Automatic Detection of Cabbage Pest Attacks Based on Leaf Images with Machine Learning Approach

Ni Wayan Surya Wardhani; Prayudi Lestantyo; Atiek Iriany; Nur Silviyah Rahmi

doi:10.70687/3szcd282

Authors

Ni Wayan Surya Wardhani Department of Statistics, Brawijaya University, Malang, Indonesia Author
Prayudi Lestantyo Management of Islamic Education, Maulana Malik Ibrahim State University Malang, Malang, Indonesia Author
Atiek Iriany Department of Statistics, Brawijaya University, Malang, Indonesia Author
Nur Silviyah Rahmi Department of Statistics, Brawijaya University, Malang, Indonesia Author

DOI:

https://doi.org/10.70687/3szcd282

Keywords:

Cabbage pest, CART, Classification

Abstract

Farmers in cabbage farming face many problems, one of which is pest attack. Plutella xylostella L. is a major pest on cabbage (known as cabbage leaf caterpillar) which can cause a decrease in production of up to 100 percent. Decision Support System (DSS) was developed to classify the attack rate of Plutella to reduce the negative effects of using various types of high doses of pesticides and short spraying intervals but causing residual effects and killing natural enemies. DSS has a role in helping farmers to make decisions regarding the time of pesticide treatment needed to minimize negative effects and increase productivity. In this study, DSS was developed to detect damage to cabbage (Brassica oleracea L) crops so that farmers can determine pesticide doses and spraying intervals based on a website. The results of the system is presented in the form of images and the percentage of damage to cabbage plants. Therefore, the CART method can be used to analyze the level of damage to plants that are attacked by pests.