Algorithm for gray scale analysis especially of fruit or nuts

• US 5,060,290 A
• Filed: 09/05/1989
• Issued: 10/22/1991
• Est. Priority Date: 09/05/1989
• Status: Expired due to Fees

• Current Assignee: Dole Dried Fruit And Nut Company
• Original Assignee: Dole Dried Fruit And Nut Company
• Inventors: Klein, Cynthia M., Kelly, Patrick L.
• Primary Examiner(s): Razavi, Michael
• Assistant Examiner(s): Cammarata, Michael
• Application Number:
  US07/402,932
• Time in Patent Office: 777 Days
• Field of Search: 382/1, 382/6, 382/8, 382/18, 382/51, 382/42, 358/106, 364/555, 356/430, 356/237
• US Class Current: 382/110
• CPC Class Codes: B07C 5/342 according to optical properties, e.g. colour according to radiation transmittivity B07C5/3416
• CPC Class Codes: B07C 5/368 actuated independently
• CPC Class Codes: G06T 2207/30128 Food products
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/0006 using a design-rule based approach

Abstract

A technique of characterizing histograms using unique serial code words for discrimination between look alike acceptable and unacceptable items. The histogram is processed by normalizing the captured histogram image pixels preferably to the largest pixel size image normally encountered. Thereafter the histogram is attributed. The preferred embodiment related to a gray scale histogram (as distinguished from a histogram relating to area, length, width, perimeter, or aspect ratio) and successively attributes each histogram into peak location, range of histogram, peak value, right of peak value, and a selected gray scale count. For each attribute, an arbitrary range of codes is generated to break the classification of each attribute into manageable two-digit code categories. These two-digit code categories are typically written to a serial look-up table opposite the actual histogram value and extracted from the look-up table in real time to create two-digit code words. The two-digit code words are in turn concatenated in real time to create a unique serial word out of the successive codes from each attribute of each individual histogram. In the case of product acceptance and rejection, the unique serial code words of acceptable product is nonredundantly written to a library with word recognition indicting acceptable product. In grading product, the unique serial code words are provided to identify the product or defect which can thereafter be summed or otherwise processed for generating the desired product grades.

Claims

1. A process for the characterization of histogram data comprising the steps of:

providing a histogram array having discrete bins for receiving like data, said discrete bins distributed over a discrete range of bin values for receiving individual data counts within those bin values and accumulating those data counts for creation of a histogram;

selecting at least three attributes from said histogram from an attribute category including, but not limited to, bin range, peak bin, peak bin value, bin value to side of peak, and selected bin value;

dividing each said selected attribute into a plurality of arbitrary range categories assigning a unique arbitrary code to each said range category, said code constituting a portion of a word for the characterization of said histogram;

determining for said selected attributes from said histogram attributes the arbitrary codes for each of said selected attributes; and

,

concatenating said arbitrary codes to form unique serial words for describing said histograms.

2. The process of claim 1 and wherein said step of selecting at least three attributes includes bin range, peak bin, peak bin value, bin value to side of peak, and selected bin value.

3. The invention of claim 2 and wherein one of said selected attributes includes said bin value to the right side of peak.

4. The invention of claim 1 and including the step of normalizing said histogram array before selecting said attributes.

5. The invention of claim 1 and wherein said histogram is a gray scale histogram.

6. The process of accumulating unique serial code words for the sorting of acceptable product comprising the steps of:

providing a video camera for observing a stream of passing product and generating signals indicative of said product for producing product histograms;

for each discrete product item passing said video camera providing a histogram array having discrete bins for receiving like data, said discrete bins distributed over a discrete range of bin values for receiving individual data counts within those bin values and accumulating those data counts for creation of a histogram;

selecting at least three attributes from said histogram from an attribute category including, but not limited to, bin range, peak bin, peak bin value, bin value to side of peak and selected bin value;

dividing each selected attribute into a plurality of arbitrary range categories;

assigning a unique arbitrary code word to each range category, said code word constituting a portion of a unique serial code word for the characterization of said histogram;

determining for said selected attributes from said histogram attributes the arbitrary codes for each of said selected attributes as directed to a particular histogram;

concatenating said arbitrary codes to form a unique serial code for describing said histogram;

writing said code words to a memory;

and passing acceptable product by said video camera to generate signals indicative of said acceptable product for producing a library of said unique serial code words.

7. The process of claim 6 including the step of nonredundantly writing said unique serial code words to a library;

and

passing said acceptable product by said video camera until said unique serial code words produced by said signals are all redundant whereby said library of unique serial code words for said acceptable product is created.

8. The process of claim 6 and wherein said histogram is a gray scale histogram and said histogram is normalized before said determining step.

9. A process for the classification of acceptable product from look alike unacceptable product comprising in combination;

passing a stream of acceptable and unacceptable product by apparatus including a video camera for creating with respect to said stream a series of histograms for each serial item in said stream;

providing a histogram array having discrete bins for receiving like data, said discrete bins distributed over a discrete range of bin values for receiving individual pixel counts within those bin values and accumulating those data counts for creation of a histogram;

selecting at least three attributes from an attribute category including, but not limited to, bin range peak value, peak bin value, bin value to side of peak and selected bin value;

dividing each select attribute range into a plurality of arbitrary categories;

assigning a unique arbitrary code to each said category, said arbitrary code constituting a portion of a unique serial word for the characterization of each said histogram;

determining for said selected attributes from said histogram the arbitrary code words for each of said selected attributes;

concatenating said arbitrary codes to form unique serial words for describing said histograms;

providing a library of unique serial code words for the classification of articles in said passing serial stream;

comparing each said serial code word to said library for determining whether said unique serial code word is written to said library;

classifying each passing article with respect to said stream based on the comparison with said library.

10. The process of claim 9 and wherein said library contains unique serial code words for acceptable product only.

11. The process of claim 9 and wherein said library contains words for acceptable and unacceptable product.

12. The process of claim 9 can and wherein said library contains classifications for said unacceptable product.