Sliced cheese is widely consumed because of its convenience, uniform shape and acceptable sensory properties. Its exposed surface and moisture content make it susceptible to quality changes during handling, distribution and storage including pH fluctuation, oxidation and visible color alteration. These surface changes are important because sliced cheese is commonly evaluated first by visual appearance before consumption. Color is one of the most direct and non-destructive indicators because it reflects physical and chemical changes on the food surface (Martins et al., 2024).
       
Natural pigment-based indicators have gained attention as alternatives to synthetic color indicators. Curcumin, the major bioactive pigment of Curcuma longa, has a strong yellow-orange color and pH-responsive behavior. Under acidic conditions, curcumin tends to appear yellow, whereas neutral to alkaline conditions may shift its color toward orange or reddish tones (Devi et al., 2025; Wang et al., 2025). Recent studies in dairy and food packaging systems also support the relevance of curcumin as a natural pigment for active and intelligent applications. Curcumin-alginate biopolymer indicators have been developed for pH-responsive freshness detection, while curcumin-enriched chitosan films have been applied to soft cheese preservation. In dairy products, Curcuma longa extract has also been evaluated as a coloring and functional ingredient in yoghurt systems (Kamble et al., 2026; Sheet et al., 2026; Benzineb et al., 2025).
       
Edible coating is a practical platform for incorporating natural indicators onto food surfaces. Gelatin forms transparent and flexible matrices, while glycerol improves flexibility and reduces brittleness. Curcumin extract obtained by Microwave-assisted extraction (MAE) offers an efficient pigment extraction approach and its incorporation into gelatin-glycerol coating may provide a color-responsive system that can be applied by pneumatic spray to obtain uniform distribution on sliced cheese (Oliveira Filho and Egea, 2022).
       
Color evaluation can be performed using CIELAB and RGB-based approaches. CIELAB provides standardized parameters, including L*, a* and b* and derived indices such as hue angle, chroma, ΔE, Whiteness Index and Browning Index (Pathare et al., 2013; Solano-Doblado et al., 2020). RGB-based analysis provides practical digital parameters such as RGB values, HEX code, luminance, HSV and RGB-based color distance. These digital metrics are useful because they translate visible color changes into numerical values that can be compared across treatments. The combination of standardized and digital color spaces is relevant for edible coating research because instrumental color values can be linked with visual interpretation, while RGB-based outputs are easier to communicate for practical applications. Although curcumin-based films and coatings have shown visible color responses, comparative evaluation of CIELAB and RGB-based metrics for curcumin-infused edible coating on cheese remains limited. This study developed a gelatin-glycerol edible coating enriched with MAE-derived curcumin extract and compared CIELAB and RGB-based color metrics in assessing its potential as a natural colorimetric indicator for sliced cheese systems.

The study was carried out at the Laboratory of Animal Product Technology, Faculty of Animal Science and Technology, Universitas Brawijaya, Malang, East Java, Indonesia. The preparation of turmeric powder was conducted at UPT Herbal Materia Medica Laboratory, Batu, East Java, Indonesia. The experiment was conducted from 21 July 2025 to 29 January 2026.
 
Materials
 
Gelatin (RandW), food-grade glycerol, ethanol 96%, distilled water and turmeric rhizomes from Bungah, Gresik, East Java. The main equipment included a hot plate magnetic stirrer (SH-2), modified microwave extractor (Sharp), pH meter (Hanna), viscometer (Brookfield) and colorimeter (CS-10).
 
Curcumin extraction (MAE method)
 
Curcumin extract was prepared using MAE. Fresh turmeric rhizomes were washed, sliced, dried at 50°C and ground into powder. Turmeric powder was mixed with ethanol at a 1:33.3 (w/v) ratio and macerated at 4°C for 24 h. Extraction was performed at 70°C for 10 min using intermittent microwave heating, consisting of 1-min heating followed by 2-min pauses to maintain temperature below 80°C. The extract was cooled, filtered and stored in a dark container at 4°C (Rahayu et al., 2015).
 
Edible coating preparation and application
 
The edible coating solution was prepared in a final volume of 100 mL. Each formulation contained 8 g gelatin, 2 g glycerol and curcumin extract according to treatment: P0 = 0 mL, P1 = 1 mL and P2 = 2 mL. Distilled water was added to reach 100 mL final volume, corresponding to approximately 90, 89 and 88 mL initial water for P0, P1 and P2, respectively. Distilled water was heated to 45°C, gelatin was gradually added and stirred for 30 min to support hydration and homogeneous dispersion, glycerol was added and homogenized for 15 min and curcumin extract was then added and stirred for another 15 min. The total mixing sequence was used to obtain a uniform film-forming solution before application. The homogeneous solution was applied uniformly onto sliced cheese using a pneumatic spray system before color measurement.
 
pH and viscosity
 
The pH of 50 mL edible coating solution was measured using a calibrated pH meter with pH 4 and 7 buffers. The electrode was immersed in the homogeneous sample and the value was recorded after stabilization according to AOAC method 981.12 (AOAC, 2019). Viscosity was measured using a Brookfield viscometer. A 75 mL sample was placed in the chamber and measured using an appropriate spindle and rotation speed. Results were expressed as cP (Brookfield Engineering Laboratories, 2017).
 
CIELAB color analysis
 
Color was measured using a CS-10 colorimeter and expressed as L*, a* and b*. The L* value indicates lightness, a* represents the red-green axis and b* represents the yellow-blue axis. The following CIELAB-derived parameters were calculated (Pathare et al., 2013; Maskan, 2001):

 
RGB, HSV and color-distance analysis
 
RGB values were obtained from the colorimeter output and converted into digital components for HEX, luminance, HSV, chromatic intensity and color-distance analysis. HEX was expressed as #RRGGBB, while luminance (Y) was calculated from R, G and B values (Gonzalez and Woods, 2018):
 

Y = 0.299R + 0.587G + 0.114B

 
Before conversion to HSV, the RGB values are normalized to the range 0-1: Before HSV conversion, RGB components were normalized as R′ = R/255, G′ = G/255 and B′ = B/255. Cmax, Cmin and Δ were determined from the normalized RGB values. Hue was calculated according to the maximum normalized component, while Saturation and Value were calculated as follows (Gonzalez and Woods, 2018):

If Δ = 0, then: H = 0

ΔHue = |H_i - H₀|



V = Cmax × 100

ΔRGB - Intensity = max(R, G, B) - min(R, G, B)

 
Multi-criteria nd statistical analysis
 
The De Garmo method was used to select the optimal treatment based on pH, viscosity, b*, Chroma, Saturation, ΔE and ΔRGB. Each parameter was normalized and weighted according to its criterion type and the highest total score indicated the optimal formulation (DeGarmo et al., 1997). Data were expressed as mean ± standard deviation from three replicates. Treatment effects were analyzed using one-way ANOVA at p<0.05, followed by Duncan’s multiple range test. Pearson correlation was used to evaluate associations between pH and color parameters. PCA visualized treatment separation based on combined physicochemical and color variables. This analysis was included to identify whether individual trends observed in pH, viscosity and color parameters produced a consistent multivariate pattern among treatments. Statistical analysis and visualization were performed using Python in Google Colab and figures were exported at 300 dpi.

Edible coating characteristics
 
Table 1 shows that curcumin extract significantly affected pH (p<0.05). P0 had the highest pH (5.83±0.04), while P1 and P2 decreased to 5.75±0.02 and 5.71±0.05, respectively. This decrease may be related to phenolic and slightly acidic components in curcumin extract, which can modify the chemical environment of the gelatin-glycerol matrix (Estrella-Osuna et al., 2024). Although the numerical difference was small, the consistent decrease indicates that curcumin incorporation contributed to the physicochemical response of the coating solution.


       
Viscosity increased from 6.45±0.28 cP in P0 to 7.88±0.21 cP in P1 and 8.06±0.28 cP in P2. The higher viscosity suggests stronger interactions in the coating matrix after curcumin incorporation, possibly through hydrogen bonding between gelatin, glycerol and phenolic compounds in the extract. Appropriate viscosity may support coating deposition and layer stability during spray application (Ramos et al., 2016). Therefore, P2 showed the most pronounced physicochemical modification, with the lowest pH and highest viscosity, which is relevant for producing a more stable color-responsive coating layer.
 
Colorimetric response of curcumin-based edible coating
 
Curcumin extract clearly modified the CIELAB color parameters of the coating (Fig 1). P0 showed the highest L* value, indicating a brighter pale-cream surface, while L* decreased in P1 and P2 due to yellow-orange pigment deposition. The a* value shifted toward zero, indicating reduced green tone and movement toward the red-yellow region. The most pronounced change was observed in b*, which increased markedly after curcumin incorporation, confirming stronger yellowness. This increase shows that the main visual effect of the extract was expressed through yellow pigment dominance rather than general darkening alone. This result is consistent with the chromophoric structure of curcumin, which contributes to yellow-orange coloration (Han et al., 2023; Wang et al., 2025).


       
Chroma and Hue angle further supported the color shift (Fig 2). Chroma increased in the curcumin-treated samples, indicating stronger color saturation, while Hue angle decreased toward a more intense yellow-orange tone. The increase in ΔE (Fig 3) confirmed visible color differences from the control and demonstrated that curcumin addition produced color changes that were not only instrumentally detectable but also visually relevant. A large color difference is important for indicator development because the response must be sufficiently clear to be distinguished from the original cheese surface. Fig 4 shows that the Whiteness Index decreased, while the Browning Index increased. This BI increase should not be interpreted as product deterioration, but as a color-derived response to increase a* and b* values from curcumin pigment deposition (Pathare et al., 2013; Botalo et al., 2024).






       
Digital color analysis supported the CIELAB results (Table 2). P0 showed high and balanced R, G and B values, reflecting a pale cream appearance. Curcumin-treated samples showed reduced blue intensity and stronger red-green dominance, producing a more intense yellow color. HEX values shifted from pale cream-yellow tones in P0 to more saturated yellow tones in P1 and P2. The most intense color was observed in P2, particularly #DCBC3D, categorized as Strong Marigold Yellow. This HEX conversion provides a practical visual reference for describing the observed color change in digital form. Luminance decreased from P0 to P2, consistent with lower L* values and reduced reflected light as curcumin pigment intensity increased (Ranjbar et al., 2023; Sabat et al., 2025).


       
HSV results provided additional digital interpretation (Table 3). Hue values remained within the yellow range, indicating that curcumin strengthened the yellow tone without changing the primary color domain. Saturation increased markedly from P0 to P1 and P2, confirming a more vivid and distinguishable yellow color. Value decreased slightly but remained relatively high, indicating that the coating maintained adequate brightness while increasing color intensity. RGB chromatic intensity also increased with curcumin addition (Table 4). P0 showed low chroma, whereas P1 and P2 were categorized as high chroma treatments. The agreement among HSV saturation, RGB chromatic intensity and CIELAB-derived Chroma confirms that the increased color response was consistent across different color spaces. These results consistently show that curcumin enhanced yellowness, saturation and visual contrast (Phuangsaijai et al., 2021; Xie et al., 2024).




 
Relationship between color systems and treatment separation
 
The comparison between CIELAB and RGB metrics showed consistent treatment trends (Fig 5 and Fig 6). Hue_Lab had a wider numerical range than Hue_RGB, reflecting differences between perceptual and digital color spaces. Nevertheless, both systems indicated yellow dominance in curcumin-treated samples. A positive relationship was also observed between ΔE_Lab and ΔRGB. Although ΔRGB produced larger values because it is calculated using the broader RGB scale, both metrics increased from P0 to P2. This parallel trend is important because it suggests that RGB-based analysis can capture the same treatment response shown by CIELAB measurement. Therefore, RGB-based color distance can complement CIELAB analysis for evaluating curcumin-induced color changes (Rodríguez-Pulido et al., 2021; Ortiz et al., 2023).




       
The correlation heatmap (Fig 7) showed that pH was negatively associated with color intensity parameters, including b*, Chroma_Lab, Saturation and ΔRGB. Lower pH was therefore related to stronger yellow color expression in the curcumin-treated samples. In contrast, pH was positively associated with brightness parameters, such as L*, Value and luminance, indicating that brighter samples tended to have higher pH. Strong relationships among Chroma_Lab, Saturation and ΔRGB indicate that these parameters responded similarly to curcumin pigment intensity. The positive association between ΔE_Lab and ΔRGB also supports the complementary use of CIELAB, RGB and HSV approaches for evaluating color changes. These correlations suggest that physicochemical variation in the coating matrix was linked to measurable optical changes, although the relationship should be interpreted as an association rather than direct causation (Solano-Doblado et al., 2020; Li et al., 2024).


       
PCA confirmed treatment separation based on physicochemical and colorimetric parameters (Fig 8). PC1 explained 85.1% of total variance, while PC2 explained 7.5%, indicating that most variation among samples was captured by the first component. P0 was separated from P1 and P2 along PC1, showing the strong effect of curcumin incorporation on the overall coating profile. P2 occupied the most distinct position, consistent with higher b*, Chroma, Saturation, ΔRGB and viscosity, as well as lower pH. This multivariate pattern agrees with the univariate results and indicates that the measured parameters worked together in distinguishing the formulations. These findings indicate that curcumin addition produced coordinated changes in optical and physicochemical properties (Kumar et al., 2017; Han et al., 2023).


       
These findings indicate that the coating should be interpreted as a prototype color-responsive system rather than a fully validated freshness sensor. This limitation is important because shelf-life, microbial and storage validation were not included in the present study. The results demonstrate the ability of curcumin extract to generate measurable and visually distinguishable color changes on sliced cheese, while further validation is required before practical application. Future work should evaluate color stability during storage and relate visual changes to microbial or physicochemical quality indicators.
 
Optimal formulation formulation based on the De Garmo method
 
The De Garmo method integrated physicochemical and colorimetric parameters to determine the optimal formulation (Table 5). This approach is suitable for edible coating development because formulation performance is not determined by one variable only. In this study, the selected criteria represented both coating characteristics and visual response, including pH, viscosity and color intensity-related parameters. This integration was necessary because a formulation with strong color but poor coating characteristics would not be ideal for further development. P2 obtained the highest score (1.00), followed by P1 (0.85), while P0 had the lowest score (0.00).


       
The high score of P2 was mainly associated with lower pH, higher viscosity and stronger color response, as indicated by higher b*, Chroma, Saturation, ΔE_Lab and ΔRGB. The gradual score increase from P0 to P2 indicates that curcumin concentration progressively improved the coating characteristics. This result was consistent with the CIELAB, RGB, HSV, correlation and PCA findings, which all showed stronger pigment expression in P2. Therefore, the 2% curcumin extract formulation can be considered the most suitable treatment for further development as a natural color-responsive edible coating for sliced cheese applications.

This study demonstrated that gelatin-glycerol edible coating enriched with MAE-derived curcumin extract modified the physicochemical and colorimetric characteristics of sliced cheese. Curcumin addition decreased pH, increased viscosity and enhanced yellowness, saturation and visual contrast. CIELAB, RGB and HSV analyses showed consistent color trends, while correlation and PCA confirmed relationships among pH, brightness and chromatic intensity. De Garmo analysis selected 2% curcumin extract as the optimal formulation. The coating should be regarded as a prototype color-responsive system and further storage validation is needed to confirm its performance as a practical indicator. Curcumin-based edible coating has potential for further development as a natural color-responsive coating system for sliced cheese applications.

The present study was supported by Indonesia Endowment Fund for Education Agency (LPDP).
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of this content.

The authors declare that there are no conflicts of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish, or preparation of the manuscript.