Implementation of HSV Imagery with K-Nearest Neighbor for Classification of Maturity Levels in Tomatoes

Authors

  • Lailatul Husna Aulia a:1:{s:5:"ar_IQ";s:39:"Universitas Islam Negeri Sumatera Utara";}
  • Fazar Azhari Universitas Islam Negeri Sumatera Utara
  • Muhammad Dhuha Bimantara Universitas Islam Negeri Sumatera Utara

Keywords:

KNN, HSV, classification

Abstract

The tomatoes used use tomatoes (Lycopersicum esculentum Mill) which is a type of horticultural plant. One type is plum tomatoes. The process of classifying tomato ripeness is carried out manually through direct visual observation. However, this is very difficult to do because it is inconsistent. Therefore, relevant features are needed to classify tomato maturity levels based on HSV features using the KNN method. The method used in classification is K-Nearest Neighbor, this algorithm requires features to build the model. The feature used is HSV feature extraction. Based on the results of research tests carried out, it proves that Euclidean distance k=2 has a percentage value of 85%. Based on the level of accuracy, the color feature k=2 shows the best k value in classifying tomato ripeness levels measuring 400x400 pixels. To achieve a high level of accuracy, image processing time should be less

References

A. Marliah, M. Hayati, and I. Muliansyah, "Utilization of Liquid Organic Fertilizer on the Growth and Yield of Several Varieties of Tomato (Lycopersicum esculentum L.)," J. Agrista, vol. 16, no. 3, pp. 122–128, 2012.

S. Y. Riska, "Classification of Tomato Ripeness Levels Based on Differences in Image Improvement Using Average RGB and Pixel Index," J. Ilm. Technol. Inf. Asia, vol. 9, no. 2, pp. 18–26, 2015.

S. Sanjaya, M. L. Pura, S. K. Gusti, F. Yanto, and F. Syafria, "K-Nearest Neighbor for Classification of Tomato Maturity Level Based on Hue, Saturation, and Value Colors," Indonesia. J. Artif. Intel. Data Min., vol. 2, no. 2, p. 101, 2019, doi: 10.24014/ijaidm.v2i2.7975.

S. Y. Riska and P. Subekti, "Classification of Tomato Fruit Ripeness Levels Based on Color Features Using Multi-Svm," J. Ilm. Inform., vol. 1, no. 1, pp. 39–45, 2016, doi: 10.35316/jimi.v1i1.442.

N. Atrianda, "Classification of Tomato Fruit Ripeness with Variations in Color Models Using Support Vector Machine," VOCATECH Vocat. Educ. Technol. J., vol. 1, no. 2, pp. 45–52, 2020, doi: 10.38038/vocatech.v1i2.27.

R. Pratama, A. Fuad, and F. Tempola, "Detection of Tomato Fruit Ripeness Based on Color Features Using the His Color Space Transformation Method," JIKO (Journal of Information and Computers), vol. 2, no. 2, pp. 81–86, 2019, doi: 10.33387/jiko.v2i2.1318.

N. Nafi'iyah, "Classification of Mango Fruit Ripeness Based on HSV Image with KNN," J. Elektron. List. and Technol. Inf. Terap., vol. 1, no. 2, pp. 4–7, 2019.

A. Chan, P. Liem, N. P. Wong, and T. Gunawan, "Fruit Segmentation Using the K-Means Clustering Method and Identification of Ripeness Using the Color Content Comparison Method," J. SIFO Mikroskil, vol. 15, no. 2, pp. 91–100, 2014, doi: 10.55601/jsm.v15i2.156.

L. Ningsih and P. Cholidhazia, "Classification Of Tomato Maturity Levels Based on RGB And HSV Colors Using KNN Algorithm," RIGGS J. Artif. Intel. Digits. Bus., vol. 1, no. 1, pp. 25–30, 2022, doi: 10.31004/riggs.v1i1.10.

I. A. A. Angreni, S. A. Adisasmita, and M. I. Ramli, "Regarding the Accuracy Level of Road Damage Identification," vol. 7, no. 2, pp. 63–70, 2018.

D. Syahid, J. Jumadi, and D. Nursantika, "Classification System for Philodendron Leaf Ornamental Plant Types Using the K-Nearest Neighbor (KNN) Method Based on Hue, Saturation, Value (HSV)," J. Online Inform. , vol. 1, no. 1, p. 20, 2016, doi: 10.15575/join.v1i1.6.

M. G. R. Miftahus Sholihin, "Egg Quality Classification Based on Color Features Using the K-Nearest Neighbor Method," Fak. Technol. Inf. UNMER Malang, pp. 1188–1193, 2018.

H. Parvin, H. Alizadeh, and B. Minati, “A Modification of K-Nearest Neighbor Classifier,” Glob. J. Comput. Sci. Technol., vol. 10, no. 14, pp. 37–41, 2010.

M. Manchanda and R. Sharma, “Fusion of visible and infrared images in HSV color space,” 3rd IEEE Int. Conf. , pp. 1–6, 2017, doi: 10.1109/CIACT.2017.7977384.

S. Sanjaya, "Application for Recognizing Ripeness Levels of Tomatoes Using Android-Based Hsv Color Features," J. Teknoinfo, vol. 16, no. 1, p. 26, 2022, doi: 10.33365/jti.v16i1.1489.

A. Salsabila, R. Yunita, and C. Rozikin, "Flower Type Image Identification using the KNN Algorithm with HSV Color Extraction and GLCM Texture," Technomedia J., vol. 6, no. 1, pp. 124–137, 2021, doi: 10.33050/tmj.v6i1.1667.

M. A. Anggriawan, M. Ichwan, and D. B. Utami, "Recognition of Tomato Ripeness Levels Based on Color Images in a Case Study of Automatic Selection System Development," J. Tek. Inform. and Sis. Inf., vol. 3, no. 3, pp. 550–564, 2017, doi: 10.28932/jutisi.v3i3.688.

J. Harefa and M. Pratiwi, "Comparison Classifier: Support Vector Machine (SVM) and K-Nearest Neighbor (K-NN) In Digital Mammogram Images," Juisi, vol. 02, no. 02, pp. 35–40, 2016.

A. P. N, M. M. R. Wulanningrum, M. Kom, U. Nusantara, and P. Kediri, "IDENTIFICATION OF TOMATO FRUIT MATURITY USING THE K-NEARST NEIGHBOR (K-NN) ALGORITHM IKA WAHYUN NINGSIH The progress of human information is developing at a rapid pace "In technology, life and agriculture are getting better in terms of quality," vol. 01, no. 08, pp. 3–8, 2017.

M. Lumb and P. Sethi, “Texture Feature Extraction of RGB, HSV, YIQ and Dithered Images using GLCM, Wavelet Decomposition Techniques,” Int. J. Comput. Appl., vol. 68, no. 11, pp. 25–31, 2013, doi: 10.5120/11623-7087.

S. Sanjaya et al., “K-Nearest Neighbor for Classification of Tomato Maturity Level Based on Hue , Saturation , and Value Colors,” vol. 2, no. 2, pp. 101–106, 2019.

U. K. N. N. Algorithm, “Classification Of Tomato Maturity Levels Based on RGB And HSV Colors Using KNN Algorithm 1,” vol. 1, no. 1, pp. 25–30, 2022.

V. Education and T. Journal, “Vocational education and technolgy journal,” vol. 1, no. 2, 2020.

R. Pratama, A. Fuad, andF. Tempola, "Detection of Tomato Fruit Ripeness Based on Color Features Using the His Color Space Transformation Method," IKO (Journal of Informatics and Computers)., vol.2, no. 2, pp. 81-86, 2019, doi: 10.33387/jiko.v2i2.1318.

S. Aprilisa and Sukemi, "Classification of Tomato Fruit Ripeness Levels Based on Color Features," Pros. Annu. Res. Semin. 2019, vol. 5, no. 1, pp. 978–979, 2019.

K. A. Pratama, W. P. Atmaja, and V. Lusiana, "Classification of Kersen Fruit Maturity Levels Using HSI Image with the K-Nearest Neighbor (KNN) Method. Kersen is a plant that has small red and sweet fruit like cherries. Kersen plants are a very common type of tree," vol. 11, no. 1, pp. 105–108, 2022.

D. I. Muhammad, E. Ermatita, and N. Falih, "Using K-Nearest Neighbor (KNN) to Classify Starfruit Images Based on Color Features," Inform. J. Computer Science., vol. 17, no. 1, p. 9, 2021, doi: 10.52958/iftk.v17i1.2132.

S. P. Adenugraha, V. Arinal, and D. I. Mulyana, "Classification of Ambon Banana Ripeness Using KNN and PCA Methods Based on RGB and HSV Images," J. Media Inform. Budidharma, vol. 6, no. 1, p. 9, 2022, doi: 10.30865/mib.v6i1.3287.

Downloads

Published

2023-07-31

How to Cite

Aulia, L. H., Azhari, F. ., & Bimantara, M. D. . (2023). Implementation of HSV Imagery with K-Nearest Neighbor for Classification of Maturity Levels in Tomatoes. Bigint Computing Journal, 1(2), 62–69. Retrieved from https://journal.aira.or.id/index.php/bigint/article/view/779

Issue

Section

Article