A HIGH ACCURACY PEST DETECTION METHOD USING NAIVE MACHINE LERANING METHOD
Abstract
Agriculture is one of the biggest economic activities in a developing country such as Malaysia. However, pest attacks are inevitable. This problem incurs loss due to profligate pesticide spray after farmers fail to detect pests accurately. For a developing country, a simple and low-cost pest detection system is indispensable. Here, we introduce naïve machine learning into the detection method and obtained high-accuracy pest detection results. We studied and explored the effect of k-means clustering value and segmentation number parameters on detection accuracy. Our method achieved 95% accuracy in pest detection, a competitive accuracy compared to other complex machine learning methods such as convolutional neural networks (CNN) and k-nearest neighbors’ algorithm (kNN).