Ball point writing quality analysis

• US 5,467,408 A
• Filed: 04/22/1994
• Issued: 11/14/1995
• Est. Priority Date: 04/22/1994
• Status: Expired due to Fees

• Current Assignee: Gillette Company (The Procter & Gamble Company)
• Original Assignee: Gillette Company (The Procter & Gamble Company)
• Inventors: Dowst, William P.
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US08/232,522
• Time in Patent Office: 571 Days
• Field of Search: 382/18, 382/51, 382/53, 382/1, 356/91
• US Class Current: 382/112
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 7/001 using an image reference approach

Abstract

A method of analyzing the writing quality of a ball point pen for ink laydown, blobbing, and degree of inconsistency employs a computer and an image analysis software package capable of assembling a pixel gray level distribution known as a histogram. The distribution of pixel gray value for a writing sample is manipulated and analyzed, the values then being compared to a range of pre-established values.

Claims

1. A method of measuring ball point pen writing quality including the steps of:

preparing ink writing samples from a plurality of ball point pens having acceptable writing quality;

providing a computer imaging system;

transforming each ink writing sample and background into a histogram formed of a plurality of pixels by employing the computer imaging system to assign each pixel a gray value from zero which is pure black to 256 which is pure white and to sum the pixels of every gray value to produce a histogram having the values of zero and 256 along the abscissa and the number of pixels of each gray value along the ordinate, each of said histograms comprising a pair of peaks, the ink peak disposed toward the 0 value indicating the maximum pixel value of the ink and the second peak spaced from the ink peak toward the 256 value indicating the maximum pixel value of the background on which the ink sample is formed;

establishing a plurality of ranges of values derived from the histogram obtained from the acceptable writing sample;

preparing an ink writing sample from a ball point pen of unknown writing quality;

employing said computer imaging system to transform said ink writing sample from said ball point pen of unknown writing quality into a histogram formed of a plurality of pixels, each pixel being assigned a gray value from zero which is pure black to 256 which is pure white and to sum the pixels of every gray value to produce a histogram having values of zero and 256 along the abscissa and the number of pixels of each gray value along the ordinate, each of said histograms comprising a pair of peaks, the ink peak disposed toward the 0 value indicating the maximum pixel value of the ink and the second peak spaced from the ink peak toward the 255 value indicating the maximum pixel value of the background on which the ink sample is formed;

determining the slope of the ink peak toward the zero value side of each histogram by forming a line running from a point about halfway up the ink peak to the point of intersection with the abscissa, and calculating that point along the abscissa where the slope line intersects the abscissa;

determining the number of pixels from that point to the zero point of each histogram and

dividing this number by the total number of pixels in each histogram to determine the fraction of image area covered with blobs of ink material in each histogram; and

comparing the values derived from the histogram of the ball point pen of unknown writing quality with the range of values established from the acceptable writing samples to determine the acceptability of the ball point pen of unknown writing quality.

2. The method of claim 1 which further includes the steps of determining the median gray value for each of said samples by locating the maximum value of the ink peak along the abscissa of each of the histograms to determine ink laydown.

3. The method of claim 1 which further includes the steps of determining the median gray value by calculating, for example, the numerical slope of each of the histograms at all gray value points, assigning a value of one to that gray value point if the slope is greater than one and assigning a value of minus one to that gray level point if the slope is less than zero, and placing the positive ones and negative ones in a separate array;

identifying the median gray value as the first positive to negative transition occurs in said separate array, and

comparing the values to determine acceptable ink laydown.

4. A method as set forth in claim 1 including the steps of taking a ratio between the value of the height of the ink peak and the distance from the height of the ink peak to the top of the trough between said ink peaks in each of said histograms to measure consistency of each of the writing samples.

5. The method of claim 1 wherein prior to comparing each said histogram with a writing sample histogram, each of the histograms is smoothed by averaging the pixel values over a plurality of gray values greater and less than the gray value at each histogram gray value, and thereafter producing a new histogram array similar to the original by employing those average values obtained, whereby each said newly produced histogram provides a histogram curve having improved definition.