Implementation of Texture Analysis and Artificial Intelligence-Based Tomato Fruit Features Estimation Algorithm 


Vol. 48,  No. 10, pp. 1321-1329, Oct.  2023
10.7840/kics.2023.48.10.1321


PDF
  Abstract

Smart farm refers to a technology that maintains and manages a growth environment by applying ICT technology to a greenhouse, and can increase productivity compared to labor force. Recently, artificial intelligence technology has been applied to various parts of smart farm, but research related to weight and Brix degree estimation, which are fruit features that affects production management, is insufficient. In this paper, we implement an algorithm that estimates the weight and Brix degree of tomatoes, a type of crop produced in smart farms, using texture analysis and artificial intelligence. The tomato fruit features estimation algorithm uses Dissimilarity, Homogeneity, Energy, Correlation and number of tomato pixels obtained by collecting images of tomatoes harvested in the greenhouse in a laboratory environment and applying the Gray Level Co-occurrence Matrix(GLCM) that used for texture analysis. The collected features are used as input data for eXtreme Gradient Boosting(XGBoost), Deep Neural Network(DNN), Convolutional Neural Network(CNN) and machine learning ensemble methods to estimate weight and Brix degree, and Root Mean Square Error(RMSE) and R2 are used as evaluation indicators to derive the most optimal method. As a result of the experiment, it was confirmed that the machine learning ensemble obtained the best result, and deep learning-type DNN and CNN gave poor results.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

J. Lee, J. Baek, D. Im, T. Kim, M. Kim, S. Park, O. Yang, "Implementation of Texture Analysis and Artificial Intelligence-Based Tomato Fruit Features Estimation Algorithm," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 10, pp. 1321-1329, 2023. DOI: 10.7840/kics.2023.48.10.1321.

[ACM Style]

Jeong-Ho Lee, Jeong-Hyun Baek, Dong-Hyeok Im, Tae-Hyun Kim, Man-Jung Kim, Seong-Jin Park, and Oh-Seok Yang. 2023. Implementation of Texture Analysis and Artificial Intelligence-Based Tomato Fruit Features Estimation Algorithm. The Journal of Korean Institute of Communications and Information Sciences, 48, 10, (2023), 1321-1329. DOI: 10.7840/kics.2023.48.10.1321.

[KICS Style]

Jeong-Ho Lee, Jeong-Hyun Baek, Dong-Hyeok Im, Tae-Hyun Kim, Man-Jung Kim, Seong-Jin Park, Oh-Seok Yang, "Implementation of Texture Analysis and Artificial Intelligence-Based Tomato Fruit Features Estimation Algorithm," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 10, pp. 1321-1329, 10. 2023. (https://doi.org/10.7840/kics.2023.48.10.1321)
Vol. 48, No. 10 Index

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880