CD Tesis

Otomasi Sistem Pencitraan 3d Untuk Menentukan Massa Dan Volume Buah Dengan Bentuk Simetris

---

Conventional sorting of agricultural products manually uses human labor
and is subjective, automatic sorting in order to have uniform quality standards is
required. This method has several advantages, including that it is non-destructive,
faster, more accurate for repetitive work, and quantitative. Imaging methods can
be used to detect, evaluate or inspect the external and internal state of a fruit or
vegetable. This study aims to create a system for estimating volume and mass of
agricultural products using 3D imaging with the help of a line diode laser. This
research is expected to produce a simple and economical fruit sorting tool design
which is expected to be used in small-medium scale agricultural units that require
sorting tools but have limited funding.
In this study, the method used is direct measurement and then processed
by the Simpson’s rule method and the 3D laser scanner method based on Python
programming known as the computer vision method and volume measurement
using the water displacement method as a comparison method. The samples of
this study were symmetrical fruit such as lemon, guava, papaya, orange, spherical
watermelon and oval watermelon. Mass and volume were measured by direct
measurement and then processed using the Simpson's rule method. The volume of
these fruits was also measured using the water displacement method and the
computer vision method. The mass of the sample was obtained from the density
analysis of the fruits with the volume measured by various methods.
The JB10 sample has a relative error of 20.55% for CV-manual
measurements and 24.70% for CV-WDM measurements. The SL1 sample has a
relative error of 70.44% for WDM-manual measurements and 41.09% for CVWDM
measurements. The Bland - Altman graph shows that 100% of volume data
of round fruit samples and 87.5% of volume data of oval fruit samples are within
the area 1.96 times the difference in standard deviation for volume measurement
using the manual method - CV. SB1 sample has a relative error of 65.14% for
mass analysis with the volume measured by manual method - CV and 50.21% for
mass analysis with manual volume measurement - WDM. Sample SL1 has a
relative error of 41.09% for mass analysis with the measurement volume of WDM
–CV. CV has a higher effectiveness for measuring volume and mass analysis of
small round and oval fruit.
Keywords: volume, mass, symmetrical fruits, WDM, CV

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

10 17. 220 (0020)

Penerbit

Pekanbaru : Universitas Riau – Pascasarjana – Tesis Fisika., 2020

Deskripsi Fisik

xv, 77 hlm.; ill.: 29 cm

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

10 17. 220 (0020)

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

FISIKA

Info Detail Spesifik

-

Pernyataan Tanggungjawab

FATAH

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar