See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/355393884

Replication of Screen-Printing Fabric via Ink-jet Textile Printing

Article  in  Journal of Imaging Science and Technology · July 2019

CITATIONS READS
3 585

4 authors, including:

Ming Wang
North Carolina State University
5 PUBLICATIONS   14 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Ming Wang on 19 October 2021.

The user has requested enhancement of the downloaded file.

 Journal of Imaging Science and Technology R
63(4): 040402-1–040402-10, 2019.
c Society for Imaging Science and Technology 2019

Replication of Screen-Printing Fabric via Ink-jet Textile

Printing
Ming Wang , Lisa Chapman, Marguerite Moore, and Minyoung Suh
Wilson College of Textiles, North Carolina State University, USA
E-mail: [email protected]

for quicker product production that is responsive to fashion

Abstract. Digital textile printing (DTP) is fast, flexible, and relatively changes is important to compete in consumer markets.
inexpensive for sample printing, and can be applied quickly in
response to consumer demand. The aim of this two-stage research
Printing of smaller collections, unique textile products, and
was to analyze the potential of DTP to replace traditional screen the sampling process, require greater flexibility of printers
printing for a specific textile product. In Stage One, an optimal DTP and, consequently, faster prepress and printing of color
workflow was established. The workflow included determination designs [5]. The most promising opportunity to date for
of the colorant and substrate combination, color calibration, CAD
file, and the necessity of pretreatment. In Stage Two, a visual
addressing these challenges for the printing industry is
assessment instrument and protocol were established to evaluate ink-jet digital printing.
the acceptance of replicated ink-jet printed fabric. The visual The aim of this research was to analyze the po-
assessment and protocol were designed to evaluate the acceptance tential of DTP to replace traditional screen printing for
of the ink-jet printed sample to fully replicate the screen-printed
sample via seven measured aspects. These seven aspects include:
a specific textile product. To achieve the research goal,
perceived color difference, lightness difference, overall color, scale, two experimental stages, which address three research
line quality, visual texture, and overall appearance. Data gathered objectives, were conducted: (1) To develop a workflow for
from the visual assessment was then analyzed and compared the fabric replication of a screen-printed fabric using ink-jet
using SPSS statistics software. The results indicate that DTP
demonstrates a significant potential alternative for traditional screen
textile digital printing; (2) To develop a visual assessment
printing. c 2019 Society for Imaging Science and Technology. instrument and protocol to evaluate the acceptance of
[DOI: 10.2352/J.ImagingSci.Technol.2019.63.4.040402] the replicated ink-jet printed fabric; (3) To determine if
the ink-jet printed sample was a suitable substitute for
screen-printed sample via an expert visual assessment pilot
1. INTRODUCTION testing.
As the digital world continuously creates new possibilities,
digital textile printing (DTP) represents the future direction 2. EXPERIMENTAL STAGE ONE
of technology development in the textile printing and dyeing 2.1 Material
industry [1]. Global annual consumption of printed textile A rotary screen-printed, 100 percent cotton sample for
products is expected to reach 29.8 billion square meters by interior home furniture end use was supplied by Springs
2021 with 17.5 percent annual growth rate. Screen printing Creative Products Group LLC., USA. The corresponding
is the most acceptable textile printing method, accounting digital design TIFF file, in RGB color mode, color separated
for about 35 percent of the overall market in 2017. Although into eight channels was also provided. Pretreated and un-
ink-jet printing is expected to replace screen printing as treated 100 percent cotton was used by Principal Investigator
the primary method in 2021 with an 18.5 percent annual (PI) for ink-jet digital printing. The fabric was supplied by
growth rate, screen printing remains the most widely used Premex and had a weight of 228 gsm, with a fiber content
printing method in the textile industry today [2–4]. However, of 96% cotton and 4% linen. A noncommercial eight-color
despite its productivity, the dominant rotary screen-printing nanopigment was used.
method has several limitations [1]. One limitation is that
color and pattern changes are slow and expensive because 2.2 File Preparation
rotary screen printing requires a long time to set up (up Color reduction is necessary to control color variety, clean up
to 6–8 weeks) [1–4]. The screens needed for traditional stray pixels, and ensure color consistency [6]. To guarantee
printing machines are not as durable as digital file storage,
the best replication result, the RGB digital design file was
and also require more space for storage and operation [2].
opened in Lectra Kaledo Print software, color reduced, stray
As the textile industry continues to mature, the demand
pixels cleaned up, recolored using an RGB color table, and
kept in an RGB color mode, TIFF format (Figure 1(a)).
IS&T Member.
Received Mar. 10, 2019; accepted for publication June 5, 2019; published 2.3 Color Calibration and Profiling
online Aug. 23, 2019. Associate Editor: Branka Lozo. Color calibration for all devices is a requirement that
1062-3701/2019/63(4)/040402/10/$25.00 actively contributes to a color-managed workflow, and can

J. Imaging Sci. Technol. 040402-1 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Figure 1. Facilities and process for conducting ink-jet printed sample (A) Color reduction in Lectra Kaledo, (B) Creation of enhanced color profile in
RipMaster 10.0, (C) Part of color profile for nanopigment printing on pretreated cotton for MS JP5, (D) X-Rite i1iO spectrophotometer, (E) Practix Mfg heat
calendar, (F) MS JP5 Evo Printer.

be used to create a color profile that represents the range that varies the ink volume and drop sizes from 4 to 72
of printable colors specifically for a specific substrate and picoliters. In this study, individual ink limitations were
ink combination [6, 7]. For digital textile printing, color 100 percent for cyan, red, magenta, orange, yellow, violet,
calibration provides the color gamut and controls printing and gray and 120 percent for black. The C setting is the
process accuracy [7]. As shown in Fig. 1(c), color calibration most commonly used production print mode requiring a
was performed for each colorant and substrate combination balance between production speed and print quality. The
using an X-Rite i1iO spectrophotometer (Fig. 1d) and MS JP5 pretreatment process is time-consuming, and add cost to
Evo printer. All steps for creating an enhanced color profile the final product [8, 9]. Both pretreated and non-pretreated
followed a specific workflow. cotton substrates were provided by Springs Creative Products
Group LLC., USA. The pretreatment DP-300 (45% solids) for
2.4 Variables Effecting Ink-jet Textile Printing Results cotton was developed by Lubrizol Corporation specifically
Printer setting, pretreatment, as well as the ink and substrate for pigment-based ink-jet printing on textiles. The main
combination were identified as variables affecting the quality components in the pretreatment reagents were multivalent
of ink-jet textile printed samples. For printer settings, the metal salts as ink coagulants, acrylic resin for ink anchoring
MS JP5 Evo Printer (Fig. 1f) has eight different modes to the substrate, and additives for wetting and surface tension

J. Imaging Sci. Technol. 040402-2 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Two 15 × 15 inch medium gray-colored PVC easels were

used, one for viewing the screen-printed sample and one for
viewing the digital printed sample. The lamp of the viewing
booth had a color temperature of 6500 ± 200 K and constant
illumination of approximately 1400 lx [9–16].

3.3 Observers, Frequency, and Interval

Experienced participants are recommended for visual assess-
ment rather than inexperienced participants to increase the
accuracy and shorten the total experiment time [13, 14, 16].
To streamline the assessment process, all the participants in
the assessment were experienced experts in ink-jet digital
Figure 2. Screen-printed sample (left) and ink-jet printed sample (right). printing, color matching, screen printing, apparel product
development, home interior product development, or other
related areas. AATCC provided a pair of blue dyed fabrics
control such as isopropyl alcohol, propylene glycol, and samples so that participants could practice an assessment
silicone-based compounds. before the official experiment. Moreover, the Neitz test was
taken before the official experiment to determine whether the
2.5 Conduct Ink-jet Textile Printing Trial participants had normal or abnormal vision.
The ink-jet sample was printed based on the identified According to the American Association of Textile
variables. The print trials were conducted with an MS JP5 Evo Chemists and Colorists (AATCC) standard, each participant
Printer with Kyocera Drop-on-Demand (DOD) piezoelectric would repeat the assessment three times with 24 hours
print heads. The Kyocera head has four printing heads with between each trial. The time interval was needed as
each head containing up to two colors. The eight-color ink participants may memorize their answers or get tired if
set consisted of black, cyan, red, magenta, orange, yellow, they complete three trials at once [17–19]. Pearson (Pearson
violet, and gray. The ink-jet printed sample was heat fixed Correlation Reliabilities) and ANOVA (Analysis of Variance)
after printing by Practix Mfg heat calendar (Fig. 1e) with were used to analyze the correlation and reliability of the
400 degrees Fahrenheit fixation temperature and 30 seconds three trials [13].
dwell time. After fixation, the nanopigment printed cotton
sample (Figure 2) was used for conducting the expert visual 3.4 Viewing Order
assessment in experimental stage two. Screen-printed and ink-jet printed samples were all multi-
color samples, where distinct colors would affect each other
3. EXPERIMENTAL STAGE TWO during the observation process [13, 16]. Variance could
3.1 Determination of Variables happen if an observer viewed the colors in a different
To conduct the visual assessment comparison between the order [13]. Evaluating the color in the same order can
screen-printed sample (Sample A) and ink-jet printed sample eliminate the variance [17, 20]. A color key was created by
(Sample B), variables needed to be determined for the PI which represented the main colors on the screen-printed
assessment protocol and survey instrument. Five variables sample. As shown in Figure 3, Color 1 is Black, Color 2 is
were controlled consistently during assessment process: Yellow, Color 3 is Dark Green, Color 4 is Gray, Color 5 is
illuminant, viewing environment, observers, frequency and Blue, Color 6 is Red, and Color 7 is Cyan. Observers viewed
interval, and viewing order. Seven variables were chosen to the sample based on the numbered order.
evaluate the perceived difference between the two samples:
color, lightness, overall color, scale, line quality, visual texture, 3.5 Expert Visual Assessment Instrument Variables
and overall appearance. Determination
Color difference was used to determine how close the color
3.2 Illuminant and Viewing Environment matched between the ink-jet printed and screen-printed
The visual assessment was carried out in the Color Science samples, including a comparison between color values
Lab in the Wilson College of Textiles at North Carolina matching of each single color listed on the color key [19].
State University. To minimize variability, the PI arranged Lightness is the most direct color visual effect and a very
carefully controlled viewing conditions, which were kept the important quality index for a textile product [19, 20].
same throughout the test trials. A Macbeth Spectra Light III After identifying the single color change level, participants
viewing booth with a filtered tungsten daylight-simulating compared the lightness of each color. Participants were
lamp (D65) was switched on during the experiment. This asked to rate the overall color when assessing how well
light source was the only illumination in the lab, with all the appearance of the ink-jet printed sample matched the
other sources of light turned off. The samples were placed on screen-printed sample [16, 19, 20]. Scale was used to evaluate
a 15 × 15 inch medium gray-colored PVC easel, which was the overall print size proportion of the ink-jet printed sample
set at a 45-degree angle at the center of the viewing booth. compared to the screen-printed sample [18]. Line quality was

J. Imaging Sci. Technol. 040402-3 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Figure 3. Color key for visual assessment.

Figure 4. Gray Scale for Color Change (AATCC).

used to evaluate how well the weight, clarity, and uniformity

of the print in the ink-jet printed sample were compared to participants assigned a value of 1–5 after comparing colors,
the screen-printed sample [15]. Visual texture was used to with 5 representing no color difference between samples
evaluate the ability of the ink-jet printed sample to replicate and 1 representing the highest color difference between
print combined with the woven structure of screen-printed samples [19]. The participants were able to move the AATCC
sample [15, 17]. Overall appearance was used to evaluate how Gray Scale freely when compared to the color difference
well the color, scale, line quality, and visual texture interacted but were instructed not to move the samples to be assessed.
in the ink-jet printed sample compared to the screen-printed Regarding overall color matching, line quality, visual texture,
sample [9, 16–20]. and overall appearance, five feasible options were given to
participants to describe the matching level between two
3.5.1 Development of Survey Instrument samples which were: Not at all (1), Slightly (2), Somewhat (3),
A survey instrument was developed and submitted to Mostly (4), and Exactly (5). Each option was given a number
North Carolina State University’s Institutional Review Board for statistical analysis purposes. Regarding the question of
(IRB) for approval prior to the assessment. The survey scale matching, participants gave an answer of either yes
instrument included questions relating to the three trials or no. For the color lightness, participants were asked to
the participants were to complete. The instrument included choose one of three options that best described the lightness
informed consent, demographics, the Neitz vision test and relationship between the two samples; A is lighter than B1
the practice visual assessment test, and the first assessment (A > B), B1 is lighter than A (A < B), and A1 is the same
trial (4, 5). During the second and third trials, participants lightness as B (A = B).
were asked to complete the second (6, 7) and third assessment
(8, 9), respectively. In the second and third assessment trials, 3.7 Identify Participants and Samples Preparation
the same questions used in the first assessment were asked. In A list of possible participants either from experienced
total, each participant repeated the assessment three times, industry professionals or Ph.D. students from North Carolina
but participants were not informed in advance that the same State University was created. The industry professionals were
questions would be repeated for each trial. working in the areas of fabric design, home interior prod-
uct development, color matching, ink-jet digital printing,
3.6 Scale Description or screen printing. The Ph.D. students were performing
AATCC Gray Scale was used to evaluate the perceived color research in color science field. Research information and
shade differences between two samples. As shown in Figure 4, invitations were sent out through e-mail to 25 potential

J. Imaging Sci. Technol. 040402-4 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Figure 5. Viewing Booth Setup: Screen-Printed Sample (Left), Ink-jet Printed Figure 6. Viewing set up for practice sample.
Sample (Right).

The color key, which pointed out the specific color that
participants and 12 of them finished all three trials. To needed to be observed for any noticeable changes from Color
help control for any possible effects caused by gender and 1 to Color 7, was provided to participants. The participants
occupation, six females and six male participants, six from were free to move the color key and AATCC Gray Scale to
industry and six from academia, were selected. The ink-jet identify the color change value for the seven corresponding
printed cotton fabric and original screen-printed fabric were colors between the pair of samples, but they were not allowed
both folded into 15 × 15 inch rectangles (Figure 5). Every to move or change the display of samples. The screen and the
color used in the design was shown in the pattern size digital printed sample were identified by labels on the back
selected, which was smaller than one repeat to protect the of the 15 × 15 inch medium gray-colored PVC easel (one
copyright of Springs Creative Products Group LLC., USA. for each sample) to maintain consistency during experiment.
Participants performed color visual assessment and were
3.8 Expert Visual Assessment Viewing Protocol asked to finish all the survey questions. Each participant
The participants wore gray gloves to minimize color would repeat the assessment three times with 24 hours
variability as well as to prevent damaging the samples between each trial. The same sample sets, viewing booth, and
and the AATCC gray scale. The participants sat in front viewing conditions were kept through all the trials.
of the Macbeth Spectra Light III viewing booth, and the
filtered tungsten daylight-simulating lamp (D65) was turned
on by the PI. While the participants adapted to the light 4. RESULTS AND DISCUSSION
source by sitting in the viewing booth for 2 minutes, the 4.1 Stage One Result
steps of the experimental process were explained to them. An optimal workflow for the replication of screen-printed
After participants adapted to the viewing conditions in the fabrics via ink-jet textile printing (Figure 7) was developed
lab, they were shown the informed consent form. Once during experimental stage one. Through this effort the
they gave their consent, the Neitz Color Vision Test was process was streamlined and can be effectively used for future
given to the participants and their answers were evaluated research and production applications. The ink-jet printed
by the PI immediately. While their answers were scored, sample B was selected and approved by the PI to conduct
the participants were asked to complete the demographic expert visual assessment.
information part of the survey. A pass or fail score was
assigned to each participant based on the Neitz Color Vision 4.2 Stage Two Result: Expert Visual Assessment
Test results. If a participant failed the test, the experiment After successful completion of stage one, a visual assessment
would be immediately stopped; otherwise, the experiment protocol (Figure 8) was established to evaluate the acceptance
continued to the second test. The test samples which of the replicated ink-jet printed fabric. The expert visual
were provided by the AATCC Color Change Evaluation assessment was undertaken following this process.
Proficiency Testing Program were given to participants for
viewing practice. A pair of blue 3 × 3 inch samples were 4.3 Sample Characteristics
placed in the viewing booth on the easel with a hairline A total of 25 possible subjects either working in U.S.
gap between them (Figure 6). Participants sat in front of the textile industry or performing textile-related research
viewing booth with gray gloves on and used the AATCC Gray (Ph.D. students) were e-mailed invitations to participate
Scale freely to identify the color shade change for this pair of in the study. Twelve of them (48%) completed the visual
samples. The PI stood near the participants and assisted with assessment and their answers were recorded for analysis. To
using the AATCC Gray Scale as needed to ensure correct use. eliminate potential gender bias, six males and six females

J. Imaging Sci. Technol. 040402-5 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Figure 8. Research stage two diagram: expert visual assessment protocol.

Table I. Demographic Information.

Total e-mailed Responses Response Rate

E-mail Response
25 12 48%
Figure 7. Research stage one diagram: workflow of fabric replication. Male Female Gender Ratio
Gender
6 6 1:1
participated in the assessment. The participants’ ages were >=1990 1980–1989 <1979
quite different. Young, middle aged, and senior participants 32% 53% 15%
were all involved in the experiment. As shown in Table I, 32 Minimum Maximum Mean
Year of Birth
percent of participants were born after 1990, 15 percent were 1950 1994 1979
born before 1979, and more than half (53%) of participants Range
were born between 1980 and 1989. The minimum birth year 1950–1994
was 1950 (66 years old), the maximum birth year was 1994
(approximately 22 years old), and 1978 (38 years old) was
the mean birth year identified for all participants. dyeing technician, textile color matching technician, and
Every participant selected at least one and up to three textile chemist.
work areas (Table II). Half of the participants (6) identified
4.4 Frequency and Length of Time
their occupation as a Ph.D. student (50.0%), with the other
The length of time and frequency for approving sample color
half identifying their occupation as an Industry Professional difference reflected participants’ level of experience in visual
(50.0%). Participants also identified their specialization in assessment. As shown in Figure 9, for the length of time,
the areas of fabric design (16.7%), color matching (41.7%), the majority of participants identified their sample approval
ink-jet printing (41.7%), apparel product development experience as 2–3 years (42%), 6 or more years (25%),
(16.7%), and others (25%). A total of 17 responses were 7 months to a year (17%), 1–6 months (8%), and 4–5 years
gathered, and the percentages reflect how many times this (6.7%). No one had never approved samples (0.0%). The
option was selected in the total 17 responses. Occupations standard deviation of the sample approving experience was
provided by participants include association executive, 1.63; for the frequency, most participants selected 1–3 times

J. Imaging Sci. Technol. 040402-6 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Table II. Occupation Information. Table III. Neitz and Viewing practice.

Occupation The Neitz test results

Industry professional Ph.D. Students Participants’ Response Percent (%)

6 (50%) 6 (50%) Yes 100%

Details No 0%
Area of work Percent (%) Totally 100%
Fabric Design 16.7% Color change response for test sample (viewing practice)
Color Matching 41.7% 2.5 8%
Screen Printing 0.0% 2 33%
Ink-Jet Printing 41.7% 1.5 58%
Apparel Product Development 16.7% Totally 100%
Home Interior Product 0.0%
Participants’ Response Range 1.5–2.5
Other 25%
AATCC Suggested Response Range 1.5–2.5
Note*: Each participant selected at least one and up to three areas of work
Text Response (other)
Area of Work Response number all participants that attended the next step had to pass
Dyeing 1 the test. The PI evaluated participants’ Neitz test results
Association Executive 1 and marked their results as pass or fail. A pass score was
Color Matching Testing Method Developer 1 defined as answering seven or more of the nine questions
correctly. The third version of the Neitz Test was used
for this experiment and each version is the same for the
color blindness test. All participants (100%) passed the test
(Table III), indicating there should be no negative influence
on the next test step. The goal of the second test was to
practice the visual assessment by using AATCC Gray Scale.
As shown in Table III, a majority of participants identified
their samples as 1.5 (58%); some participants identified their
samples as 2 (33%); and fewer participants identified their
samples as 2.5 (8.0%). All results fell within the suggested
color range, while only one (8%) chose 2.5. A standard color
change range based on the previous data, between 1.5 and 2.5,
was established by AATCC before the test started. Based on
the AATCC standard color change range, the PI identified
the participants’ understanding and ability to observe the
sample color change which were acceptable.

4.6 Visual Assessment Results

For the individual color visual assessment mean response,
black was perceived to have the strongest color match. Yellow,
Dark Green, Red, and Gray were in the second strongest
color match group. Blue and Cyan were perceived to have the
Figure 9. Sample approval experience and frequency. weakest color match (Figure 10).
For the overall matching visual assessment, the overall
a week (42%) and 1–3 times a month (42%). Some selected color appearance received the highest match. Overall appear-
1–3 times a year (20%), and no one selected a frequency of ance and overall matching were also consistent between the
never. The standard deviation of sample approval frequency two samples. Compared to the other aspects, the line texture
is 2.12. and visual texture received the lowest match (Figure 11). For
the scale match, all participants (100%) identified that the
4.5 Neitz Test and Viewing Practice scales of samples A and B matched.
After completing the survey demographics, two tests were Visual assessment results indicated that all participants
administered to participants. The first one was the Neitz agreed that the digital printed sample matched well with the
test, which is used for a red–green and/or yellow–blue color screen-printed sample, as judged by color difference, scale,
deficiency. To validate the responses of the visual assessment, line quality, visual texture, and overall appearance.

J. Imaging Sci. Technol. 040402-7 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Figure 10. Mean Response for Individual Color Visual Assessment.

Figure 11. Mean Response for Overall Matching Visual Assessment.

4.7 Perceived Lightness Results consistency and reliability of the results from all the three
As shown in Table IV, most participants identified Black, trials [21]. Analysis of variance (ANOVA) and Pearson’s
Dark Green, and Gray as appearing lighter in sample correlation, were used to analyze the data.
B (ink-jet printed sample). The majority of participants A significance level (denoted as α or alpha) of 0.05 was
identified Blue, Red, and Cyan as appearing lighter in sample used for statistical analyses. As shown in Tables V and VI, the
A (screen-printed sample). All participants (100 percent) significance level of the seven individual colors and overall
identified Blue and Cyan as appearing lighter in sample difference were found to be greater than 0.05, supporting the
A. For Yellow, results indicated a color shade change in
conclusion that the scores for these three trials are consistent
grayscale.
and can be interpreted with confidence.
4.8 Correlation and Reliability Statistics The Pearson correlation coefficient is a measure of
Although the viewing experiment was effectively controlled the linear dependence between two variables X and Y.
through the experimental method, variation among ob- Coefficients have values between +1 and −1 inclusive, where
servers may still exist, such as mood, fatigue, or stress level, 1 is total positive linear correlation, 0 is no linear correlation,
which may uniquely impact different trials [15, 18], influence and −1 is total negative linear correlation. As shown in
the researchers, or the experimental environment. For this Table VII, all of the correlation coefficients are positive and
reason, statistical analysis was conducted to examine the all seven colors of Trials 2 and 3 are positively correlated,

J. Imaging Sci. Technol. 040402-8 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Table IV. Perceived Lightness Difference. Table V. Individual Color Difference—Analysis of Variance (ANOVA).

Lightness Trial 1 (%) Trial 2 (%) Trial 3 (%) SS DF MS F Sig

A<B 75.0 75.0 67.0 Color 1 Between Groups 0.14 2 0.007 0.079 0.924
Color 1 B<A 17.0 25.0 33.0 Within Groups 2.896 33 0.088
A=B 8.0 0.0 0.0 Total 2.910 35
A<B 33.0 33.0 25.0 Color 2 Between Groups 0.097 2 0.049 0.405 0.670
Color 2 B<A 42.0 50.0 75.0 Within Groups 3.958 33 0.120
A=B 25.0 17.0 0.0 Total 4.056 35
A<B 75.0 83.0 83.0 Color 3 Between Groups 0.875 2 0.437 2.287 0.117
Color 3 B<A 8.0 8.0 8.0 Within Groups 6.313 33 0.191
A=B 17.0 8.0 8.0 Total 7.188 35
A<B 83.0 92.0 92.0 Color 4 Between Groups 0.431 2 0.215 1.714 0.196
Color 4 B<A 17.0 8.0 8.0 Within Groups 4.146 33 0.126
A=B 0.0 0.0 0.0 Total 4.576 35
A<B 0.0 8.0 0.0 Color 5 Between Groups 0.097 2 0.049 0.255 0.776
Color 5 B<A 100.0 92.0 100.0 Within Groups 6.292 33 0.191
A=B 0.0 0.0 0.0 Total 6.389 35
A<B 25.0 25.0 17.0 Color 6 Between Groups 0.181 2 0.090 0.317 0.730
Color 6 B<A 67.0 75.0 83.0 Within Groups 9.396 33 0.285
A=B 8.0 0.0 0.0 Total 9.576 35
A<B 0.0 8.0 0.0 Between Groups 0.500 2 0.250 1.333 0.277
Color 7 B<A 100.0 92.0 100.0 Color 7 Within Groups 6.188 33 0.188
A=B 0.0 0.0 0.0 Total 6.688 35

which indicated that these two trials appeared to be in the Table VI. Overall Difference—Analysis of Variance (ANOVA).
most agreement. This could be a result of the learning process
of the participants. In total, 14 of the 21 tests indicated SS DF MS F Sig
significant values (P values <0.05) and two additional tests
suggest significant value for the one-tail test. Overall appearance of color Between Trials 0.500 2 0.250 0.868 0.429
Within Trials 9.500 33 0.288
Total 10.000 35
5. CONCLUSIONS
This study succeeded in developing an optimal workflow for Line Quality Between Trials 0.056 2 0.028 0.061 0.941
the replication of screen-printed fabrics via ink-jet textile Within Trials 14.917 33 0.452
printing. The process is streamlined and effective and can Total 14.972 35
be used in the future by researchers and practitioners. A
Visual Texture Between Trials 0.500 2 0.250 0.611 0.549
visual assessment instrument and protocol were established
Within Trials 13.500 33 0.409
to conduct expert visual assessment for digital textile printed
Total 14.000 35
media. The ANOVA statistics and Pearson correlation
reliabilities indicated no significant difference between the Overall appearance Between Trials 0.056 2 0.028 0.121 0.887
three trials and the results are consistent in supporting Within Trials 7.583 33 0.230
validity among the observed data. For the color difference Total 7.639 35
comparison, Black was perceived to have the highest match; Overall Matching Between Trials 0.167 2 0.083 0.148 0.863
however, the target color was not a true black. Yellow, Dark
Within Trials 18.583 33 0.563
Green, Gray, and Red were in the middle group. Blue and
Total 18.750 35
Cyan had the lowest perceived match. Overall, the primary
colors chosen for evaluation from the screen-printed sample
were well matched by the ink-jet printed sample. For the
assessment of scale, line quality, visual texture, and overall However, the mean of line texture and visual texture was
appearance, expert participants agreed there was an adequate slightly lower than the other aspects, which may indicate that
match between two printed samples based on data analysis. digital printing technology at this stage cannot completely

J. Imaging Sci. Technol. 040402-9 July-Aug. 2019

 Wang et al.: Replication of screen-printing fabric via ink-jet textile printing

Table VII. Perceived Color Lightness Difference—Pearson Correlation Reliabilities screen-printed fabric, digital files, and raw material. Hunts-
man is acknowledged for providing technical expertise.
T1&T2 T1&T3 T2&T3 Lubrizol is acknowledged for providing the pretreatment.

Color 1 Pearson Correlation 0.444 0.204 0.683

Sig. (2-tailed) 0.149 0.525 0.014 REFERENCES
1 C. Kan and C. Yuen, Res. J. Textile Apparel 16, 1 (2012).
N 12 12 12 2 T. Abe, Adv. Mater. Res. 441, 23 (2012).
3 J. L. Parsons, Clothing Textiles Res. J. 33, 280 (2015).
Color 2 Pearson Correlation 0.736 0.502 0.750
4 J. Ren, G. Chen, and X. Li, Comput. Industry 92–93, 152 (2017).
Sig. (2-tailed) 0.006 0.096 0.005 5 M. Kanik and P. J. Hauser, Coloration Technology 119, 230 (2003).
N 12 12 12 6 C. W. M. Yuen, S. K. A. Ku, P. S. R. Choi, C. W. Kan, and S. Y. Tsang,

Color 3 Pearson Correlation 0.449 0.449 1.000 ‘‘Determining functional groups of commercially available ink-jet
printing reactive dyes using infrared spectroscopy,’’ Res. J. Textile Apparel
Sig. (2-tailed) 0.143 0.143 0.000 9, 26–38 (2005).
N 12 12 12 7 G. Savvidis, E. Karanikas, N. Nikolaidis, I. Eleftheriadis, and
E. Tsatsaroni, ‘‘Ink-jet printing of cotton with natural dyes,’’ Coloration
Color 4 Pearson Correlation 0.674 0.674 1.000 Technology 130, 200–204 (2014).
Sig. (2-tailed) 0.016 0.016 0.000 8 S. Ervine, B. Siegel, and K. S. Meyer, ‘‘APPLIED TECHNOLOGY-A
Simple, Universal Approach to Ink Jet Printing Textile Fabrics,’’ Textile
N 12 12 12
Chemists and Colorists and American Dyestuff Reporter 32, 26–27
Color 5 Pearson Correlation 0.110 0.496 0.721 (2000).
9 D. Clark, ‘‘Applications of digital ink-jet printing on textiles,’’ AATCC
Sig. (2-tailed) 0.733 0.101(.05)* 0.008 Review 3, 14–16 (2003).
N 12 12 12 10 H. Liu, S.-H. Gao, J.-S. Cai, C.-L. He, J.-J. Mao, T.-X. Zhu, Z. Chen,
J.-Y. Huang, K. Meng, K.-Q. Zhang, S. S. Al-Deyab, and Y.-K. Lai, ‘‘Re-
Color 6 Pearson Correlation 0.949 0.735 0.775 cent progress in fabrication and applications of superhydrophobic coating
Sig. (2-tailed) 0.000 0.006 0.006 on cellulose-based substrates,’’ Materials 9, 124 (2016).
11 T. T. Lamminmäki, J. P. Kettle, and P. A. C. Gane, ‘‘Absorption and ad-
N 12 12 12
sorption of dye-based inkjet inks by coating layer components and the
Color 7 Pearson Correlation 0.349 0.451 0.508 implications for print quality,’’ Colloids Surf. Physicochem. Eng. Aspects
380, 79–88 (2011).
Sig. (2-tailed) 0.266 0.141 0.092(.046)* 12 M. Singh, H. M. Haverinen, P. Dhagat, and G. E. Jabbour, ‘‘Inkjet
N 12 12 12 printing—process and its applications,’’ Adv. Mater. 22, 673–685 (2010).
13 R. W. G. Hunt and M. R. Pointer, Measuring Colour (John Wiley & Sons,
*Note: Significant value for one-tail test
Hoboken, NJ, 2011).
14 M. D. Fairchild, Color Appearance Models (John Wiley & Sons, Hoboken,
NJ, 2013).
15 A. F. Cabalar and Z. Karabash, ‘‘California bearing ratio of a sub-base
express the thick layering effect produced by traditional material modified with tire buffings and cement addition,’’ J. Testing
printing technology [5, 14, 20, 21]. In general, all participants Evaluation 43, 1279–1287 (2014).
16 R. Shamey, L. M. Cardenas, D. Hinks, and R. Woodard, ‘‘Comparison of
agreed that the digital printed sample well matched the naïve and expert subjects in the assessment of small color differences,’’ J.
screen-printed sample, as judged by color difference, scale, Opt. Soc. Am. A 27, 1482–1489 (2010).
line quality, visual texture, and overall appearance. From 17 J. P. Hwang, S. Kim, and C. K. Park, ‘‘Development of a color matching
this perspective, the results suggest promise for using DTP algorithm for digital transfer textile printing using an artificial neural
network and multiple regression,’’ Text. Res. J. 85, 1076–1082 (2015).
samples as substitutes for screen-printed samples in the 18 Annual Book of ASTM Standards, West Conshohocken, PA, USA, 6 (1),
future which can support quicker response to consumer 13-23 (2016).
preferences and contribute to advances in the product 19 ‘‘AATCC Technical Manual,’’ Evaluation Procedure 1 Gray Scale
development process for a vast range of textile products. for Color Change,’’ (AATCC, Research Triangle Park, NC, 2013),
pp. 407–408.
20 M. Wang, ‘‘Replication of screen printing via ink-jet textile printing,’’ M.S.
ACKNOWLEDGMENT thesis (North Carolina State University, 2017).
21 S. Mihok, D. A. Carlson, E. S. Krafsur, and L. D. Foil, ‘‘Performance of the
The Walmart Innovation Fund is acknowledged for funding Nzi and other traps for biting flies in North America,’’ Bull. Entomol. Res.
this work. Spring Creative is acknowledged for providing the 96, 387–397 (2006).

J. Imaging Sci. Technol. 040402-10 July-Aug. 2019

View publication stats