BEST PRACTICE RECOMMENDATION

FOR THE CAPTURE OF MUGSHOTS

September 23, 1997

The original version of the "Best Practice Recommendation" was initiated at the Mugshot and Facial Image Workshop which was held in Gaithersburg, MD on October 23-25, 1995. Developed as a recommendation, the implementation of the practices and principles described in that document makes the conversion of existing and ongoing photographic collections more uniform. It contains a suggested set of procedures and equipment specifications for organizations considering the purchase of new systems or the upgrade of current systems. The recommendation is not designed to render current and legacy mugshot collections unacceptable. Rather, it is intended as a means of establishing or improving interoperability between mugshot systems.

The information contained in this updated revision of the "Best Practice Recommendation", Version 2.0, does not alter any of the individual points that were consensually agreed upon and included in the original version of this recommendation. It does provide additional details and clarifications for many of those points and has been supplemented with information regarding depth-of-field and exposure considerations.

This recommendation reflects a minimum set of common denominators. The provisions of this recommendation are keyed to the quality aspects associated with the unaltered captured mugshot image. For new mugshot images being captured, the specifications contained in this recommendation are equally applicable to realtime electronic capture of mugshots as well as the electronic conversion of photographic images. For conversion of legacy files of photographs, most of the provisions of this recommendation are also still applicable. In the future, it should be possible to add additional specifications without contradicting any of the current contents of the recommendation.

POSE

The full-face or frontal pose is the most commonly used pose in photo lineups and shall always be captured. This pose is in addition to profiles or intermediate angled poses captured to acquire perspective and other information. For subjects who normally wear eyeglasses, a frontal mugshot image should be captured of the subject without glasses. This is required due to the glare from external illumination. An additional image can optionally be captured of the subject wearing eyeglasses.

The subject's captured facial image shall always be in focus from the nose to the ears. Although this may result in the background behind the subject being out of focus, it is not a problem. For optimum quality of the captured mugshot, the f-stop of the lens should be set at two f-stops below the maximum aperture opening when possible.

The facial image being captured (full-face pose) shall be positioned to satisfy all of the following conditions:

• The approximate horizontal mid-points of the mouth and of the bridge of the nose shall lie on an imaginary vertical straight line positioned at the horizontal center of the image.
• An imaginary horizontal line through the center of the subject's eyes shall be located at approximately the 55% point of the vertical distance up from the bottom edge of the captured image.
• The width of the subject's head shall occupy approximately 50% of the width of the captured image. This width shall be the horizontal distance between the mid-points of two imaginary vertical lines. Each imaginary line shall be drawn between the upper and lower lobes of each ear and shall be positioned where the external ear connects to the head.

Subject illumination shall be accomplished using a minimum of three (3) point balanced illumination. Appropriate diffusion techniques shall also be employed and lights positioned to minimize shadows, and to eliminate hot spots on the facial image. These hot spots usually appear on reflective areas such as cheeks and foreheads. Proper lighting shall contribute to the uniformity of illumination of the background described in the exposure requirement.

The subject whose image is being captured shall be positioned in front of a background which is 18% gray with a plain smooth flat surface. A Kodak or other neutral gray card or densitometer shall be used to verify this 18% gray reflectance requirement.

The exposure shall be keyed to the background. Several areas of the recorded 18% gray background shall be used to verify the proper exposure. The averages of the 8-bit Red, Green, and Blue (RGB) components within each area shall be calculated. Each of the RGB means shall fall between 105 and 125 with a standard deviation of plus or minus 10. Furthermore, for every area examined, the maximum difference between the means of any two of the RGB components shall not exceed 10.

The Width:Height aspect ratio of the captured image shall be 1:1.25.

The minimum number of pixels in an electronic digital image shall be 480 pixels in the horizontal direction by 600 pixels in the vertical direction. It should be noted that the image quality of the captured mugshots and facial images will be improved as the number of pixels in both directions are increased. However, as images are captured with an increased number of pixels, the 1:1.25 (Width:Height) aspect ratio will be maintained.

Two considerations must be noted regarding this aspect of the recommendation. First, the normal orientation of many available cameras is the landscape format which specifies a greater number of pixels in the horizontal than in the vertical direction. Unless these cameras capture at least 600 pixels in the vertical direction, it may be necessary to rotate the camera 90 degrees. Second, the 480x600 capture format exceeds the VGA display format of 640x480. Therefore, at a minimum, an SVGA specification of 800x600 pixels will be required to display the facial image. The image will occupy less than the total number of available horizontal pixels.

Captured electronic color facial images are required. Digital images shall be represented as 24-bit RGB pixels. For every pixel, eight (8) bits will be used to represent each of the Red, Green, and Blue components. The RGB colorspace is the basis for other colorspaces including the Y, Cb, Cr and YUV. Additional color management techniques are available from the International Color Consortium. Information regarding these techniques can be downloaded from the following URL: www.color.org

Digital cameras and scanners used to capture facial images shall use square pixels with a pixel aspect ratio of 1:1.

The algorithm used to compress mugshot and facial images shall conform to the JPEG Sequential Baseline mode of operation as described in the specification approved by the ANSI X3L3 Standards committee. The target size for a JPEG compressed color mugshot image file shall be 25,000 to 45,000 bytes.

The JPEG File Interchange Format (JFIF) shall contain the JPEG compressed image data. The JFIF file shall then be part of the transaction file for interchange which conforms to the requirements as contained in ANSI/NIST-CSL 1-1993 and ANSI/NIST-ITL 1a-1997.

Bradford Wing
NIST
100 Bureau Drive, Stop 8940
Gaithersburg, MD, 20899.
voice (301) 975-5663
fax (301) 975-5287
email-> bwing@nist.gov

Created May 22, 1996.
Last modified September 15, 2009.
Contact webmaster@magi.nist.gov with corrections/comments.