Quality Estimation of Indian Chili Using Affordable IoT-Enhanced Image Processing

Internet of Things (IoT) based growth parameter estimation of Indian vegetables or spices is very beneficial for commercial purposes in case of harvesting. This study presents an IoT-based system for real-time monitoring of plant growth parameters of chili after identifying and measuring the changes in color, length, shape, and other parameters of chili, the quality of the product can be judged. Using Machine Learning (ML), the data is analyzed and the optimum harvesting time for using the product is identified. This automated estimating system shows the measured values of color, shape, and other parameters, which can easily indicate the proper ripening stage of chili. The proposed IoT-based method uses the ESP32-CAM module to capture real-time images in field conditions, Google Colab for data storage, and machine learning algorithms for predicting accurate harvesting time by maintaining the quality. The results obtained from the remote system after employing ML are promising. The system's performance has been evaluated using stratified 10-fold cross-validation with an 82.14% accuracy measure using the Metamodel (Gradient Boost) and Random Forest, SVM, Logistic Regression, and Naive Bayes which are used as the base models. Whereas the highest accuracy is obtained as 76.38% for both SVM and KNN when used individually. Based on the implementation of machine learning algorithms, it can be confidently asserted that the system can be relied upon to provide accurate and trustworthy results for quality estimation through image processing of Indian Chilli using the proposed IoT-based device.