Vision recognition system for produce labeling

• US 8,456,646 B2
• Filed: 09/12/2011
• Issued: 06/04/2013
• Est. Priority Date: 09/13/2010
• Status: Expired due to Fees

• Current Assignee: Sinclair Systems International LLC
• Original Assignee: Sinclair Systems International LLC
• Inventors: Cronk, Stephen, Dale, Jason, Holland, Phillip, Hawkes, Richard
• Primary Examiner(s): Ton, Tri T
• Application Number:
  US13/199,876
• Publication Number:
  US 20120069354A1
• Time in Patent Office: 631 Days
• Field of Search: 356601-613, 356 52- 67
• US Class Current: 356/612
• CPC Class Codes: B65C 9/40 Controls; Safety devices
• CPC Class Codes: G01B 11/0608 Height gauges
• CPC Class Codes: G01B 11/2522 the position of the object changing and being recorded

Abstract

A vision recognition system is provided for use with a high speed, automatic produce labeling machine. The system uses laser profiling to direct a sheet of light transversely to the longitudinal axis of a produce feed conveyor. The sheet of light periodically impacts, and generates laser profiles of, the surfaces of the produce items, such as pears, being fed by the conveyor to one or more labeling machines. The laser profiles are used to generate real world (x,y) coordinates of the domes of the incoming produce items, which are passed to the labeling machine or machines. Real world height (or z) coordinates may also be created and passed to the labeler.

Claims

1. A vision recognition method for use with a high speed, automatic produce labeling machine, wherein a conveyor feeds incoming produce items along an axis A-A toward said labeling machine, wherein each produce item has a dome portion of its surface which occurs at the greatest height of said produce item'"'"'s surface above said conveyor, comprising the steps:

generating a sheet of light from a laser source or other high intensity light source,

directing said sheet of light transversely to said axis A-A of said conveyor,

causing said sheet of light to periodically impact and reflect from the surfaces of said produce items to create instantaneous laser profiles of said surfaces,

continuously imaging said laser profiles with a high speed camera locating the dome of each produce item by comparing the relative brightness of the reflections of said sheet of light from different portions of the surface of each produce item, and by identifying the portion of said surface that reflects the most light and shows the greatest relative brightness,

creating real world coordinates (x,y) for the instantaneous location of said domes of said produce items horizontally on said conveyor, and

transmitting said instantaneous real world horizontal (x,y) coordinates periodically to said labeling machine to provide advance data to said labeling machine regarding the real world horizontal location (x,y coordinates) of the dome of each incoming produce item to be labeled.

2. The method of claim 1 comprising the further step:

creating a height or z coordinate for at least some of said domes by comparing the brightness of said reflection of light from said dome with known reference brightness measurements, and

transmitting said height or z coordinates to said labeling machine.

3. The method of claim 2 comprising the further step:

selecting the largest height or z coordinate from a selected number of said height or z coordinates, and

transmitting said largest height or z coordinate to said labeling machine.

4. The method of claim 2 wherein said height or z coordinates are created for all of said domes and transmitted to said labeling machine.

5. The method of claim 1 wherein said different portions of said surface each form topographic rings showing the shape of the upper half of each said produce item.

6. The method of claim 1 comprising the further step:

generating a 2-D map of said produce items from said laser profiles.