Metode sortir sayuran secara pengamatan langsung oleh manusia, sering
kali tidak konsisten terutama pada pengamatan berskala besar. Untuk
mengatasi permasalahan tersebut, peneliti mengusulkan sistem deteksi
tingkat kesegaran kubis putih secara otomatis dan real-time dengan
menggunakan algoritma deep learning YOLOv8. Sistem ini akan
mengidentifikasi berdasarkan tiga tingkatan kesegaran, yaitu kubis segar,
kubis kurang segar, dan kubis busuk. Dataset yang digunakan terdiri dari
3009 citra kubis putih hasil dari proses pre-processing berupa resize, dan
rotasi, serta proses augmentasi citra berupa brightness untuk
meningkatkan variasi sampel sehingga diperoleh proporsi 93% (3009
citra) data latih, 3% (106 citra) data validasi, dan 3% (106 citra) data uji.
Hasil pengujian sistem terhadap 106 citra data uji menghasilkan nilai rata
rata recall sebesar 0,847, recall sebesar 0,878 dan f-measure sebesar
0,863. Kemudian hasil pengujian sistem secara real-time terhadap citra
multi objek (berisi tiga kelas secara acak) yang diambil sebanyak 1 kali
per 10 detik selama 5 menit pada jarak 20 cm dihasilkan nilai recall
sebesar 0,919, recall sebesar 0,902, dan f-measure sebesar 0,903,
sedangkan pada jarak 30 cm diperoleh nilai recall sebesar 0,889, recall
sebesar 0,911, dan f-measure sebesar 0,888. Dari hasil pengujian tersebut
membuktikan bahwa sistem dapat bekerja dalam mendeteksi kesegaran
kubis putih secara real-time paling baik pada jarak 20 cm. Hal ini
membuktikan bahwa usulan sistem dapat menjadi solusi efektif untuk
seleksi mutu produk kubis putih secara otomatis dan real-time.
Kata kunci: Kesegaran Sayur Kubis Putih, Identifikasi Objek,
Yolov8, Deep learning, Real-time.

Vegetables are frequently not consistenly sorted by direct human observation, particularly when large-scale observation are involved. In order to solve this issue, researcher suggested a system that used the YOLOv8 Deep learning algorithm to determine automatically and in real-time the freshness, namely fresh cabbage, less fresh cabbage, and rotten cabbage. The dataset used consisted of 3009 images of white cabbage resulting from the pre-processing process in the form of resize, and rotation, as well as the image augmentation process in the form of brightness to increase sample variation so that a proportion of 93% (3009 images) of training data, 3% (106 images) of validation data, and 3% (106 images) of test data was obtained. The results of system testing on 106 test data images produced an average precision value of 0.847, recall of 0.878 and f- measure of 0.863. Then the result of real-time system testing on multi-object images (containing three classes randomly) taken 1 time per seconds for 5 minutes at a distance of 20 cm produced a precision value of 0.919, recall of 0.909, and an f-measure of 0.903, while at a distance of 30 cm obtained precision value of 0.889, a recall of 0.911, and an f-measure of 0.888. The test results proved that system could work in detecting the freshness of white cabbage in real-time best at a distance of 20 cm. This finding proves that the proposed system can be an effective solution for automatic and real-time selection of white cabbage product quality. Keywords: White Cabbage Freshness, Object Identification, YOLOv8, Deep Learning, Real-time.