whether or not each of the objects included in the holdover set is respectively counted in association with the acquired image;

(c) means for counting the objects in the downstream uncertainty region if (I) the holdover set for a previous acquired image includes the objects and (II) the included objects are not counted in association with the previous acquired image.

8. The system as set forth in claim 1 wherein the counter is further constructed and arranged to count only the objects located fully in the downstream overlapping section and including means for selectively counting the objects located in the downstream uncertainty region of the acquired image if the objects are located in an upstream uncertainty region of a directly previous acquired image and not counted in association with the previous acquired image.

9. A method for counting objects on an elongated surface comprising:

defining, with a camera, an area of interest with respect to an elongated direction of the surface, the camera and the surface being in relative movement with respect to each other in an upstream to downstream direction;

identifying, with a vision system, locations of objects occurring within the area of interest within a predetermined interval;

triggering the acquiring of an image by the vision system within the area of interest each time the surface moves a predetermined distance along the elongated direction, wherein each acquired image overlaps an adjacent acquired image by a predetermined overlap distance along the elongated direction, thereby defining, in each area of interest, a central section, an upstream overlapping section upstream of the central section, and a downstream overlapping section downstream of the central section; and

counting (a) objects identified that are located at least partially in the central section and (b) objects identified that are located fully within only one of either the upstream overlapping section and the downstream overlapping section in the area of interest for each acquired image.

10. The method as set forth in claim 9 further comprising moving the surface and generating, based upon movement of the surface for the predetermined distance at the predetermined times, a signal, and wherein the triggering step occurs in response to the signal.

11. The method as set forth in claim 10 further comprising operating a shutter assembly, operatively connected to the camera, to selectively acquire an image at the predetermined times based upon the triggering step.

12. The method as set forth in claim 10 further comprising illuminating the area of interest with a flash to thereby acquire an image by the camera at the predetermined times in response to the signal.

13. The method as set forth in that claim 9 wherein the counting step includes defining an upstream uncertainty region overlaid on the upstream overlapping region and a down stream uncertainty region overlaid on the downstream overlapping region, based upon a predetermined range of expected erroneous movement of the objects in the elongated direction between each acquired image in the area of interest.

14. The method as set forth in claim 13 further comprising defining each of the upstream uncertainty region and the downstream uncertainty region as a distance in the elongated direction that is a difference between a maximum expected erroneous movement of the objects and a minimum expected erroneous movement of the objects and so that each of the upstream uncertainty region and the downstream uncertainty region is at a respective outer upstream boundary and outer downstream boundary of the area of interest and each of the upstream uncertainty region and downstream uncertainty region has a width from the respective outer upstream boundary and outer downstream boundary to a respective inner upstream boundary and inner downstream boundary, in the elongated direction, of at least one half the difference.

15. The method as set forth in claim 14 wherein the counting step includes (A) counting only the objects located fully in the downstream overlapping section and (B) defining a counting region between each of the inner upstream boundary of the upstream uncertainty region and the inner downstream boundary of the downstream uncertainty region, and further including:

(a) counting the objects located in the upstream uncertainty region when the objects therein are located only partially in the upstream uncertainty region and that have respective predetermined origin points thereof located in the counting region;

(b) defining a holdover set including the objects having the respective predetermined origin points thereof located in the upstream uncertainty region, the holdover set including, for each of the objects located therein, a respective location thereof, and determining whether or not each of the objects included in the holdover set is respectively counted in association with the acquired image;

(c) counting the objects in the downstream uncertainty region if (I) the holdover set for a previous acquired image includes the objects and (II) the included objects are not counted in association with the previous acquired image.

16. The method as set forth in claim 9 wherein the counting step comprises counting only the objects located fully in the downstream overlapping section and including selectively counting the objects located in the downstream uncertainty region of the acquired image if the objects are located in an upstream uncertainty region of a directly previous acquired image and not counted in association with the previous acquired image.

17. A method for counting objects experiencing relative movement with respect to an area of interest in an upstream-to-downstream direction, wherein images of the objects are acquired at respective predetermined time intervals based upon the relative movement, the method comprising:

(A) acquiring images in succession within the area of interest, wherein each of the images includes an downstream overlapping region with respect to a directly preceding of the images and an upstream overlapping region with respect to a directly succeeding of the images;

(B) counting all the objects that have a respective origin point located between each of the downstream overlapping region and the upstream overlapping region for each of the images;

(C) counting all the objects that have the respective origin point thereof located in only one of either the upstream overlapping region and the downstream overlapping region for each of the images.

18. The method as set forth in claim 17 further comprising defining an upstream uncertainty region overlaid on the the upstream overlapping region and a downstream uncertainty region overlaid on the downstream overlapping region, based upon a predetermined range of expected erroneous movement of the objects in the upstream-to downstream direction between each acquired image in the area of interest.

19. The method as set forth in claim 18 further comprising defining each of the upstream uncertainty region and the downstream uncertainty region as a distance in the elongated direction that is a difference between a maximum expected erroneous movement of the objects and a minimum expected erroneous movement of the objects and so that each of the upstream uncertainty region and the downstream uncertainty region is at a respective outer upstream and outer downstream boundary of the area of interest and each of the upstream uncertainty region and downstream uncertainty region has a width from the respective outer upstream boundary and outer downstream boundary to a respective inner upstream boundary and inner downstream boundary, in the upstream-to-downstream direction, of at least one half the difference.

20. The method as set forth in claim 19 wherein the counting step (B) includes counting only the objects located fully in the downstream uncertsainty region and defining a counting region between each of the inner upstream boundary of the upstream uncertainty region and the inner downstream boundary of the downstream uncertainty region, and further including:

(a) counting the objects located in the upstream uncertainty region when the objects therein are located only partially in the upstream uncertainty region and that have respective predetermined origin points thereof located in the counting region;

(b) defining a holdover set including the objects having the respective predetermined origin points thereof located in the upstream uncertainty region, the holdover set including, for each of the objects located therein, a respective location thereof, and determining whether or not each of the objects included in the holdover set is respectively counted in association with the acquired image;

(c) counting the objects in the downstream uncertainty region if (I) the holdover set for a previous acquired image includes the objects and (II) the included objects are not counted in association with the previous acquired image.

21. A method for counting objects experiencing relative movement with respect to an area of interest in an upstream-to-downstream direction, wherein images of the objects are acquired at respective predetermined time intervals based upon the relative movement, the method comprising:

(A) acquiring images in succession within the area of interest, wherein each of the images includes a downstream overlapping region with respect to a directly preceding of the images and an upstream overlapping region with respect to a directly succeeding of the images;

(B) defining each of an upstream uncertainty region and a downstream uncertainty region as a distance in the elongated direction that is a difference between a maximum expected erroneous movement of the objects and a minimum expected erroneous movement of the objects, the upstream uncertainty region being bounded by a respective outer upstream boundary and downstream uncertainty region being bounded by an adjacent outer downstream boundary, each of the upstream uncertainty region and the downstream uncertainty region having a width from the respective outer upstream boundary and outer downstream boundary to a respective inner upstream boundary and inner downstream boundary, in the elongated direction, of at least one half the difference;

(C) counting each of the objects having a respective origin point located between the upstream uncertainty region and the downstream uncertainty region;

(D) defining, for each of the objects having a respective origin point located in the upstream uncertainty region, a holdover set that stores the objects respective location, and assigning to the holdover set a respective count indicator for each of the objects that is counted in the upstream uncertainty region; and

(E) counting each of the objects having a respective origin point located in the downstream uncertainty region that correspond to respective objects in the holdover set for the directly preceding of the images for which a count indicator is absent.

22. The method as set forth in claim 21 wherein the difference comprises approximately one-half a width of one of the objects.