Study of no-reference image quality assessment algorithms on printed images

Abstract

Abstract. Measuring the visual quality of printed media is important since printed products have an important role in everyday life. Finding ways to automatically predict the image quality has been an active research topic in digital image processing, but adapting those methods to measure the visual quality of printed media has not been studied often or in depth and is not straightforward. Here, we analyze the efficacy of no-reference image quality assessment (IQA) algorithms originally developed for digital IQA with regards to predicting the perceived quality of printed natural images. We perform a comprehensive statistical comparison of the methods. The best methods are shown to accurately predict subjective opinions of the quality of printed photographs using data from a psychometric study.

From pattern recognition methods to machine vision applications

“…probabilistic methods for image quality assessment [35][36][37][38], and pattern recognition methods for nonvisual information [39].…”

“…Methods based on the overall visual quality index (VQI) of the whole image have been proposed in [35][36][37]. Tuytelaars et al have given a survey on VOC [31], and partly inspired by their article, methods based on the regions of interest in an image using VOC have been proposed in [27][28][29][30], enabling a content-based VQI where the regions of the image are weighted.…”

“…Based on these data, hypotheses can be verified as shown in Figure 11.7(a) where the experience of quality decreases when mottling increases. See [35][36][37] for further information. In the figure, the nth bin of the cumulative histogram tells how many times (percent of all samples) the ground truth match (human evaluation) is contained in the set of n closest samples in the second evaluation space (computational).…”

Assessment of the print quality parameters’ impact on the high-quality flexographic print visual experience

“…11 In a study by Tchan et al, a computer model was proposed for estimating quality of electrographic and inkjet prints by visual assessment. 12 The model was developed using neural networks, which were based on image analysis.…”