A non-proprietary image file header label must be used.

If a proprietary image file label is used, a bridge to a non-proprietary image file header label or a detailed definition of image file header label structure must be provided.

13 TAC 6.96 (f) (1)  

System hardware and/or software must provide a quality assurance capability that verifies information written to the storage media.

Provide written instruction and training for staff to use and verify.

13 TAC 6.96 (f) (2)  

Scanned image quality must be evaluated according to the standard procedures in ANSI/AIIM MS44*.

Detailed in the table below for ANSI/AIIM MS-44-1988 (R1993) standard.

13 TAC 6.96 (f) (3)  

A visual quality control evaluation must be performed for each scanned image and related index data.

Detailed in the table below for ANSI/AIIM MS44-1988 (R 1993) standard.

13 TAC 6.96 (f) (4)  

A scanning resolution with a minimum of 200 Dots per image (DPI) is required for recording documents that contain no type font smaller than 6 point.  For documents with a type font smaller than 6 point, scanning resolution must be adequate to ensure that no information is lost.

Detailed in the table below for ANSI/AIIM MS44-1998 (R1993) standard.

13 TAC 6.96 (f) (5-6)  

The selected scanning resolution must be validated with tests on actual documents.

Detailed in the table below for ANSI/AIIM MS44-1998 (R1993) standard.

13 TAC 6.96 (f) (7)  

The use of International Telecommunication Union-Technical (ITU-T) Group 3 or Group 4 compression techniques is required for document images without continuous tonal qualities.

If use of a propriety compression technique is unavoidable, a gateway to either Group 3 or Group 4 compression techniques must be provided.

13 TAC 6.96 (f) (8)  

*American National Standard for Information and Image Management-Recommended Practice for Quality Control of Image Scanners (most current version of ANSI/AIIM MS44) 

Use standard original targets.

Purchase targets from AIIM.  Do NOT use copies of targets!

A photocopy of Target 1 will render it useless for determining at what level of gray the scanner decides a point is black.

 A photocopy of Target 2 will destroy the usefulness of size, placement, black, and halftone test areas.

ANSI/AIIM MS44-1988 (R 1993) 7. Target 1: IEEE

8.Target 2: AIIM Scanner Target

Determine number of document types. 

Inventory documents to determine how many different types of documents are to be scanned.  Consider different sizes, colors, amount of white space, etc.

ANSI/AIIM MS44-1988 (R 1993) 

4.3 Establishing a Quality Reference. 

Establish a quality reference for “good  output” for each type of document.

Scan all targets, output to hardcopy, and examine.  Keep hardcopy and store digital images.  Note scanner adjustment on settings on the hardcopy.

ANSI/AIIM MS44-1988 (R 1993) 

4.3 Establishing a Quality Reference.

Determine when to change scanner settings.

Batch documents change scanner settings for batch; or (2) Change scanner settings for each document scanned.

ANSI/AIIM MS44-1988 (R 1993) 

5. Frequency of Testing

Testing for batch documents  

Perform a test run before and after scanning for each document type.

ANSI/AIIM MS44-1988 (R 1993) 

4.3 Establishing a Quality Reference 

4.4 Pre and Post-testing

Testing after each scanned document

Determine a small number of allowable settings for each document type

ANSI/AIIM MS44-1988 (R 1993)

4.3 Establishing a Quality Reference.

Test after maintenance

After recalibration by a technician, perform a test run of all targets.

ANSI/AIIM MS44-1988 (R 1993)

4.3 Establishing a Quality Reference.

Recordkeeping 

Each test run should be recorded on a log sheet. Log can spot a problem earlier; ex: threshold settings keep changing in one direction.  If it is an equipment problem, note scanner settings on the back of the printout and save for technician.

ANSI/AIIM MS44-1988 (R 1993) 

4.5 Record Keeping 

Frequencies of test run

Perform test run before and after each batch is scanned.  No limit to batch size, but frequent testing recommended.  Terminate each batch at end of a shift.

ANSI/AIIM MS44-1988 (R 1993) 5. Frequency of Testing

Set up scanner parameters

Set up parameters to match original scan parameters of the most recent quality reference.

ANSI/AIIM MS44-1988 (R 1993) 

6.1 Setup of Scanner Parameters

Placement of targets

Place targets on scanner in same manner as original documents will be processed.

ANSI/AIIM MS44-1988 (R 1993)

6.2 Placement of Target on the scanner 

Scanning

Scan targets using same procedures for original documents.  Do not change settings between targets.  After printed image is evaluated, electronic test images may be erased or stored.

ANSI/AIIM MS44-1988 (R 1993)

6.3 Scanning

Examine targets on screen

Test targets may be viewed on screen until a good test run is achieved; however, there is no easy way to compare a current scan against reference scan.  Also, display may not show full page.  Easier to compare paper.

ANSI/AIIM MS44-1988 (R 1993) 

6.4 Examination of Targets on Screen 

Printing targets

Test scans should be output to paper.  Provides end-to-end system check.  Can quickly determine if output is acceptable.

ANSI/AIIM MS44-1988 (R 1993) 

6.5 Printing of Targets

Determination of problem areas

If printed output is not acceptable, check printer by retrieving a quality reference and printing.  If output is not acceptable, it is probably a printer problem.  If output is acceptable, it is probably a scanner problem.

ANSI/AIIM MS44-1988 (R 1993)

6.6 Determination of Problem Areas

Target 1: IEEE Std 167-A1987 (Institute of Electrical and Electronic Engineers)

This target is a facsimile machine test target, which provides a continuous-tone photograph, gray scale, precision measurement marks, resolution charts and test characters.

ANSI/AIIM MS44-1988 (R 1993)

7. Target 1: IEEE 

Target 1 (continued)

Gray scale shows the threshold pint at which the scanner decides that a given area is black rather than white.  Run the target and observe the point at which the bars turn from white to black.  The continuous wedge will have ragged appearance at transition area.  In daily testing, observe if the white-to-black transition points are in the same place as the reference copy.  If so, threshold setting is properly adjusted.  Use Patterns #7 and #8 on the target to test gray scale.

ANSI/AIIM MS44-1988 (R 1993) 

7.1 Thresholds for Gray Levels

Target 1 (continued)

(2) Resolution is a measurement of the output quality of an image, usually in terms of samples, pixels, dots, or lines per inch.  At the center the lines are very narrow and will blur at varying distances from the center.  The distance of the blurring from the center is an indication of the resolution of the total system.  Use Patterns #12 and #13 to test scanner resolution.

ANSI/AIIM MS44-1988 (R 1993) 

7.2 Resolution

Target 1 (continued)

(3) Linearity and Rectangularity involves testing the calibration to make sure that the system is not distorting images.  The length of the line on the copy should match the length of lines on the original.  All lines should be straight, and lines on opposite lines should be equal.  The rectangularity of the image is perfect if the length of opposite sides are diagonal corners are identical

ANSI/AIIM MS44-1988 (R 1993) 

7.3 Linearity and Rectangularity

Target 1 (continued)

(4) The text can be tested by examining the smallest legible type.  This should not vary between the daily test runs and the quality reference.  Though explained under Target 1, it recommends the use of Target 2 for this test.

ANSI/AIIM MS44-1988 (R 1993) 

7.4 Text

Target 2: AIIM Scanner Target

This target is an ink-on-paper target that simulates conditions that may cause scanner problems.

ANSI/AIIM MS44-1988 (R 1993) 

8. Target 2: AIIM Scanner Target

Target 2 (continued)

Check to ensure that the scan area is the proper size.  If the black boxes in the corners of the target do not run off the edges of the scanned image, the scanner area may be too large.  If the “0” digit in the line of numbers at the corners of each page are visible on the scanned image, the scan area is too small, or the target was not properly aligned.  Note: Some printers do not print to the edge of the paper – this should be taken into consideration.

ANSI/AIIM MS44-1988 (R 1993) 

8.1 Size of scan area

Target 2 (continued)

(2) Check the alignment of the page.  If the scan area is the proper size and target is perfectly aligned, the “0” digit will show all corners and at the center of each edge.  If the image size or printout size is reduced, look for the same numbers appearing at all points.

ANSI/AIIM MS44-1988 (R 1993)

8.2 Alignment of Page

Target 2 (continued)

(3) Check the text by examining the small characters and punctuation to determine where scanning problems may occur.  Look for eligibility and detail of the small characters.  Examine the News Gothic Bold Reverse font for character lines that may be filed with black.  It is important to know at what type size the scanner will lose the distinction between lower case letters such as a, e, c, and o.  A properly adjusted scanner with 300 dots per inch should preserve these characters in a 4-point type.

ANSI/AIIM MS44-1988 (R 1993)

8.3 Text

Target 2 (continued)

(4) There should be five horizontal and five vertical lines on the page.  Verify that the thinnest line is visible.  Note that stair-stepping in the lines is normal if the target is not exactly parallel to the scan lines.

ANSI/AIIM MS44-1988 (R 1993)

8.4 Horizontal and Vertical Lines

Target 2 (continued)

(5) The diagonal line across the target is a test for uniform transport movement.  The line should be smooth and straight within the capability of the scanner and recorder.  Breaks in the line may indicate that a mechanical transport is not working smoothly or is being forced to pause and restart.

ANSI/AIIM MS44-1988 (R 1993)

8.5 Diagonal Line

Target 2 (continued)

(6) The isolated characters simulate a page number or part of a mathematical equation.  Because of the white space around each character, some scanners will see these characters as dirt specks and eliminate them.  Some scanners will fail on the degree symbol (last column, center row) and display it as a solid dot.

ANSI/AIIM MS44-1988 (R 1993)

8.6 Isolated Characters