Coffee is one of the agricultural products that has a fairly high value, making it one of the export commodities. Ironically, however, the amount of exports from 2021 to 2024 has continued to decline, partly due to monkey infestations. Efforts to prevent monkey attacks have been made by designing monkey detection based on PIR-HC-SR04 sensors, ultrasonic sensors, and combined with a microcontroller as a controller, but the results have not been able to distinguish between monkeys, cows, and humans passing through the plantation area. For this reason, the use of cameras will be tried as a means of sensing the plantation area and replacing the use of old sensors. Therefore, the purpose of this study is to optimize the detection of monkey pests based on digital images. To achieve this research objective, the convolution neural network (CNN) method is used. In order to obtain maximum results, the CNN will be trained beforehand so that it can correctly distinguish between monkeys, cows, humans, and so on. The results obtained are not yet optimal due to the occurrence of monkey pest attacks.
Copyrights © 2025