Spoilage detection of tomatoes using a convolutional neural network

With the increasing productivity in agriculture, it has become extremely essential to look for an advanced technique that will help to minimise losses. Recently, deep learning has outperformed the task of recognition and classification of fruits and vegetables automatically from images, finding applicability in this study. This work, thus, attempts to develop an automatic spoilage detection CNN model for tomatoes. In this work, a deep learning-based CNN model is trained and validated on a self-prepared dataset for classifying tomatoes as edible and spoilt is proposed. The dataset consisted of 810 images, out of which 572 images were considered for training and 238 images for validation. The model is also trained iteratively with varying epoch and batch sizes to evaluate the model in giving the highest classification accuracy. The highest accuracy of 99.70% was achieved at epoch 20 and batch size 32. Further evaluating the performance of the developed model using a confusion matrix, a precision, recall and accuracy of 100%, 87% and 95%, respectively, was obtained for the spoilage detection of tomatoes. Also, on establishing Pearson's correlation between the predictive model and the sensory evaluation results, a Pearson correlation of 0.895 was obtained, showing that there is strong linear correlation between them.