Fuzzy logic to classify date palm trees based on some physical properties related to precision agriculture

Abstract

Precision agriculture can be cost effective for date palm groves because the tree positions are known and fixed, the groves are mostly structured and many of agricultural operations are applied manually. Therefore, the new technology tools of precision agriculture are not essential. This study was done to improve date palm yield using maps of the variation in tree properties. Data on five Mozafati tree properties, such as sex, age, yield, visual appearance and fruit length were measured and recorded for each tree in five groves near the city of Bam in Iran. Tree positions were defined and the above properties were mapped. It was difficult to judge patterns in the variation because of tree to tree variability. Therefore, the Mamdani fuzzy inference system (MFIS) was used to classify the productive trees based on yield, fruit length and visual appearance, and to produce a tree total quality map (TTQM) for each grove. Based on the TTQM the trees were graded as excellent, good, medium, poor and very poor. The trees were also graded by the date growers who were experts. The evaluation for all groves by MFIS showed 87% general agreement with the results from the human expert indicating that it is a feasible method for classifying date palms based on quality. The results indicted that there are young and old trees in these groves that need different treatments, such as the amount of fertilizer required. Some date growers have low incomes, therefore, the overall TTQM is suitable for them because they can identify which trees have the greatest needs and apply limited resources to only those. This case study illustrates the use of computer and mathematical software to enhance date palm production and reduce the costs.

Fuzzy Knowledge-Based Model for Prediction of the Terminal Velocities of the Chickpea and Dry Bean Seeds

“…Electronic maps of tree features such as tree type, age, yield, visual appearance and fruit length were created by fuzzy expert system for managing the groves. The average evaluation results for all five groves obtained from the FES showed 87 % general conformity with the results from the human expert (Mazloumzadeh et al, 2010).…”

Exploring Precision Farming Scenarios Using Fuzzy Cognitive Maps

“…FIS applications in precision agriculture also include optimal reservoir operation for irrigation of multiple crops [35], and optimum nitrogen rates for corn crop [36]. FISs were also used by Mazloumzadeh et al [37] to classify date palm trees, and by Tagarakis et al [38] to model grape quality in vineyards. However, the majority of these studies focused on crop yield prediction, and did not offer tools to explore future scenarios, or to perform simulations by generating projections from prospective management options in PA.…”

Application of the ANFIS model in deflection prediction of concrete deep beam

“…Artificial intelligence (AI) system approaches such as artificial neural network (ANN), fuzzy inference systems (FIS), and neuro-fuzzy/ fuzzy-neural systems have been used successfully for modelling in many engineering applications as well as in agricultural (Mazloumzadeh et al 2010 andAlavi et al 2010), soil science (Yilmaz and Kaynar 2011), stability and serviceability of structures (Bilgehan 2011, Hakim et al 2011, Mohammadhassani et al 2013.…”