DIGITAL IMAGE-PROCESSING .1. EVALUATION OF GRAY-LEVEL CORRECTION METHODS IN-VITRO

The aims of this study were a) to assess in an in vitro model the amou nt of density changes measured in digitally subtracted images due to e lectronic noise and image alignment error, and b) to test the accuracy of different gray level correction procedures in the reduction of den sitometric image mismatches. A section of a pig mandible in which a ho llow cylinder ITI Bonefit(R) implant had been placed was used to obtai n pairs of standardized radiographs. Series of radiographs were obtain ed with different exposure times (0.34, 0.39, 0.44, 0.51, 0.58 s). The radiographs were captured through a video camera, digitized and store d in a personal computer. The same radiographic image was recorded and subtracted from itself 10 times to study the error of the method due to electronic transformations of the images and image alignment. The n oise due to the analog-to-digital transformation of the radiographic i mages was calculated to be +/-2 gray levels i.e., 2% of the scale of g ray levels. This kind of error was reduced up to 40% by capturing the images more than once and averaging the values per pixel. The manual s uperimposition of the images to be subtracted caused an increase of th e error to +/-3 gray levels (2.7%). Seven methods of gray level correc tion based either on a linear least squares approximation or on the cu mulative density function (CDF) were tested. The group based on the CD F algorithm gave significantly better results than any other method. P ixels yielding differences smaller or equal +/-7 gray levels (5.5% of the scale of gray levels) should be excluded from further calculations in order to eliminate (false-positive) errors due to the normalizing algorithms. Furthermore, the CDF method on an arbitrarily chosen area of the image or on the wedge seems to give to subtraction images the a bility of revealing real subtle changes in tissue density (fewer false -negative errors). The use of reference structures did not futher impr ove the ability of the normalization methods to correct gray level mis matches between radiographic pairs.