Print quality (PQ) is most important in the printing industry. It plays a role in users’ satisfaction with their products. Page quality will be degraded when there are print quality defects on the printed page, which could be caused by the electrophotographic printer (EP) process and associated print mechanism. To identify the print quality issue, customers have to consult a printer user manual or contact customer service to describe the problems. In this paper, we propose a comprehensive system to analyze the printed page automatically and extract the important defect features to determine the type and severity of defects on the scanned page. This system incorporates many of our previous works. The input of this system is the master digital image and the scanned image of the printed page. The comprehensive system includes three modules: the region of interest (ROI) extraction module, the scanned image pre-processing module (image alignment and color calibration procedure), and the print defect analysis module (text fading detection, color fading detection, streak detection, and banding detection). This system analyzes the scanned images based on different ROIs, and each ROI will produce a printer defect feature vector. The final output is the whole feature vector including all the ROI feature vectors of the printed page, and this feature vector will be uploaded to customer service to analyze the printer defect.
Print quality (PQ) is most important in the printing industry. To detect and analyze print defects is an effective solution to improve print quality. As the different types of print defects appear in different regions of interest (ROI) in the digital image of a scanned page, extracting the different ROIs helps to detect and analyze the printer defect. This paper proposes a method to extract different ROIs based on the digital image object map [1], which includes three different labels: raster (images or pictures), vector (background and smooth gradient color areas), and symbol (symbols and texts). Our ROI extraction method will extract four kinds of ROIs based on these three labeled objects. So we need to distinguish the background area and smooth gradient color area (color vector) from other vector objects. The process of the ROI extraction method includes four parts; and each part will extract one kind of ROI. For the color vector and background ROI extraction part, we develop two approaches: one is to obtain the maximum area rectangular ROI; and the other approach is to extract the deepest rectangular ROI. With both of these two methods, we use a greedy algorithm to gather additional useful ROIs. In the final result of the ROI extraction process, we only save the left top and right bottom positions for each ROI. In the end, we design a Matlab GUI Tool and label the ROI ground truth manually. We calculate the intersection over union (IoU)) between the ROI extraction result and the ROI manually labeled ground truth to evaluate our ROI extraction algorithm, and check whether it is good enough to crop different ROIs from the image of the scanned page to detect and analyze print defects.
Macro-uniformity is an important factor in the overall quality of prints from inkjet printers. The International Committee for Information Technology Standards (INCITS) defined the macrouniformity for prints, which includes several printing defects such as banding, streaks, mottle, etc. Although we can quantitatively analyze a certain kind of defect, it is difficult to assess the overall perceptual quality when multiple defects appear simultaneously in a print. We used the Macro-uniformity quality rulers designed by INCITS W1.1 as experimental references, to conduct a psychophysical experiment for pooling perceptual assessments of our print samples from subjects. Then, calculated features can describe the severity of defects in a test sample; and we trained a predictive model using these data. The predictor can automatically predict the macro-uniformity score as judged by humans. Our results show that the predictor can work accurately. The predicted scores are similar to the subjective visual scores (ground-truth). Also, we used 6-fold cross-validation to confirm the efficacy of our predictor.