Healthy/Unhealthy Tomato Fruit Grading Using Deep-Learning with Features Fusion

Food products automatic grading is a necessary process for appropriate selection, packaging, and marketing. Food grading based on artificial-intelligence (AI) technique for automating the whole process. This work considered the tomato for the study and this procedure is employed to categorize vegetables according to their grade. Proposed work implements a Deep-Learning (DL) based technique to grade the tomato. The stages of this research includes; image collection and resizing, feature extraction with a chosen model, feature reduction with 50% dropout and serial features integration to generate Fused-features Vector (FV), and classification using 3-fold cross validation and confirmation. The efficacy of this tool is evaluated using different DL-model using the chosen image data and this work achieved a detection accuracy >98% when FV based classification is executed using SoftMax. This result confirms that the proposed scheme helps to achieve a better result on the chosen tomato image data.