Banana is one of the most consumed fruits globally and is a rich source of vitamins, minerals and carbohydrates. With the many benefits that bananas have, many farmers cultivate this fruit. The problem that occurs when the harvest is produced on a large scale is the process of selecting bananas that are still unripe or ripe. Usually farmers carry out the selection process manually by visually identifying ripeness based on the color of the fruit skin. However, direct observation has several drawbacks such as subjectivity, takes a long time and is inaccurate. For this reason, we need a system that can help determine the maturity level of bananas automatically through a series of banana image processing processes. One way that can be used to determine the maturity level of bananas is by looking at the color and texture of the bananas. This study aims to classify the maturity level of bananas based on the color and texture characteristics of the banana image using the Gray Level Co-occurrence Matrix and K-Nearest Neighbor methods for the classification process. Based on the results of the research analysis that has been carried out, using the parameter k which has a value of 3 obtains very high accuracy.