The Journal of the Convergence on Culture Technology (문화기술의 융합)

The International Promotion Agency of Culture Technology (국제문화기술진흥원)
권호 표지
DOI QR코드

DOI QR Code

Analysis of Apple Colors and Sugar Contents Using Linear Regression

선형회귀를 이용한 사과의 색상과 당도 분석

  • 김선종 (부산대학교 IT응용공학과)
  • Received : 2021.12.27
  • Accepted : 2022.01.08
  • Published : 2022.01.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, the relationship between RGB, HSV, La*b* colors and sugar content was analyzed using linear regression on apples harvested in the same region. First, as a result of examining the correlation coefficient with sugar content according to each color level, it was found that the (+) region having a positive coefficient and a (-) region having a negative coefficient were separated according to the color level. Also, the correlation coefficient between color and sugar content, represented by the average value, was 0.342 in the La*b* color space, which was higher than the coefficient in the RGB and hsv space. That is, this means that the sugar content is related to the color in the La*b* space. Also, in the complex color composed of regions with high sugar content, it was found to be R2=0.3627, indicating that it is related to sugar content. In all nine color spaces, it was found to be R2=0.3668. In this case, it was found that the coefficients of v and b* had an effect on the sugar content. Due to this, it was possible to confirm the validity of the empirical prediction that the higher the b* representing yellow, the higher the sugar content.

본 논문에서는 같은 지역에서 수확된 사과 영상에 대해 선형회귀를 이용하여 RGB, HSV, La*b* 색상과 당도와의 연관 관계를 분석하였다. 먼저, 각 색상 레벨에 따른 당도와의 상관계수를 조사한 결과, 색상 레벨에 따라 양의 계수를 갖는 (+) 영역과 음의 계수를 갖는 (-) 영역으로 구분됨을 알 수 있었다. 또한 평균값으로 대표되는 색상과 당도와의 상관계수는 La*b* 색상 공간에서 0.342로 RGB, hsv 공간에서의 계수보다 높게 나타났다. 즉, 이는 당도는 La*b* 공간에서의 색상과 관계가 있다는 것을 의미하고 있다. 또한 당도와 관련이 높은 영역으로 구성된 복합 색상에서는 R2=0.3627로 나타났으며, 이는 당도와 관련이 있음을 보여주고 있다. 9개 모든 색상 공간에서는 R2=0.3668로 나타났다. 이 경우 v와 b*의 계수가 당도에 영향이 있음을 알 수 있었다. 이로 보아 노란색을 대변하는 b*가 높을수록 당도도 높게 나타난다는 경험적인 예측의 타당성을 확인할 수 있었다.

Keywords

Acknowledgement

이 과제는 부산대학교 기본연구지원사업(2년)에 의하여 연구되었음.

References

  1. Bhargava A. and Bansal A. "Fruits and vegetables quality evaluation using computer vision: A review," Journal of King Saud University -Computer and Information Sciences (2018). https://doi.org/10.1016/j.jksuci.2018.06.002.
  2. Hayrettin Toylan and Hilmi Kuscu, "A Real-Time Apple Grading System Using Multicolor Space," Hindawi Publishing Corporation The Scientific World Journal Volume 2014. https://doi.org/10.1155/2014/292681
  3. G. C. Bora, R. Pathak, M. Ahmadi, and P. Mistry, "Image processing analysis to track colour changes on apple and correlate to moisture content in drying stages," Food Quality and Safety, 2018, 2, 105-110. https://doi.org/10.1093/fqsafe/fyy003
  4. Z. A. Kus, B. Demir, I. Eski, F. Gurbuz, and S. Ercisli, "Estimation of the Colour Properties of Apples Varieties Using Neural Network," Erwerbs-Obstbau (2017) 59:291-299. DOI:10.1007/s10341-017-0324-z
  5. S. N. Subhashree, S. Sunoj, J. Xue, and G. C. Bora, "Quantification of browning in apples using colour and textural features by image analysis," Food Quality and Safety, 2017, 00, pp. 1-6. https://doi.org/10.1093/fqsafe/fyx021
  6. X. Cheng, T. Pang, H. Tao, M. Lin, and H. Yang "Prediction of apple internal qualities using hyperspectral imaging techniques," Int. Conf. Adv. Mechatronics Syst., 450-455, 2017, DOI:10.1109/ICAMechS.2017.8316516
  7. J.-H. Lee, J.-H. Park, and S.-J. Kim, "A New Similarity Measurer for Color-Texture and Its Clustering for Apple," International Journal of Signal Processing Systems, Vol. 8, No. 3, pp. 59-63, September 2020. DOI:10.18178/ijsps.8.3.59-63.
  8. Soo-Mok Jung, "Color Image Watermarking Technique using Adjacent Pixels and Spatial Encryption Technique," The Journal of the Convergence on Culture Technology (JCCT) Vol. 7, No. 3, pp.863-867, Aug. 2021. https://doi.org/10.7236/IJIBC.2021.13.1.187
  9. J.-W. Oh, H.-K. Kim, and I.-T. Kim, "Design and Implementation of Fruit harvest time Predicting System based on Machine Learning," Smart Media Journal, Vol. 8, No. 1, pp. 74-81, 2019. https://doi.org/10.30693/SMJ.2019.8.1.74
  10. R. Lu, "Predicting Firmness and Sugar Content of Sweet cheeies Using Near-infrared Diffuse Reflectance Spectroscopy" Transactions of the ASAE, Vol. 44(5): 1265-1271, 2001. DOI: 10.13031/2013.6421
  11. S.-J. Kim, "Association analysis of sugar contents based on color information of apple fruit images," Journal of KIIT, Vol. 18, No. 4, pp. 41-49, 2020. DOI:10.14801/jkiit.2020.18.4.41