The Real-Time Fresh Fruit Bunch (FFB) Ripeness Detection System for Oil Palm is an application used to assist in detecting the ripeness of fresh oil palm fruit bunches during the grading process in palm oil factories in real-time, using a smartphone with video input. This process is crucial for maintaining the quality of the palm oil produced, as unripe or under-ripe FFBs can result in a high level of Free Fatty Acids (FFA) in the palm oil. Therefore, this system is used to detect unripe or under-ripe FFBs, ensuring that each FFB to be processed is ripe, thus guaranteeing the quality of the resulting palm oil.

The palm fruit ripeness detection process in this application utilizes video input, which is used for sorting FFB ripeness levels into six categories: raw, under-ripe, ripe, over-ripe, empty bunches, and abnormal bunches. This application is modeled using quantized deep-learning techniques for implementation on Android-based smartphones.