本研究採用分光式多頻譜掃描儀進行多種頻譜演算模式分析，從390nm~730nm內取樣間隔10nm的方式取出35個色頻波段，運用主成份分析法(PCA)期望能依其特性建立最佳演算分析模式，並且簡化其複雜的分析流程，由於頻譜的特性決定對色彩，所以在分析法上面的特徵分析更為重要。本研究採用較先進之分光式多頻譜技術進行研究，相較於以往多頻譜影像技術在於平面不同，以主成份分析進行分析，並且從實驗上的結果發現，色塊在PCA分析中有固定的差異量，並運用此實驗結果分析漸層色票。以主成份分析觀察其差異，藉由此差異量分析混色比例嘗試建立最佳的物體主色多頻譜複製。色彩複製上很容易會遇到同色異譜的現象，因此有相關文獻開始採用以頻譜複製的方式來進行色彩複製，希望能透過重建與複製物體的反射光譜，達到在不同的觀測光源環境下，依舊可以重現或預測其原始色彩。結果顯示，運用此頻譜的重建向量中的權重來預估網點面積是可行的方向。而出現在第二成份中材質的螢光特性會影響重建的準確度是另外值得關注的議題。
In this study, it is assumed that spectral model can be used to predict the dot area of ink jet printing. A grating-type multi-spectral imaging instrument is used to analyze the spectral data of ink jet output. The spectral reflectance values from 390nm to 730nm in 10nm increment are sampled and then analyzed through principal component analysis (PCA). The spectral absorptions of color ramps for each primary color (C, M, Y and K) are used to derive the virtual spectral primary from the first component in the PCA. Regression analysis is used to find out the scaling factor and to reconstruct the spectral reflectance of each color ramp according to the spectral primary respectively. Color differences are computed to judge the goodness of the reconstruction. In the end, dot area for each color ramp is calculated by the ratio of scaling factor in each ramp to the 100% tint. The results indicate that it is possible to estimate the dot area by this proposed method. Furthermore, the second component can be influenced by the fluorescent characteristics in the material, which deserves some attention as well.
