Development of a Recognition System for Spraying Areas from Unmanned Aerial Vehicles Using a Machine Learning Approach

Abstract

Unmanned aerial vehicle (UAV)-based spraying systems have recently become important for the precision application of pesticides, using machine learning approaches. Therefore, the objective of this research was to develop a machine learning system that has the advantages of high computational speed and good accuracy for recognizing spray and non-spray areas for UAV-based sprayers. A machine learning system was developed by using the mutual subspace method (MSM) for images collected from a UAV. Two target lands: agricultural croplands and orchard areas, were considered in building two classifiers for distinguishing spray and non-spray areas. The field experiments were conducted in target areas to train and test the system by using a commercial UAV (DJI Phantom 3 Pro) with an onboard 4K camera. The images were collected from low (5 m) and high (15 m) altitudes for croplands and orchards, respectively. The recognition system was divided into offline and online systems. In the offline recognition system, 74.4% accuracy was obtained for the classifiers in recognizing spray and non-spray areas for croplands. In the case of orchards, the average classifier recognition accuracy of spray and non-spray areas was 77%. On the other hand, the online recognition system performance had an average accuracy of 65.1% for croplands, and 75.1% for orchards. The computational time for the online recognition system was minimal, with an average of 0.0031 s for classifier recognition. The developed machine learning system had an average recognition accuracy of 70%, which can be implemented in an autonomous UAV spray system for recognizing spray and non-spray areas for real-time applications.

A Machine Learning Method for Vision-Based Unmanned Aerial Vehicle Systems to Understand Unknown Environments

“…The detection result could enhance the performance of an UAV, like alarming the operator after detecting the target, triggering the next mission, and participating in the decision making. This makes the UAV more intelligent, compared with the method proposed by Gao et al [22]. The generic object detection capability enlarges the application scope of our method, compared with the work presented by Hummel et al [22].…”

“…This makes the UAV more intelligent, compared with the method proposed by Gao et al [22]. The generic object detection capability enlarges the application scope of our method, compared with the work presented by Hummel et al [22]. Clearly, the computational performance of hardware, especially smartphones, is increasing every year.…”

“…Smartphones are more accessible than an embedded system with ROS. Therefore, our method has better generality and scalability than the method proposed by Gao et al [22].…”

“…Gao et al [22] built a system that can detect agricultural croplands and orchard areas by online and offline methods, which is not the same task definition as ours. The online method in [22] represents that learning and recognition are at the same time, and offline means the phases of learning and recognition are separated. Their algorithm is conducted on the ground station with the help of MATLAB to process the images which are collected from low (5 m) and high (15 m) altitude by the UAV.…”

Design of Plant Protection UAV Variable Spray System Based on Neural Networks

“…An artificial neural network (ANN) is a large-scale parallel nonlinear dynamic system. And neural network is a mathematical model used to find the relation between input and output datasets with complex relations [ 18 , 19 ]. With its successful application in various fields, ANN technology has entered the agricultural field and been successfully applied to various agricultural production problems.…”