Mango (Mangifera indica L.) is a high-value tropical fruit whose quality is influenced by its shape, size, texture, and skin color. Skin color is important for assessing ripeness and commercial value. During harvest, mangoes are classified based on quality for distribution to the market. Traditional classification by workers is time- consuming and can lead to spoilage if mangoes are delayed in the warehouse.
This research develops a mango fruit quality detection system by integrating color data and image processing using a Raspberry Pi. The system utilizes color
masking, converts colors from RGB to HSV, and performs classification with LED
lighting. The dataset includes mangoes with various ripeness levels and was tested
under both daytime and nighttime lighting conditions. The results show that the system
can identify mango ripeness with 90% accuracy in 20 trials and is effective during the
day, although there are challenges with low-light conditions at night. Benefits of this
system include increased efficiency in the classification process, reduced risk of
spoilage due to delays, and potential for widespread use in the agricultural industry to
enhance the quality and commercial value of mangoes.
Keywords: Mango, Image Processing, Ripeness Detection, Color Masking, Raspberry Pi, RGB to HSV, Day and Night Lighting