Soil color analysis based on a RGB camera and an artificial neural network towards smart irrigation: A pilot study. Issue 1 (January 2021)

• Record Type: Journal Article
• Authors: Al-Naji, Ali; Fakhri, Ahmed Bashar; Gharghan, Sadik Kamel; Chahl, Javaan
• Is Part Of: Heliyon. Volume 7:Issue 1(2021)
• Publication Date: 2021-01
• Subjects: Smart irrigation -- Computer vision system -- RGB color analysis -- Artificial neural network -- Feed-forward back propagation neural network; Research -- Periodicals; Medical sciences -- Periodicals; Natural history -- Periodicals; Social sciences -- Periodicals; Earth sciences -- Periodicals; Physical sciences -- Periodicals; 507.2
• DOI: 10.1016/j.heliyon.2021.e06078
• Languages: English
• ISSNs: 2405-8440

Abstract

Abstract: Irrigation operations in agriculture are one of the largest water consumers in the world, and it has been increasing due to rising population and consequent increased demand for food. The development of advanced irrigation technologies based on modern techniques is of utmost necessity to ensure efficient use of water. Smart irrigation based on computer vision could help in achieving optimum water-utilization in agriculture using a highly available digital technology. This paper presents a non-contact vision system based on a standard video camera to predict the irrigation requirements for loam soils using a feed-forward back propagation neural network. The study relies on analyzing the differences in soil color captured by a video camera at different distances, times and illumination levels obtained from loam soil over four weeks of data acquisition. The proposed system used this color information as input to an artificial neural network (ANN) system to make a decision as to whether to irrigate the soil or not. The proposed system was very accurate, achieving a mean square error (MSE) of 1.616 × 10 −6 (training), 1.004 × 10 −5 (testing) and 1.809 × 10 −5 (validation). The proposed system is simple, robust and affordable making it promising technology to support precision agriculture. Abstract : Smart irrigation; Computer vision system; RGB color analysis; Artificial neural network; Feed-forward back propagation neural network.