Enhancing the accuracy of detecting rotten fruit using visual geometry group-16 comparison with naive bayes.

The main goal of this research is to enhance the accuracy of identifying and classifying fresh and decayed fruits using the Novel Visual Geometry Group-16 network and the Naive Bayes algorithm. Improving the detection and classification procedures will achieve this goal. Using machine learning approaches to increase fruit recognition accuracy while decreasing detection loss is the primary goal. What We Need and How It Works: The Novel Visual Geometry Group-16 and Naive Bayes algorithms are computed using ClinCalc. The parameters include a subsample size of 10 and a total sample size of 20, with Gpower maintained at 0.8, a pretest power of 0.8, and an alpha value of 0.05. The dataset is subsequently processed using these algorithms to identify fruit quality, which achieves an average accuracy of 94.60%. The dataset was comprised of 577,086 images of fruit quality detection that were collected from Kaggle. Each and every one of these images includes 224 columns and rows all at once. One way to tell if fruit is of high quality is by measuring its accuracy and loss. With a significance level of 0.014, the statistical analysis showed that VGG-16 had a more substantial impact. It was determined that this result was acceptable. The results of the independent sample T-test showed that the two algorithms were significantly different from one another. The p-value was below than the significance level of 0.05, coming in at 0.014. In conclusion, this study aims to identify the shortcomings of the existing methods and provide an enhanced approach to distinguishing fresh from spoiled fruit samples. The p-value of 0.014 (p<0.05) indicates that the Novel Visual Geometry Group-16 and the Naive Bayes technique have significantly different mean accuracy. [ABSTRACT FROM AUTHOR]