A deep learning approach for Moroccan dates types recognition.

Abstract

The growth of the date fruit market, it becomes necessary to use artificial intelligence techniques to recognize date fruits categories. In this work, we use computer vision approaches to classify date fruits produced in Morocco according to their type. To do this, A dataset has been curated, comprising images of the seven most prevalent types of date fruit found in Morocco. Distinguishing itself from other prominent datasets in the field, our dataset poses a challenge to the model due to its inclusion of images captured under varying conditions. For the recognition algorithm, two computer vision approaches are compared and evaluated in terms of performance. Both approaches are based on transfer learning of a convolutional neural network CNN. These two approaches are standard feature extraction where deep features are used to train a machine learning classifier, we compare four classifiers and show that SVM gives the best results. The second approach is fine-tuning where we fit the pre-trained model to our dataset. The approaches used in this work achieve outstanding performance on our dataset, with a classification precision of 97 %. We employ the GradCam technique to visualize the features of our model, revealing that the model primarily emphasizes the texture of the date fruit in its predictions.