Evaluation of Color Development Pattern on Pepper (Capsicum Annuum) Surface

Transcription

1 Evaluation of Color Development Pattern on Pepper (Capsicum Annuum) 1. Introduction Surface L. Baranyai, L.D. Dénes, G. Papucsek, J. Felföldi Corvinus University of Budapest, Department of Physics and Control, Hungary, Pepper (Capsicum annuum L.), or paprika, is important horticultural produce in Hungary. Production of paprika was 168,944 MT in Hungary in 2009 (FAOSTAT, 2009). Consumers are looking for vegetables with long shelf life, firm flesh, pungent taste and promising appearance. Carotenoid pigments of three mexican varieties were analyzed using an absorbance detector adjusted to 450 nm (Collera-Zúniga et.al, 2005). Important differences were found in both color and carotenoids composition. Carotenoid pigments were extracted from various sources and CIELAB a*b* color parameters were measured by Meléndez-Martínez et.al (2007). Alteration in functional groups resulted in significant difference in color. Additionally, 5 major color groups of 17 carotenoids were found based on their location on a*b* plane. Flavonoids, compounds important in antioxidant activity, were also analyzed using spectrophotometer at 515 nm (Ghasemnezhad et.al, 2011). Flavonoid concentration varied greatly among varieties at stages of full maturity and full ripeness. Chlorophyll fluorescence imaging was used to monitor changes in Chlorophyll activity of pepper during storage (Zsom et.al, 2010). It was found that storage temperature affected Chlorophyll activity and content the most. The optimal storage condition of 8 C and 90%RH was recommended (Guerra et.al, 2011) for CIELAB a* color improvement and minimal weight loss of 1-1.6% within 10 days. Acoustic stiffness method was used to follow firmness changes of 5 sweet pepper varieties during storage at 20 C (Muha et.al, 2005). Significant softening was found, especially after 5 days storage. During this experiment, total weight loss of 8-16% was observed, depending on the variety. The objective of the presented work was to analyze color changes on the surface of pepper and investigate the effect of storage condition and surface location on color development. 2. Materials and Methods 2.1. Materials Pepper (Capsicum annuum L.) sample of 70 pieces was received from retail. The sample was split into two equal groups using random selection. The first group was moved into fridge where temperature was adjusted to 10 C and controlled within ±2 C. This setup simulated the typical storage conditions applied by consumers. The second group was placed in room temperature of 24±4 C in order to induce color changes. Both groups were measured for 12 days on daily basis, except weekends.

2 2.2. Vision system The machine vision system consisted of a camera (Hitachi HV-C20 3CCD), illumination system (12 halogen lamps of 20W power) and zoom lenses (Canon TV). Digital images of pixel size with mm/pixel resolution were acquired and saved in bitmap format. Peppers were placed on black background in order to help automatic segmentation. Average intensity value of red, green and blue color components of the surface and derived parameters of hue, saturation and value were computed. Color standards, MOMCOLOR (white), (red), (green) and (blue), were also captured each day to control color stability and make necessary adjustments Spectrophotometers The ColorLite sph850 spectrophotometer scanned the spectra of nm in 10 nm steps and reported observed values in various color spaces: XYZ, L*a*b*, L*uv and xy. The light source of D65 was used with 2 incident angle. Measurements were performed in three replicates on each pepper to minimize noise. The Ocean Optics USB4000 spectrophotometer scanned the spectra of nm. This instrument was used to save the spectra for reference and wavelength selection. The average spectra of each day and storage group was calculated and compared to select the most appropriate wavelength ranges. Additionally, due to the small sampling area of spectrophotometers, three regions were used in measurements: tip cap, middle and shoulder Statistical analysis The open source software of R (version , R Foundation for Statistical Computing, Vienna, Austria) was used to perform statistical analysis, produce summary reports and charts. The effect of storage time, temperature, location and their interaction was investigated using ANOVA (analysis of variances). 3. Results and Discussion The surface color of pepper changed significantly during storage. Table 1 presents the summary of ANOVA analysis performed on color attributes. It was observed that red, green and blue intensity values changed more with storage time than HSI attributes, except intensity. This is in agreement with the dependence of intensity on the basic color components. Storage temperature affected hue and saturation the most, followed by green and blue color components. The significant values of time temperature interaction show that all measured and calculated features followed different tendency depending on the storage temperature. Figure 1 shows the mean values with 95% confidence interval for hue and saturation, the most affected color attributes. According to the hue values, both sample groups started from yellow color (approximately 60 ) and minimal changes were observed for cool stored (10±2 C) compared to the group of room temperature (24±4 C). The group stored at room temperature changed its color toward red (0 ) and this color became more and more vivid since saturation

3 was increased at the same time. Figures of hue and saturation show that groups were similar in color until 8 days of storage. Table 1. Statistical effect (ANOVA) of factors on color attributes Color attribute Time Temperature Interaction RGB Red ** ** RGB Green ** 8.91* 61.76** RGB Blue ** 8.83* 44.53** HSI Hue ** 35.33** 89.22** HSI Saturation ** 59.76** 49.47** HSI Intensity ** ** Significance codes: ** p<0.001, * p<0.01, + p<0.1 Figure 1. Changes of HSI Hue (left) and Saturation (right) during storage Similar tendency was found with linear normalization of red and green color components. Table 2 presents the Pearson correlation between parameters. Strong relationship was found between hue and normalized color components.

4 Table 2. Pearson correlation between normalized RGB and HSI components Color components Hue Saturation RGB normalized Red RGB normalized Green Average reflectance spectra were also calculated for sample groups for each storage days. It was observed that curves were running almost parallel. No difference was observed around 670 nm, the specific wavelength of Chlorophyll. The highest difference can be found around 540 nm, which is close to the specific wavelength of anthocyanins and flavonoids (Ghasemnezhad et.al, 2011; Lee et.al, 2008). Due to the small covered area of the optical sensor of spectrophotometers, three locations were compared. The result of ANOVA test is shown in Figure 2. Figure 2. Observed effects on reflectance readings at 540 nm by ANOVA Time Time x Location Temperature x Time Temperature x Location Location Temperature Temperature x Time x Location The effect of location on the surface was significant. This main effect is the second strongest contributing factor, while its interaction with storage time is the following. The reason of such difference between tip cap, middle and shoulder can be the semi transparent thin wall and cavity of fruit. This internal structure cannot influence digital image processing due to the calculation of average values for the whole visible surface. Statistical parameters of image attributes, such as deviation or contrast, may reflect color variations on the surface.

5 4. Conclusions Digital image processing technique was used to monitor changes of pepper samples during storage. Surface color measured with red, green and blue intensity values was primarily affected by storage time. Pieces became darker during the experiment, especially on room temperature (24±4 C). Significant effect of storage temperature was observed on HSI hue and saturation parameters. Analysis of spectra, collected by reference instruments, pointed out that sample groups differed the most around 540 nm. Probably the internal structure of fruit made spectrophotometer readings location dependent. This behavior should be taken into account in case of pepper. Machine vision system was not affected by local differences due to the calculated averages on the visible surface. Digital image processing is a rapid and robust technique can be used for monitoring purposes. Acknowledgements This research was supported by the New Széchenyi Plan TÁMOP-4.2.1/B-09/1/KMR. References Collera-Zúniga, O., F.G. Jiménez and R.M. Gordillo (2005): Comparative study of carotenoid composition in three mexican varieties of Capsicum annuum L. Food Chemistry, 90: FAOSTAT (2009): Food and Agricultural commodities production. Website visited 18 November Ghasemnezhad M., M. Sherafati and G.A. Payvast (2011): Variation in phenolic compounds, ascorbic acid and antioxidant activity of five coloured bell pepper (Capsicum annum) fruits at two different harvest times. Journal of Functional Foods, 3: Guerra, M., R. Magdaleno and P.A. Casquero (2011): Effect of site and storage conditions on quality of industrial fresh pepper. Scientia Horticulturae, in Press, doi: /j.scienta Lee, J., C. Rennaker and R.E. Wrolstad (2008): Correlation of two anthocyanin quantification methods: HPLC and spectrophotometric methods. Food Chemistry, 110: Meléndez-Martínez, A.J., G. Britton, I.M. Vicario and F.J. Heredia (2007): Relationship between the colour and the chemical structure of carotenoid pigments. Food Chemistry, 101: Muha, V., S. Istella and D. Tompos (2005): Storability of paprika varieties measured by nondestructive acoustic method. International Journal of Horticultural Science, 11(2): Zsom, T., V. Zsom-Muha, L. Baranyai, W.B. Herppich, J. Felföldi and Cs. Balla (2010): Nondestructive Determination of Post-harvest Ripening of Capsicum x annum Kárpia. Proceedings of the Third International Conference Postharvest Unlimited 2008, Potsdam, 4-7 November, Acta Horticulturae (ISHS 2010), 858: