EP0115546B1 - Automatic print quality assessment and adjustment of a printer - Google Patents

Description

The invention relates to a method for regulating the print quality of characters printed out by printers.

In the article "The monitoring of a rotogravure printing press with the help of a process computer (4th IFAC / IFIP International Conference on Digital Computer Applications to Process Control, Zurich, Switzerland, March 19-22, 1974, Part II, pages 477-487, Springer-Verlag , Berlin, DE.) While it is mentioned at the beginning that the optical color density is a criterion for assessing the print quality, in particular only one color density controller for the optical density is described.

However, this ignores the issue of print quality, since e.g. a low color density does not necessarily have to represent a poorer print quality in the subjective sense. In this document it is claimed that the optical color density would be a criterion for assessing the print quality; however, it is not the print quality that appears as the control variable, but only the optical density.

The print quality is a subjective variable and can be determined empirically from a statistically relevant number of individual assessments of a number of equally good or of gradually or continuously increasing (or also decreasing) templates by a likewise statistically relevant number of observers. In this way, as detailed tests have shown in good agreement with one another, one can actually come to a reasonably reliable statement about the print quality, especially if one categorizes the results into a number of graded quality classes, for example six quality classes.

Above all, the new method is intended to constantly monitor and, if necessary, correct the print image of a line printer, raster printer or the like. There are a number of parameters for print quality, which have different meanings or importance depending on the printing process and printer. With partially cut letters, for example, the control of the hammer flight time is incorrect or the type carrier moves too quickly or too slowly, the hammer is triggered too early or too late. Unclear or too strong pressure can indicate that the stop of the print hammer or printhead is too strong or too weak, for example. This can also be an indication that the ink ribbon or blanket is brand new or, on the contrary, has been used up so much that the structure of the ink ribbon or ink can be seen in the print. To overcome these difficulties, for example, a medium print quality will be specified, which still has to be described as very good, and attempts will be made to adjust to this medium print quality, which is certainly the optimum for long-term operation.

(R_i = Reflexionswert des i-ten Bildelementes
R_B = Reflexionswert des Hintergrundes)
ermittelt und aus allen Bildkontrastsignalwerten des Zeichens der arithmetische Mittelwert zu PCS bestimmt wird, daß aus den Bildkontrastsignalwerten unter Bildung der Standardabweichung

und Bestimmung eines Druckqualitätsfaktors

ein objektiver Druckqualitätswert zu

${\text{DQ}}_{\text{o}}{\text{≧ a}}_{\text{o}}\text{+ a₁ lnx₂}$

(a_o ≧ schlechteste Wertzahl)
a₁ wird so gewählt (als negativer Wert), daß sich bei einer besten Druckqualität für DQ_o die beste Wertzahl ergibt)
bestimmt wird, und daß bei einer Abweichung des objektiven Druckqualitätswertes DQ_o vom vorgegebenen subjektiven Druckqualitätswert DQ_s ein Nachregeln des DQ_o- auf den DQ_s-Wert dadurch erfolgt, daß die die objektive Druckqualität des Druckers beeinflussenden Größen entsprechend verändert werden.This is achieved according to the invention in that the subjective print quality is rated by human evaluators (jury) on the basis of their subjective impression with value numbers (for example 1 = best; 1.5; 2; ... 6 = worst), that for a special printer the required subjective print quality predetermined by a certain value number DQs is stored in it, that in a manner known per se for a printed character by opto-electronic scanning each an image contrast signal value of its picture elements i (i = 1 to n) along the skeleton line of the character and the background

(R _i = reflection value of the i-th picture element
R _B = reflection value of the background)
determined and the arithmetic mean value from all image contrast signal values of the character PCS it is determined that from the image contrast signal values to form the standard deviation

and determining a print quality factor

an objective print quality value

${\text{DQ}}_{\text{O}}{\text{≧ a}}_{\text{O}}\text{+ a₁ lnx₂}$

(a _o ≧ worst value number)
a₁ is chosen (as a negative value), so that the best value for DQ _o results in the best print quality)
is determined, and in that when a deviation of the objective print quality value DQ _o from the predetermined subjective print quality value _s DQ readjustment of the DQ _o - on the DQ _s value is effected in that the objective print quality of the printer variables influencing be changed accordingly.

Further embodiments of the invention can be found in the subclaim.

The invention will now be described in more detail using an exemplary embodiment in conjunction with the accompanying drawing, to which reference is made in the text.

To determine a subjective print quality factor DQ _s , a representative set of print templates, which may be of the same and / or different quality, is submitted to an equally representative group of evaluators for assessment. The evaluators are asked to rate the quality with a factor between 1 and 6 with the possibility of intermediate values 1.5, 2.5, etc. These assessments are evaluated using statistical procedures. This results in a formula which allows the assessments to be mathematically predicted with sufficient accuracy if the corresponding print quality parameters have been determined by measurement. One of these parameters is the Print Contrast Signal (PCS). This is determined along the skeleton line in accordance with DIN 66 223 and also over the template background by scanning using the above-mentioned scanning method. This parameter results from the different reflection values of the skeleton line to the background.

• 1. PCS (Mittelwert),
• 2. PCS Standardabweichung oder σ_PCS und
• 3. PCS (Fehlstellen).

This scan gives a different value depending on the quality of the characters scanned. The variation of this parameter on the print template can be described in three expressions:

• 1. PCS (Average),
• 2. PCS standard deviation or σ _PCS and
• 3rd PCS (Missing parts).

so ergibt sich eine hinreichend genaue Beschreibung des Beurteilungsverhaltens dieser Gruppe von menschlichen Bewertern bezüglich der Druckqualität der Vorlage. Dies zeigt die Zeichnung. Dort sind auf der Ordinate die Beurteilungswerte abgetragen, während der Druckqualitätsfaktor x₂ auf der Abszisse angegeben ist. Dabei ist unterstellt, daß die Gesamtheit der bewerteten Druckvorlagen durch diese Untermenge mit einer im Rahmen liegenden Genauigkeit hinreichend repräsentiert wird. Außerdem wird vorausgesetzt, daß die Standardabweichung einen von Null verschiedenen Wert aufweist. Dies ist im Normalfall immer der Fall.Accordingly, the value (1- PCS defects ) can also be called the completeness of a character as an average value. To summarize these terms in the following formula:

this results in a sufficiently precise description of the assessment behavior of this group of human evaluators with regard to the print quality of the template. This is shown in the drawing. There, the evaluation values are plotted on the ordinate, while the print quality factor x 2 is indicated on the abscissa. It is assumed that the totality of the evaluated print templates is adequately represented by this subset with an accuracy that is within the framework. It is also assumed that the standard deviation has a non-zero value. This is usually the case.

Dies gilt immer entlang der Skelettlinie.This x₂ value can be linearized using the natural log and then you get a linear scale of 1-6. An extrapolation to values less than 1 and greater than 6 is then possible. In most cases, the completeness of the characters is ensured; so that the formula for x₂ simplifies too

This always applies along the skeleton line.

In the unlikely event that the PCS standard deviation is zero, this formula is of course not applicable. The reason for this is that the true function of x _{2 is, as} expected, logarithmic and that the formula for classifying the value of the print quality is only an approximation from a series development, so that the formula does not become too abstract. For examinations with very good print quality or font quality, x₂ should be replaced by ln x₂. This gives a linear dependency.

This formula can be used for all print templates in which the print contrast, the fluctuations in the print contrast along the skeleton line, the fluctuations in the print contrast between different characters and the completeness of the individual characters as well as the edge sharpness of the individual characters are the essential components for the print quality factor.

The control process runs in such a way that the parameters accessible in the printer of a control system are stored in a list memory and that the control variables effect the readjustments of the individual parameters, which are subordinate to their efficiency, one after the other until the optimum print quality is achieved by constant comparison with the stored value is approximated or reached.

In principle, the sampled values can be obtained either from the characters of a print line itself or from special characters to be printed outside the typeface, which are particularly suitable for precise adjustment to the optimal print quality. These patterns can be, for example, dot grids, line grids, or grids with only one dot or grids with concentric circles or the like. In order to carry out the method, the scanned values are expediently recorded with a photoelectronic scanner mounted directly above the print line in a very small space and the print quality factor DQ is derived using a simple arithmetic circuit.

Claims (2)

1. Method of controlling the print quality of characters printed on printers, characterized in that values (e.g. 1 = best; 1,5; 2; ... 6 = worst) are assigned to the subjective print quality by human evaluators (jury) on the basis of their subjective impression, that the subjective print quality required for a particular printer is defined by a predetermined value DQ_s and stored therein, that, in a manner known per se, an image contrast signal value

   

   (R_i = reflection value of the ith picture element
   R_B = reflection value of the background)
   
   is determined for a printed character by optoelectronically scanning each of its picture elements i (i = 1 to n) along the skeleton line of the character and the background and the arithmetic mean value of PCS is derived from the various image contrast signal values of the character, that on the basis of the image contrast signal values, forming a standard deviation of

   

   and defining a print quality factor

   

   an objective print quality value is determined
   
   ${\text{DQ}}_{\text{o}}{\text{= a}}_{\text{o}}\text{+ a₁ lnx₂}$

   

   
   
   (a_o ≧ worst value)
   a₁ (as a negative value) being selected such that the best value is obtained at the maximum print quality DQ_o),
   
   and that if the objective print quality value DQ_o deviates from a predetermined subjective print quality value DQ_s, the DQ_o value is adjusted to DQ_s by modifying the variables influencing the objective print quality of the printer.
2. Method as claimed in claim 1,
   characterized in that in the case of voids in the skeleton line of the characters to be evaluated, a standardized average PCS void value of
   
   PCS _void
   
   is calulated proportional to the missing picture elements relative to the totality of the picture elements, and that the print quality factor is

   

   taking account of the voids.

Images