The water requirement of chili plants is an important thing, considering that red chili is a plant that is very sensitive to water. If there is a lack of water in these plants, the development of branches will be hampered, the leaves will turn yellow and shrink if this occurs in the vegetative phase, and if this occurs in the generative phase, then the chili plants will experience flowers, leaves, and fruit falling and being damaged. In addition, this plant is also very vulnerable if excess water causes root rot and is susceptible to various diseases that cause crop failure. So that the irrigation system for water distribution for these plants must be precise and in accordance with what is needed by chili plants. Therefore, a computational model is needed that is able to detect water needs in plants with precision. One of the proposed algorithms for detecting water needs is convolutional neural networks (CNN). This algorithm will detect the water needs of plants based on the condition of the leaves. The expected goal of this research is to serve as a basis for building a prototype of an intelligent irrigation system capable of meeting the precise water needs of red chili plants.