Automatic digital object counting and verification system and associated method

• US 7,570,786 B2
• Filed: 08/30/2004
• Issued: 08/04/2009
• Est. Priority Date: 08/30/2004
• Status: Expired due to Fees

• Current Assignee: Antoun Ateya
• Original Assignee: Antoun Ateya
• Inventors: Ateya, Antoun
• Primary Examiner(s): Dang; Duy M
• Application Number:
  US10/929,293
• Publication Number:
  US 20060045323A1
• Time in Patent Office: 1,800 Days
• Field of Search: 382/103, 382/107, 382/143, 382/149, 382/190, 382/192, 532/44, 221/92, 221/123, 221/129, 702/128
• US Class Current: 382/103
• CPC Class Codes: G06M 1/101 by electro-optical means
• CPC Class Codes: G06M 11/00 Counting of objects distributed at random, e.g. on a surface
• CPC Class Codes: G06V 10/141 Control of illumination
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 20/66 Trinkets, e.g. shirt buttons or jewellery items recognising microscopic objects G06V20/69

Abstract

A system for automatic counting of non-overlapping objects irrespective of shape, size, and color is provided by imaging and computer subsystems. Objects to be counted are placed on a transparent surface disposed on a diffusing surface uniformly irradiated by electromagnetic radiation sources. Low intensity object shadows and high intensity object background regions are digitally imaged. The digital image is converted by a computing unit to a binary image and subjected to the Distance Transform to determine a count of the objects. Object identification verification is provided by comparing identification information obtained from a bar code associated with a supply container of the objects and identification information obtained from a digitally imaged written request.

Claims

1. An object counting system for counting non-overlapping objects, said system comprising:

at least one radiation source providing electromagnetic radiation;

a digital imaging unit for providing a digital image by detecting at least part of said electromagnetic radiation, at least part of said digital image being associated with a plurality of objects to be counted, said plurality of objects being non-overlapping;

a radiation diffusing surface positioned between said digital imaging unit and said at least one radiation source, said radiation diffusing surface being exposed to said electromagnetic radiation substantially uniformly;

an object holding surface substantially transparent to said electromagnetic radiation, said object holding surface supporting said plurality of objects placed thereon to be counted automatically, and said object holding surface being supported by said radiation diffusing surface; and

a computing unit interfaced with said digital imaging unit for receiving said digital image, and including software for converting said digital image to a binary digital image, and for analyzing said binary digital image to automatically count said plurality of objects, wherein said software for analyzing said binary digital image employs a Distance Transform to segment clusters of objects of said plurality of objects and to count said plurality of objects.

2. The system of claim 1, wherein said electromagnetic radiation includes at least a first portion and a second portion,

wherein said first portion interacts with said plurality of objects, thereby generating regions (shadow regions) associated with each object of said plurality of objects, said shadow regions being directed toward said digital imaging unit, and said shadow regions being associated with one or more intensity levels of said electromagnetic radiation,

wherein said second portion passes through regions (background regions) of said object holding surface to said digital imaging unit without interacting with said plurality of objects, said background regions being associated with one or more intensity levels of said electromagnetic radiation,

wherein said one or more intensity levels associated with said shadow regions are lower than said one or more intensity levels associated with said background regions, and

wherein said digital image includes a digital representation of said one or more intensity levels associated with said shadow regions and said background regions.

3. The system of claim 2, wherein said binary digital image includes a first binary state and a second binary state, and

wherein said converting to said binary digital image includes converting said one or more intensity levels associated with said shadow regions to said first binary state and converting said one or more intensity levels associated with said background regions to said second binary state.

4. The system of claim 1, wherein said plurality of objects are counted irrespective of attributes of said objects, said attributes including at least shape, color and size of said objects.

5. The system of claim 1, wherein said plurality of objects are counted without employing a priori knowledge of attributes of said plurality of objects, and without deriving attributes from a sample of said plurality of objects.

6. The system of claim 1, wherein said automatic count of said plurality of objects is initiated by said plurality of objects being placed on said object holding surface, and said object holding surface being placed on said diffusing surface.

7. The system of claim 1, wherein said at least one radiation source includes a plurality of electromagnetic radiation sources.

8. The system of claim 7, wherein said plurality of electromagnetic radiation sources includes two cold cathode fluorescent light (CCFL) sources.

9. The system of claim 1, wherein said software for analyzing said binary digital image determines approximate sizes of said plurality of objects, a mean size of the determined sizes, and a standard deviation associated with the determined sizes to facilitate identifying fragments of objects of said plurality of objects, and identifying foreign objects counted by said software, wherein said foreign objects are mistakenly included in said plurality of objects to be counted.

10. The system of claim 1, further comprising a radiation integrating cavity disposed under said radiation diffusing surface,

wherein said electromagnetic radiation sources reside in said cavity,

wherein said radiation integrating cavity includes interior surfaces to provide diffusive and multiple reflections of said electromagnetic radiation within said cavity, and

wherein said reflections facilitate said radiation diffusing surface being substantially uniformly exposed to said electromagnetic radiation.

11. The system of claim 1, wherein said digital imaging unit is selected from a group consisting of a digital camera and a linear lens array aligned with a linear photodiode array.

12. The system of claim 1, wherein said object holding surface comprises a tray, said tray being removable from said radiation diffusing surface, and wherein said radiation sources provide radiation exposure to said radiation diffusing surface by continuous exposure or by flash exposure.

14. The system of claim 1, wherein said plurality of objects are removed from a supply container prior to being placed on said object holding surface, and wherein said system further comprises an object identification verification subsystem, said verification subsystem comprising:

a removable transparent written request holder for placement on said radiation diffusing surface;

a written request for placement in said written request holder, said written request including information (first identification information) identifying objects whose count is requested by said written request;

an illumination source for illuminating said written request placed in said written request holder, thereby reflecting light from said written request to said digital imaging unit to facilitate generating a digital image of said written request; and

a bar code scanner for scanning a bar code associated with said supply container, said bar code including information (second identification information) identifying said plurality of objects removed from said supply container,

wherein said computing unit receives and analyzes said digital image of said written request to extract said first identification information,

wherein said computing unit receives said second identification information, and

wherein said software compares said first identification information to said second identification information, thereby verifying if said plurality of objects removed from said supply container and placed on said object holding surface match said objects whose count is requested by said written request.

13. An object counting system for counting non-overlapping objects, said system comprising:

at least one radiation source providing electromagnetic radiation;

a digital imaging unit for providing a digital image by detecting at least part of said electromagnetic radiation, at least part of said digital image being associated with a plurality of objects to be counted, said plurality of objects being non-overlapping;

a radiation diffusing surface positioned between said digital imaging unit and said at least one radiation source, said radiation diffusing surface being exposed to said electromagnetic radiation substantially uniformly;

an object holding surface substantially transparent to said electromagnetic radiation, said object holding surface supporting said plurality of objects placed thereon to be counted automatically, and said object holding surface being supported by said radiation diffusing surface;

a computing unit interfaced with said digital imaging unit for receiving said digital image, and including software for converting said digital image to a binary digital image, and for analyzing said binary digital image to automatically count said plurality of objects

wherein said motion of said plurality of trays is provided by an apparatus selected from a group consisting of a moving conveyor and a rotary mechanism,

wherein said motion of said plurality of trays is selected from a group consisting of stop-and-go motion and continuous uniform motion,

wherein radiation sources provide radiation exposure to said radiation diffusing surface by continuous exposure,

if said digital imaging unit includes lens and photodiode arrays and said motion of said trays is said continuous uniform motion, or

if said digital imaging unit is a CCD-based digital camera and said motion of said trays is said stop-and-go motion, and

wherein radiation sources provide radiation exposure to said radiation diffusing surface by flash exposure if said digital imaging unit is a CCD-based digital camera and said motion of said trays is said continuous uniform motion.

15. A method of counting non-overlapping objects, said method comprising:

placing a plurality of non-overlapping objects on a substantially transparent object holding surface to automatically count said objects, said object holding surface supported by a radiation diffusing surface disposed between a digital imaging unit and at least one radiation source;

providing, by said at least one radiation source, electromagnetic radiation to expose said radiation diffusing surface substantially uniformly;

detecting at least part of said electromagnetic radiation by a digital imaging unit to provide a digital image, at least part of said digital image being associated with said plurality of objects to be counted;

receiving, by a computing unit interfaced with said digital imaging unit, said digital image;

converting, by software included in said computing unit, said digital image to a binary digital image; and

analyzing, by said software, said binary digital image to automatically count said plurality of objects, wherein said analyzing by said software employs a Distance Transform to segment clusters of objects of said plurality of objects and to count said plurality of objects.

16. The method of claim 15, wherein said electromagnetic radiation includes at least a first portion of electromagnetic radiation and a second portion of electromagnetic radiation,

wherein said first portion interacts with said plurality of objects, thereby generating regions (shadow regions) associated with each object of said plurality of objects, said shadow regions being directed toward said digital imaging unit, and said shadow regions being associated with one or more intensity levels of said electromagnetic radiation,

wherein said second portion passes through regions (background regions) of said object holding surface to said digital imaging unit without interacting with said plurality of objects, said background regions being associated with one or more intensity levels of said electromagnetic radiation,

wherein said one or more intensity levels associated with said shadow regions are lower than said one or more intensity levels associated with said background regions, and

wherein said digital image includes a digital representation of said one or more intensity levels associated with said shadow regions and said background regions.

17. The method of claim 16, wherein said binary digital image includes a first binary state and a second binary state, and

wherein said converting to said binary digital image includes converting said one or more intensity levels associated with said shadow regions to said first binary state and converting said one or more intensity levels associated with said background regions to said second binary state.

18. The method of claim 15 wherein said plurality of objects are counted irrespective of attributes of said objects, said attributes including at least shape, color and size of said objects.

19. The method of claim 15, wherein said plurality of objects are counted without employing a priori knowledge of attributes of said plurality of objects, and without deriving attributes from a sample of said plurality of objects.

20. The method of claim 15, wherein said automatic count of said plurality of objects is initiated by said placing said plurality of objects on said object holding surface, and placing said object holding surface on said diffusing surface.

21. The method of claim 15, wherein said at least one radiation source includes a plurality of electromagnetic radiation sources.

22. The method of claim 21, wherein said plurality of electromagnetic radiation sources includes two cold cathode fluorescent light (CCFL) sources.

23. The method of claim 15, wherein said software for analyzing said binary digital image determines approximate sizes of said plurality of objects, a mean size of the determined sizes, and a standard deviation associated with the determined sizes to facilitate identifying fragments of objects of said plurality of objects, and identifying foreign objects counted by said software, wherein said foreign objects are mistakenly included in said plurality of objects to be counted.

24. The method of claim 15, wherein providing said electromagnetic radiation further comprises providing said electromagnetic radiation in a radiation integrating cavity disposed under said radiation diffusing surface,

wherein said electromagnetic radiation sources reside in said cavity,

wherein said radiation integrating cavity includes interior surfaces to provide diffusive and multiple reflections of said electromagnetic radiation within said cavity, and

wherein said reflections facilitate said radiation diffusing surface being substantially uniformly exposed to said electromagnetic radiation.

25. The method of claim 15, wherein said detecting to provide a digital image further comprises detecting by a digital imaging unit selected from a group consisting of a digital camera and a linear lens array aligned with a linear photodiode array.

26. The method of claim 15, wherein said placing further comprises placing said plurality of objects on tray, said tray including said object holding surface, said tray being removable from said radiation diffusing surface, and wherein said radiation sources provide radiation exposure to said radiation diffusing surface by continuous exposure or by flash exposure.

28. The method of claim 15, further comprising:

removing said plurality of objects from a supply container prior to being placed on said object holding surface; and

verifying identifications of a plurality of objects requested and said plurality of objects counted, wherein said verifying comprises;

placing a written request in a removable transparent written request holder, said written request including information (first identification information) identifying objects whose count is requested by said written request;

lighting, by an illumination source, said written request placed in said written request holder, thereby reflecting light from said written request to said digital imaging unit;

generating, by said digital imaging unit, a digital image of said written request;

scanning, by a bar code scanner, a bar code associated with said supply container, said bar code including information (second identification information) identifying said plurality of objects removed from said supply container;

receiving and analyzing, by said computing unit, said digital image of said written request to extract said first identification information;

receiving, by said computing unit, said second identification information; and

comparing, by said software, said first identification information to said second identification information, thereby verifying if said plurality of objects removed from said supply container and placed on said object holding surface match said objects whose count is requested by said written request.

27. A method of counting non-overlapping objects, said method comprising:

placing a plurality of non-overlapping objects on a substantially transparent object holding surface to automatically count said objects, said object holding surface supported by a radiation diffusing surface disposed between a digital imaging unit and at least one radiation source;

providing, by said at least one radiation source, electromagnetic radiation to expose said radiation diffusing surface substantially uniformly;

detecting at least part of said electromagnetic radiation by a digital imaging unit to provide a digital image, at least part of said digital image being associated with said plurality of objects to be counted;

receiving, by a computing unit interfaced with said digital imaging unit, said digital image;

converting, by software included in said computing unit, said digital image to a binary digital image;

analyzing, by said software, said binary digital image to automatically count said plurality of objects,

wherein said motion of said plurality of trays is provided by an apparatus selected from a group consisting of a moving conveyor and a rotary mechanism,

wherein said motion of said plurality of trays is selected from a group consisting of stop-and-go motion and continuous uniform motion,

wherein said radiation sources provide radiation exposure to said radiation diffusing surface by continuous exposure,

if said digital imaging unit includes lens and photodiode arrays and said motion of said trays is said continuous uniform motion, or

if said digital imaging unit is a CCD-based digital camera and said motion of said trays is said stop-and-go motion, and

wherein said radiation sources provide radiation exposure to said radiation diffusing surface by flash exposure if said digital imaging unit is a CCD-based digital camera and said motion of said trays is said continuous uniform motion.