Analysis and evaluation of printing image quality of Unusual Daughter Birthday Cards

Analysis and evaluation of printing image quality of Unusual Daughter Birthday Cards

 

I. Introduction

Printed matter is the result of text and text reproduction, the comprehensive effect of the original manuscript, materials, equipment, imaging and transfer process and other factors to form a printed image. The vector properties of graphics and text are crucial to page description, typography and RIP interpretation. In particular, the output of graphics, text and images must be treated differently to obtain the most reasonable results. However, the vector properties of graphics and text have become meaningless to the printed result because these objects lose their vector characteristics once they are transferred to the surface of the paper. Therefore, it is not necessary to distinguish the dot matrix description and vector description characteristics of objects when evaluating printing quality, and it is reasonable to think that printing quality evaluation is equivalent to printing image quality evaluation (hereinafter referred to as image quality evaluation).

 

Human's research and application of image quality evaluation technology has never stopped. Because of the continuous development of image quality improvement activities in the field of imaging technology, a lot of progress has been made. For example, image quality evaluation technologies in traditional printing, silver halide photography, color television and digital printing fields, many of which show good correlation with physical characteristics and subjective evaluation results, and image quality evaluation systems suitable for different fields have been developed. However, it is difficult to accept a measure only by subjective evaluation.

 

Color offset printing is the main means of modern graphic printing. For a long time, the evaluation of the quality of printed images is determined by the visual and subjective impression of customers. This is a very poor stability and very difficult work. Therefore, how to objectively evaluate and fully apply the technical process means to achieve color printing image quality control is very important. Print image quality assessment refers to the use of data to represent the correlation between the product and the manuscript. With certain data and indicators, the permissible range of product quality error and quality decline can be controlled to ensure the reproducibility and quality stability of image reproduction.

 

There are three key factors in the quality of the printed image: (1) the clarity of the image, that is, whether the change of the edge density of the image is consistent with the visual sensitivity of the human, mainly depends on the screen printing and printing dot expansion. (2) The level of the image, that is, the visual discernable density of the image, mainly depends on the expansion and reduction of the contrast between pre-press and printing. (3) The color of the image, that is, the reproduction accuracy and neutral gray balance of various colors of the image, mainly depends on the color separation mechanism, sensitive materials, paper, ink and other quality.

 

Two, the traditional printing image quality evaluation of the main methods

No matter how to evaluate the quality of the printed image, it must come down to the definition, level and color of the image. The evaluation methods mainly include subjective evaluation, objective evaluation and a comprehensive method based on visual evaluation.

 

1. Subjective evaluation method

The subjective evaluation of printed images is a method to evaluate the quality of printed images according to experience. Subjective evaluation methods often use visual evaluation method and qualitative index evaluation method. Visual evaluation method refers to that under the same evaluation environment conditions (such as consistent light source and illumination) by a number of experienced managers, technicians and users to observe the original manuscript and printed matter, and then based on each person's experience, emotions and preferences, according to the excellent, good, medium, difference grade of each printed matter, and statistics of the frequency of each grade, the unanimous praise is good, good, and vice versa. Qualitative index evaluation method is to list the important factors affecting the quality of each index according to a certain qualitative index, and score by several experienced evaluators. The quality of the high score is good, and that of the low score is poor.

 

① Multidimensional scaling method:

Multidimensional scaling is a scaling technique based on mathematical statistics. When comparing the difference between the samples in pairs or determining the satisfaction degree of the samples, the multidimensional scaling method can be used to analyze and identify the main parameters used in people's evaluation. The relative importance of the main parameters of printing quality can be determined when printing samples are evaluated by this method. The value obtained from the evaluation can make the internal correlation between subjective evaluation and objective evaluation or with the paper property. The reliability of the quality evaluation of each printed matter and the consistency of the evaluation of each evaluator (such as printing and papermaking experts, readers, advertising personnel, etc.) with the evaluation team can also be obtained.

Multidimensional scaling was developed by Thorgesen, which states that if there is a perceived difference between two elements, the difference can be expressed by a geometric distance. If this difference is recorded on a linear scale, the scale value on the scale shows this distance, which can then be used to build more than one dimension geometric models reflecting the relationships between the samples.

 

An important feature of multidimensional scaling technique is that it can be used to scale the weighted situation of subjective psychological factors of the evaluator. The function of each parameter in an evaluation can be expressed by an expectation vector.

 

② Pair comparison method:

People have subjective characteristics when judging the quality of printed matter, and different people will make completely different conclusions. This kind of objective inconsistency cannot be ignored as deviation or randomness. Even if there are reference objects that can be used as the comparison basis in the evaluation, there will be inconsistency in the evaluation results. The inconsistency in the subjective evaluation can not be ignored as deviation and randomness.

 

In many cases there is no reference for comparison at all. At this time, internal comparison can only be made between the objects to be judged. There are two common methods: one is to arrange the samples to be judged in a certain order; Second, each sample in a group of evaluated samples is compared with other evaluated samples one by one. On the basis of comparison, scores are scored and evaluations are made according to the integral. This is the paired comparison method. It belongs to a subjective evaluation method.

 

When implementing the paired comparison method, it is important to ensure that no unrelated factors interfere with the comparison when the judges are simultaneously comparing two samples. The printed matter to be compared should be in exactly the same standard lighting condition, the background should be neutral, and the room used for comparison should be able to ensure that the judges concentrate, without interrupting the judging work or interfering with the judges. The samples to be compared should be provided to the judges in a random order so as to make the comparison process unbiased. It must be clearly and simply explained to the judges on what basis the judgment will be made, and there must be no ambiguity.

 

Such subjective comparison experiment is only the comparison between samples, and the comparison results are only relevant to the samples being compared. Whether the results of pair comparison are reliable can be known by examining the "reliability of judgment" of the judge, that is, by calculating the reliability coefficient to measure.

 

2. Objective evaluation method

① Evaluation of the reappearance of the tone (level) :

The contrast range of the transmission manuscript is very different, and the color print should be reproduced as a basically consistent density range of the picture, the density range is mostly lower than the density range of the color manuscript, in the separation of color plate, it is necessary to make compression adjustment. On the one hand, the redistribution and adjustment of each tone level depends on the level distribution of the original manuscript, and on the other hand, it is related to people's visual perception. Among them, there are Munselming value factor of visual response and subjective factor of people's visual psychological requirements. To copy and adjust the level of tone of the original manuscript, it is necessary to comprehensively summarize people's visual psychological needs, add the physical value of visual response, and consider the level distribution of the original manuscript to obtain the density level of tone level reproduction curve of the printed picture. Then, the evolution data in the process of replication and reproduction are included to design the level of tone reproduction curve of the specific original manuscript.

 

The evaluation of the step reproduction of printed images, if only from the perspective of print quality inspection, that is: to measure the printing field density of various colors of ink layer; Measure and calculate the overprint rate of each color ink layer; Measure the step increase of printing dot or measure and calculate the printing K value; Check the ink dot transfer quality; Measure and describe the density level reproduction curve of the printed matter to the original manuscript. Through these objective technical data measurement, and then compared with the quality standards formulated by the department, we can determine the quality grade of specific color printing products.

 

② Evaluation of color reproduction:

Color reproduction, there are three different concepts. The first is the physical sense of color reproduction, the reproduction color and the original color in each color point spectral distribution is exactly the same. But printed matter is for visual viewing, it is difficult to achieve the physical sense of the same spectrum color reproduction, there is no much need. Second, reproduction in the sense of colorimetry, so that the printing reproduction image and the original color points in the color of the same or similar, that is, different spectrum with the same color effect, which is a realistic objective evaluation of color reproduction standards. The third is the psychological sense of color reproduction, that is, the color of printing reproduction may have some difference with the color of the original manuscript in chrominance, but in color effect may achieve visual psychological satisfaction, here add subjective evaluation factor.

 

Due to the lack of chrominance performance characteristics of printing materials (paper, ink, etc.), the imperfection of color separation reproduction means and equipment performance, as well as the deficiency of color representation in the printing reproduction mode itself, the realistic printing technology can not faithfully restore all the colors of the original manuscript or original scenery, even if the reproducible part can not reach the degree of faithful restoration, only relatively close. This brings some difficulties to the objective evaluation of color reproduction quality of printed matter. [next]

Only from the print to the original manuscript or original scene color approximation, through the results of chrominance measurement to be compared. Between similarity and dissimilarity, we can make a comprehensive and comprehensive evaluation of color reproduction in printed matter by combining people's psychological requirements for color vision, namely the degree of psychological reproduction.

 

If the objective technical measurement standard is set for the close degree of reproduction of the printed color to the original color, it should include: measurement and inspection of the color reproduction range of printing ink, measurement and inspection of the printing gray balance reproduction, measurement and inspection of the absolute reproduction accuracy of the printed color to the original color, and measurement and calculation of the relative reproduction degree.

 

③ Evaluation of articulation reproduction:

The sharpness of color print is an important quality index of image reproduction. In addition to the special artistic conception of the image, each picture should always have a part of the level (subject or background) is clear. There are three aspects related to the evaluation of the clarity of the printed picture: the firmness of the contour of the image layer; The clarity of the contrast change of two adjacent layers of the image is the subtle contrast; The resolution of the level of the original manuscript or printed picture, that is, the fineness of its fine level, is the essence of the objective scenery, which is the so-called texture.

 

3. Visual evaluation method

After years of research on human visual system (HVS), remarkable progress has been made in understanding the function of visual system from neurophysiology, psychophysics and other aspects. The important features related to image quality include: Weber-Fechner law, that is, the threshold of the perceived stimulus signal is proportional to the background intensity; The modulation transfer function (MTF) is band-pass characteristic, and the maximum response is between 2-5c/d. Region masking effect, that is, the signal visibility on the overthreshold background decreases; Multichannel resolvability of area, frequency and direction of visual signals.

 

A commonly derived objective distortion measurement method related to vision is to use the early information processing stage of HVS, that is, from the retina through the optical nerve, optic chiasma and lateral geniculate nucleus to the cerebral cortex, to convert the error image into the "perceptual domain", so that the error of the same amplitude at this time has the same visual perception. Finally, the errors are summed up into a distortion scale by a suitable distance measure. The complexity of the types of HVS and mass damage brings many difficulties to the study of this problem. In the past, some scholars have used visual properties obtained from relatively simple mental visual experiments, such as frequency selection and masking properties, to build visual models for quality prediction. With the continuous development and requirements of coding technology and application, the research on quality measurement based on vision is also deepening and has made remarkable progress.

 

S. Kerasekera & N. Kerbery proposed a distortion measure for damage. This damage is commonly seen in vector quantization and DCT coding and consists of vertical and horizontal edge distortions. Firstly, a subjective experiment was carried out on the synthetic test image, and the visual sensitivity of the edge damage was measured by adjusting four parameters: edge amplitude, edge length, background brightness and background activity, and a visual model was established. It is worth mentioning that visual sensitivity is measured by the response time to the stimulus rather than the usual visual threshold method, and the model parameters are determined by matching the constant-sensitive points. The model for the prediction of vertical edge error is (similar in horizontal direction). The edge information is first extracted from the error image by a low-pass filter with appropriate bandwidth, and then the activity and region masking effects are considered. The activity masking is determined by the weighted energy sum of the N×N square around the edge plus the window Fourier transform coefficients. Then the nonlinear transformation is carried out to adapt to the nonlinear characteristics of HVS, and finally the total error is calculated (taking the absolute average). In the test of real natural images, we design some images with the same MSE but different distribution of block structures. The model prediction agrees well with the subjective test, and it also shows that MSE is not a good distortion measure.

 

① Image quality evaluation method based on visual perception:

The effect of image degradation on human vision is determined by the sensitivity of human vision system, which is determined by the visual cells of human eye. In addition, the sensitivity of human visual system is also affected by the local spatial frequency of the image. A large number of experimental results show that the factor affecting the visibility of pixel error is the local regional environment around the error, rather than the background environment of the whole image.

 

According to the above visual characteristics, various HVS models have been established to evaluate the quality of images. The most typical HVS model is shown in Figure 1, which simulates three salient characteristics of visual perception, namely, visual nonlinearity (Weber's law), visual sensitivity bandpass, visual multichannel and masking effect.

 

② Image quality evaluation method based on visual interest:

The coding technology based on image content has inspired people to study the image quality evaluation method based on visual interest. From the perspective of visual psychology, vision is a kind of positive feeling behavior, which is not only related to physiological factors, but also depends on psychological factors to a considerable extent. When people observe and understand images, they tend to be unconsciously interested in some areas, which are called "ROI (Region of Interest)". The visual quality of the overall image often depends on the quality of the ROI, and the degradation of the area of disinterest is sometimes imperceptible. In real life, due to the influence of cultural background, surrounding environment and emotions, people's evaluation of the same image will have a large deviation, but there is a common feature for the areas of concern in the image, they concentrate on conveying most of the objective information to be expressed by the whole image.

 

The measurement method based on visual interest opens up a new way for image quality evaluation, but it is still in the early stage of research at present, and there are still many problems to be further studied. For example, how to determine the area of interest in an image; If the test image contains multiple areas of interest, how to determine the interest weight of these areas and so on.

 

Iii. Concluding Remarks

The correct evaluation of image quality is of great significance to the development of the whole image information engineering. It is believed that with the rapid development of multimedia information technology, the study of image quality evaluation will be paid more and more attention by people.

In the traditional fusion image quality evaluation methods, there are some defects, such as the subjective evaluation method is too cumbersome and cannot be repeated, the objective evaluation results do not match with the actual image quality or even contradict each other, and the existing evaluation methods are mostly suitable for the image fusion evaluation with strong correlation, not suitable for the fusion image evaluation in the field of entertainment photography. Therefore, it is an effective way to solve this problem by introducing HVS feature into the fusion image measurement method and organically combining the traditional objective evaluation method with the subjective evaluation method, which is also the development direction of the fusion image quality evaluation.