Research On Evaluating Method For Print Mottle Based On Digital Image Processing Technology

Print mottle is the common defects in the printing industry. It is a critical factor affecting the printing quality. Print mottle will seriously affect the appearance of the product packaging, and also compromise the quality of the printed matter. It is really necessary to provide a suitable model for predicting the print mottle.This paper is to study the print mottle evaluation method based on digital image processing technology. Firstly, we analyze the important factors affecting the print mottle. Mainly includes three aspects: the human visual system, the paper properties and the image noise. Secondly, we carry out the print mottle testing experiments based on the traditional IGT printability tester, and introduce the specific visual subjective evaluation method. Compared with the print mottle mathematical model of the traditional methods, we propose new methods of measurement based on the scanner and built a new model of print mottle. Print mottle evaluation model is based on wavelet analysis. There are mainly two methods, the wavelet de-noising theory and the wavelet detail energy theory. With the Matlab simulation, combined with the human eye evaluation system to validate the print mottle model in order to analyze the feasibility and accuracy.The paper is to explore the measuring principle of print mottle measurement system, and research the inkjet paper mottle situation. Meanwhile, research the print mottle situation of different types of papers such as newspaper, offset paper, coated paper and inkjet paper. The modified coefficient of variation(Mod Co V) value is better for the same paper on the same wavelength range in the same direction to reflect the difference between each other. In the study of print mottle orientation occurs, print mottle levels are different for the same kind of paper on different wavelengths and different orientations, and Mod Co V values can better reflect the difference in different directions of the print mottle.Based on the wavelet denoising theoretical analysis method, selecting the bior2.4 wavelets and the fourth decomposition level can get the best denoising effection. Using print mottle evaluation method based on wavelet denoising theory, the optimal parameter is the sym5 wavelets of secondary decomposition level. In the same conditions, within the entire wavelength range, using this method can get the desired results. However, the results obtained in a short octave band are not very good, especially in the larger wavelength.Based on the wavelet detail energy theory analysis method of print mottle, using 12 kinds of wavelets, from the first decomposition level to the fifth decomposition circumstances, can achieve a high degree of correlation.While the correlation of the sixth and the seventh decomposition level is slightly worse. The most optimal parameter is the db2 wavelet of the fifth decomposition level. In the same conditions, using print mottle evaluation method based on the theory of wavelet detail energy can get the desired results in the entire wavelength range. Compared to the results obtained by the method of noise theory, the results within a shorter range octave band is more ideal. The results showed that the method based on the wavelet detail energy theory is the most optimal.