Palm leaf nutrient deficiency detection using convolutional neural network (CNN)

Document Type : Research Paper


Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA Cawangan Melaka (Kampus Jasin), 77300 Merlimau, Melaka, Malaysia


Palm oil, scientifically known as Elaeis guineensis, is a rapidly growing commercial sector in Southeast Asia with a diverse economic composition. Palm oil plantations are crucial in economic activities and growth, as they generate employment in managing the palm oil quality. However, the lack of nutrients can affect the growth and quality of the crops. The manual detection of palm leaf nutrient deficiency can be one of the challenges as the visual symptoms of the deficiency demonstrate a similar representation. Thus, in this study, the palm leaf nutrient deficiency detection using Convolutional Neural Network (CNN) is proposed. CNN or ConvNet is a branch of deep neural networks in Deep Learning that is commonly used in analysing images and has proven to produce better feature extraction from dataset. A total of 350 images of healthy leaf and six types of palm leaf nutrient deficiency are Nitrogen, Potassium, Magnesium, Boron, Zinc, and Manganese were tested. The application of CNN to a variety of testing datasets returned good detection accuracy at 94.29\%. It can be deduced that the proposed implementation of CNN for palm leaf nutrient deficiency detection is found to be successful. Nonetheless, the number of datasets could be increased in the future to improve the detection performance.


[1] D. Adyanti, D. Novitasari and A.F. Aris, Support vector machine multiclass using polynomial kernel for osteoporosis detection, Proc. Int. Conf. Math. Islam 1 (2018) 384–390.
[2] S. Akram, M.U.D. Dar and A. Quyoum, Document image processing- A review, Int. J. Comput. Appl. 10(5)
(2010) 35–40.
[3] J. Amara, B. Bouaziz and A. Algergawy, A deep learning-based approach for banana leaf diseases classification,
BTW 2017 – Workshopband, Lecture Notes in Informatics (LNI), (2017) 79–88.
[4] B.A.M. Ashqar and S.S. Abu-naser, Image-based tomato leaves diseases detection using deep learning, Int. J.
Academic Engin. Res. 2(12) (2018) 10–16.
[5] P. Balasubramaniam and V.P. Ananthi, Segmentation of nutrient deficiency in incomplete crop images using
intuitionistic fuzzy C-means clustering algorithm, Nonlinear Dyn. 83(1–2) (2016) 849–866.
[6] R. Fern´andez-Escobar, M. Guerreiro, M. Benlloch and M. Benlloch-Gonz´alez, Symptoms of nutrient deficiencies
in young olive trees and leaf nutrient concentration at which such symptoms appear, Sci. Hortic. (Amsterdam).
209 (2016) 279–285.
[7] N. Leena and K.K. Saju, Vision based nutrient deficiency classification in maize plants using multi class support
vector machines, Int. Conf. Elect. Elect. Mater. Appl. Sci. (2018) 020048.
[8] J. Liu and X. Wang, Plant diseases and pests detection based on deep learning: A review, Plant Meth. 17(22)
[9] M.S. Munir and F.T. Anggraeny, Normal leaf knowledge on soybean plant based on threshold, Proc. IJCST 1(1)
(2017) 462–466.
[10] L.G. Nachtigall, R.M. Araujo and G.R. Nachtigall, Classification of apple tree disorders using convolutional neural
networks, Proc. IEEE 28th Int. Conf. Tools with Artif. Intell. 2016 (2016) 472–476.
[11] A.K. Nafiu, M.O. Abiodun, I.M. Okpara and V.O. Chude, Soil fertility evaluation: a potential tool for predicting
fertilizer requirement for crops in Nigeria, African J. Agric. Res. 7(47) (2001) 6204–6214.
[12] A.E.F. Neto, K.V.F. Boldrin and N.S. Mattson, Nutrition and quality in ornamental plants, Ornamental Hort.
21(2) (2015) 139–150.
[13] E. Recep, Science, Ecology and Engineering Research in the Globalizing World, St. Kliment Ohridski University
Press, 2018.[14] J.C. Russ, The Image Processing Handbook, Sixth Edition, CRC Press, 2011.
[15] N. Salkind, Ch5: Measurement reliability and validity, Explor. Res. 9th ed., (2015) 95–111.
[16] B.H. Shraddha, R.M. Shet, P. Nikita and N.C. Iyer, Nutrient detection for maize plant using noninvasive technique, 2018 Int. Conf. Adv. Commun. Comput. Technol. 2018 (2018) 504–509.
[17] T. Subetha, Dyadic human interaction recognition from videos using multi-layer 3D CNN, Int. J. Emerg. Tech.
11(3) (2020) 1033–1040.
[18] K. Sukvichai, C. Utintu and W. Muknumporn, Automatic speech recognition for thai sentence based on MFCC
and CNNs, Second Int. Symp. Instrument. Control, Artif. Intell. Robot. (2021) 1–4.
[19] S. Suresh and S. Lal, Modified differential evolution algorithm for contrast and brightness enhancement of satellite
images, Appl. Soft Comput. 61 (2017) 622–641.
[20] V. Vijay, P. Stuart, J. Clinton, S. Sharon and C. Erin, Palm oil is everywhere – but where did it come from?,
Environ. Sci. J. (2016) 1–6.
[21] P. Wang, E. Fan and P. Wang, Comparative analysis of image classification algorithms based on traditional
machine learning and deep learning, Pattern Recogn. Lett. 141 (2021) 61–67.
[22] D. Xiao, B.T. Le and T.T.L. Ha, Iron ore identification method using reflectance spectrometer and a deep neural
network framework, Spectrochimica Acta Part A: Molecular Biomol. Spectros. 248 (2021) 119168.
Volume 13, Issue 1
March 2022
Pages 1949-1956
  • Receive Date: 02 August 2021
  • Revise Date: 21 September 2021
  • Accept Date: 27 September 2021
  • First Publish Date: 21 November 2021