Using Yolo V7 to Detect the Health and Quality of Fruits

EasyChair Preprint 10546

Abstract

This study explores the potential of YOLOv7 in detecting the health and quality of fruits. The aim is to develop a dependable automated system that can detect the quality and safety of fruits to ensure that only the best and safest fruits are delivered to consumers. The methodology involves linking YOLOv7 with image enhancement techniques for efficient fruit detection and classification. The performance of YOLOv7 is evaluated in detecting the health and quality of fruits using a specifically designed dataset. Results show that YOLOv7 has an accuracy of 83.5% in detecting fresh and rotten apples, indicating its potential as a useful tool in the fruit management industry. The high accuracy rate in detecting fruit quality can improve the efficiency of fruit sorting and grading, leading to higher productivity and better quality final products. Future research can focus on optimizing the algorithm for specific use cases and validating its performance in other scenarios. Overall, this study demonstrates the potential of YOLOv7 in fruit detection and classification.

Keyphrases: Agriculture Technology, Convolutional Neural Networks, YOLO, computer vision, fruit detection, image processing, object detection

@booklet{EasyChair:10546,
  author    = {Theodore Aditya Oei and Marcelius Surya Wijaya and Johannes Batistuta Simanjuntak and Erna Francisca Angela Sihotang and Edy Irwansyah},
  title     = {Using Yolo V7 to Detect the Health and Quality of Fruits},
  howpublished = {EasyChair Preprint 10546},
  year      = {EasyChair, 2023}}