COFFEE MATURITY CLASSIFICATION USING CONVOLUTIONAL NEURAL NETWORK AND TRANSFER LEARNING
Coffee is one of the most popular beverages in the world, and the quality of coffee products largely depends on the maturity level of the coffee cherries used. Accurate and efficient classification of coffee maturity is essential for the coffee industry to ensure the quality and consistency of coffee products. Traditional methods of coffee maturity classification, such as visual inspection, can be time-consuming, subjective, and prone to errors. Therefore, the development of automated methods for coffee maturity classification is highly desirable. In recent years, deep learning has shown great promise in various image recognition tasks, including object detection, segmentation, and classification. Convolutional neural networks (CNNs) have been widely used for image classification tasks due to their ability to automatically learn features from images. In this study, we propose a deep learning approach for coffee maturity classification using images of coffee cherries. The proposed method consists of two main stages: image preprocessing and CNN-based classification. In the preprocessing stage, images of coffee cherries are first resized and normalized to reduce the variability in image size and lighting conditions. In the classification stage, a CNN is trained on the preprocessed images to classify coffee cherries into different maturity levels. To evaluate the performance of the proposed method, we collected a dataset of coffee cherry images at different maturity stages. The dataset was divided into training, validation, and testing sets, and the CNN was trained on the training set and evaluated on the testing set. The results show that the proposed method achieved high accuracy and outperformed other traditional machine learning techniques for coffee maturity classification.