Claim Missing Document
Check
Articles

Found 1 Documents
Search

PERBANDINGAN PERFORMA YOLOv11n DAN YOLOv12n UNTUK DETEKSI MULTI-OBJEK BUAH DIPASAR MENGGUNAKAN ROBOFLOW Larasati Angelica Nurfadillah
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 10 (2026)
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Object detection has become one of the most important applications of computer vision in supporting automation across various sectors, including retail. However, detecting multiple fruit objects in market environments remains challenging due to complex backgrounds, varying illumination, and object occlusion. This study aims to compare the performance of the YOLOv11n and YOLOv12n models for multi-object fruit detection using the Roboflow platform. The research adopted the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology, which includes the stages of data understanding, data preparation, modeling, evaluation, and deployment. A dataset consisting of 2,974 images representing nine fruit classes, namely Apple, Banana, Grapes, Kiwi, Mango, Orange, Pineapple, Sugarapple, and Watermelon, was utilized. Data preparation involved image annotation using bounding boxes, image resizing to 640 × 640 pixels, normalization, and augmentation techniques including flipping, rotation, brightness, contrast, and zoom. Both models were trained under identical experimental settings to ensure an objective comparison. Performance evaluation was conducted using Precision, Recall, F1-Score, and mean Average Precision at IoU 0.5 (mAP50). The experimental results demonstrate differences in detection performance between the two models, enabling the identification of the most suitable lightweight model for real-time multi-object fruit detection in market environments. The findings provide valuable insights for the development of efficient computer vision applications in smart retail systems .