An Introduction to A Smart Durian Musang King and Durian Kampung Classification

Fruit classification is a crucial implementation process for identifying the types of fruit in the agricultural field. Traditionally the identifying of the fruit was conducted manually using human power. Classification with the manual method requires a systematic, continuous, and even human eye recognition technology to maintain quality consistency. However, this process and step are considered repetitious and uninteresting for the worker, which infringes a prone to human mistakes. In Malaysia's local exercises, The Malaysian Federal Agricultural Marketing Authority (FAMA) controls a set of six indexes for categorizing various types and categories of tropical fruit available locally. Although a reference for classifying the tropical fruit is available here, a method and reliable classifier that could be implemented universally are still desired for the agricultural player. This paper introduces a simple idea employing CNN to classify the local Durian in Malaysia, Durian Kampung, and the well-known durian, Musang King. Although the work is still in progress, we produced a comparable and reliable result with 80% classification accuracy utilizing the Deep Learning approach.