Sequential Model Utilization for Fruits Classification Using Optimized Deep Learning Methods on Adam Optimizer On Fine-Tuned Epochs

• Published in: 2023 Global Conference on Information Technologies and Communications (GCITC) pp. 1 - 5
• Main Authors: Gill, Kanwarpartap Singh, Anand, Vatsala, Gupta, Rupesh
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2023
• Subjects: Agriculture; Artificial Intelligence; Biological system modeling; Business; Convolutional Neural Network; Data visualization; Deep learning; Fruit Classification; Predictive models; Sequential Model; Task analysis; Training and Validation prediction
• ISBN: 9798350308143
• DOI: 10.1109/GCITC60406.2023.10426392

Fruit classification is the process of grouping various fruit species and variations according to their distinctive traits, such as form, size, colour, texture, and other distinguishing properties. In a variety of fields and businesses, such as agriculture, biology, food science, nutrition, and technology, this category has a number of uses. Fruit categorization is to develop organised methods that make different fruit varieties simple to recognise, examine, and use. A common use of computer vision and machine learning methods as suggested in this research is the categorization of fruits using deep learning. Overall, classifying fruits using deep learning is a difficult problem that needs a significant quantity of annotated data and sophisticated deep learning methods. Nonetheless, it has a sustainable wide range of useful applications, including quality control and inventory management in the food business. Moreover, fundamental machine learning methods are applied to illustrate the outcomes. The work is highly helpful in locating fruits among several categories with good accuracy in prediction in addition to visualisation. Hence, those involved in the fruit and vegetable sector will ultimately benefit greatly from this social policy. The suggested sequential model's highly accurate accuracy of 98 percent prediction for suitable visualisation is quite clear as it uses Adam optimizer and epoch of value 35 and other optimized parameters.