A Novel and Effective Ensemble Machine Learning Model for Identifying Healthy and Rotten Fruits

The world's food supply is dwindling at an alarming rate because of inefficient measures used to cope with food waste at all levels of the food system, from farms to households to restaurants. Ripe fruit may now be picked. Predicting when fruit will be ripe and of better condition is crucial not just for farmhand and the food business, but also for grocery shops and corner markets. The term “fruit maturity classification” refers to the method through which fruits are categorized based on their stage of development. Machine learning (ML) is being used in a wide variety of smart agricultural use cases, including soil and water management, seeding crops, early diagnosis, weed elimination, crop dispersal, crop collection, and productivity prediction. The purpose of this research is to identify the most effective deep learning algorithms for estimating fruit ripeness and quality to extend their storage life. Banana fruit datasets are utilized in this investigation. To increase the total no of images in the collection, image augmentation is utilized. For both datasets, we propose an Ensemble Machine Learning Technique (EMLT) that combines SVM with ResNets. Our analysis led us to the conclusion that the suggested EMLT model is the most efficient technique for determining the quality of bananas and classifying their ripeness.