Mittuniversitetet

Reflectance models such as the monochrome Murray–Davies (MD) and the Neugebauer color equations make inaccurate predictions owing to changes in reflectance or tristimulus values (TSVs) of halftone dots and the paper between the dots. In this paper, we characterize the change of micro-TSVs as a function of printed area in spectral halftone image by a power function and compare its prediction efficiency using theoretically and experimentally measured limiting TSVs assuming dots of uniform thickness. We found that experimentally accounting for dot thickness variations as solid and mixed areas more precisely explained the single-model parameter that captured the observed lateral light scattering effect. The results showed that incorporating empirically modeled TSVs of the dots and the paper between dots, as well as introducing a new term addressing mixed area in the MD equation, produced CIE ΔE* ab in the range 1.22–1.76, and the overall gain was more than 1 ΔE* ab.

We propose expanding the Murray-Davies formula by adding the effect of edges of solid inks in a halftoned image. The expanded formula takes into account the spectral reflectance of paper white, full tone ink and mixed area scaled by the fractional area coverages. Here, mixed area mainly refers to the edge of an inked dot where the density is very low, and lateral exchange of photons can occur. Also, in such area the paper micro components may have higher scattering power than ink, especially, in uncoated paper. Our methodology uses cyan, magenta and yellow separation ramps printed on different papers by impact and non-impact based printing technologies. The samples include both frequency and amplitude modulation halftoning methods of various print resolutions. Based on pixel values, the captured microscale halftoned image is divided into three categories: solid ink, mixed area, and unprinted paper between the dots. The segmented images are then used to measure the fractional area coverage that the model receives as parameters. We have derived the characteristic reflectance spectrum of mixed area by rearranging the expanded formula and replacing the predicted term with the measured value using half of the maximum colorant coverage. Performance has clearly improved over the Murray-Davies model with and without dot gain compensation, more importantly, preserving the linear additivity of reflectance of the classical physics-based model.

A method has been proposed, whereby k-means clustering technique is applied to segment microscale single color halftone image into three components—solid ink, ink/paper mixed area and unprinted paper. The method has been evaluated using impact (offset) and non-impact (electro-photography) based single color prints halftoned by amplitude modulation (AM) and frequency modulation (FM) technique. The print samples have also included a range of variations in paper substrates. The colors of segmented regions have been analyzed in CIELAB color space to reveal the variations, in particular those present in mixed regions. The statistics of intensity distribution in the segmented areas have been utilized to derive expressions that can be used to calculate simple thresholds. However, the segmented results have been employed to study dot gain in comparison with traditional estimation technique using Murray-Davies formula. The performance of halftone reflectance prediction by spectral Murray-Davies model has been reported using estimated and measured parameters. Finally, a general idea has been proposed to expand the classical Murray-Davies model based on experimetal observations. Hence, the present study primarily presents the outcome of experimental efforts to characterize halftone print media interactions in respect to the color prediction models. Currently, most regression-based color prediction models rely on mathematical optimization to estimate the parameters using measured average reflectance of a large area compared to the dot size. While this general approach has been accepted as a useful tool, experimental investigations can enhance understanding of the physical processes and facilitate exploration of new modeling strategies. Furthermore, reported findings may help reduce the required number of samples that are printed and measured in the process of multichannel printer characterization and calibration.

Conventional 3-channel color images have limited information andquality dependency on parametric conditions. Hence, spectral imaging andreproduction is desired in many color applications to record and reproduce thereflectance of objects. Likewise RGB images lack sufficient information tosuccessfully analyze diabetic retinopathy. In this case, spectral imaging may bethe alternative solution. In this article, we propose a new supervised techniqueto detect and classify the abnormal lesions in retinal spectral reflectance imagesaffected by diabetes. The technique employs both stochastic and deterministicspectral similarity measures to match the desired reflectance pattern. At first, itclassifies a pixel as normal or abnormal depending on the probabilistic behaviorof training spectra. The final decision is made evaluating the geometricsimilarity. We assessed several multispectral object detection methodsdeveloped for other applications. They could not proof to be the solution. Theresults were interpreted using receiver operating characteristics (ROC) curvesanalysis.