A deep learning model for crop classification

Hairudin, Noradila Hanis (2022) A deep learning model for crop classification. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Preview

Text (Full Text) A deep learning model for crop classification.pdf - Submitted Version Download (2MB) | Preview

Abstract

Given the importance of food status in modern civilization and its economic contribution, it becomes more critical to improve fruit freshness, but manual operation is time demanding. On the one hand, each crop falls into a number of difficult-to-categorize categories. The purpose of this project is to determine how successfully an artificial intelligence can classify them. The eggplant was chosen as the crop for this research. The algorithm is designed to be capable of evaluating fruit quality based on metrics such as size, colour, shape, and intensity, however colour and size remain the most critical factors in fruit grading and sorting. Thus, this project will develop a model that capable of automatically determining the freshness of eggplants, which will save time, labour, and provide a higher level of accuracy than manual sorting. The primary objective is to classify the eggplants produced in terms of their quality. When three grades of eggplants are input into the deep learning system. It is self-adjusting in order to classify (grade) eggplants depending on sample sets. When an eggplant has characteristics of one of the three categories, it is categorised and graded appropriately. This technique is quick and dependable; more importantly, it produces consistent results. A total of 148 eggplant datasets were collected in order to obtain the results.

Actions (login required)

View Item