Optimized deep learning model for mango grading: Hybridizing lion plus firefly algorithm

This paper intends to present an automated mango grading system under four stages (1) pre-processing, (2) feature extraction, (3) optimal selection and (4) classification. Initially, the input image is subjected pre-processing phase, where reading, sizing, noise removal segmentation process happens. Subsequently, features are extracted from pre-processed image. To make more effective, features, selected using a new hybrid optimization algorithm termed lion assisted firefly (LA-FF), which combination of LA FF, respectively. Then, given for classification process, optimized deep convolutional neural network (CNN) deployed. As major contribution, configuration CNN fine-tuned via selecting count layers. obviously enhances accuracy in system. For fine-tuning layers CNN, LA-FF used so that classifier optimized. The evaluated on basis healthydiseased, ripeunripe bigmediumvery big cases with respect type I II measures performance proposed model compared over other state-of-the-art models.