+44-7897-074717George Ifeoluwa Pele* and Simeon Adedoyin Adeyemi
The quality change in climacteric fruits is one of the factors militating against post-harvest handling and the preference of consumers. The present study was carried out to develop a predictive model for post-harvest changes of some selected climacteric. Selected samples of matured climacteric fruits (Apple, Banana and Guava) were captured in JPEG format and concurrently evaluated for physicochemical properties, proximate and vitamin compositions from maturity to senescence. A predictive model to map the images of the selected fruits with the identified quality attributes was developed by employing Convolutional Neural Network (CNN) algorithm. The predictive model was implemented as mobile application using python programming language while the performance of the implemented model was evaluated using Accuracy, Precision and Recall as Metrics. All the cropped images were resized and the images splitted into 70%, 20%, and 10% for training, validation and testing set, respectively. The training images were augmented while the model training was carried out using Google Colab using “Adam” optimizer. Two Inference applications were developed, one for the web channel and the other for mobile channel. The web channel was implemented using Flask, HTML, CSS and JavaScript. The mobile application part was implemented using Flutter framework and Dart programming language. The results showed that 61 apple fruits and 65 Guava fruits were correctly classified. Out of the 79 Banana fruit images evaluated, there were 77 correct classifications with 2 misclassified as Guava. The trained model was able to achieve 100%, 97.5%, 100% predictive accuracies for Apple, Banana, Guava, respectively.
Share this article