Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : JURNAL INTEGRASI

Pendeteksian Objek Hasil Pengepresan Kaleng dan Botol dengan Metode You Only Look Once (YOLO) yang Diaplikasikan pada Mesin Sortir Pembelajaran PBL Diono, Diono; Wicaksono, M. Jaka Wimbang; Jefiza, Adlian; Prayudha, Dimas Rama
JURNAL INTEGRASI Vol 16 No 1 (2024): Jurnal Integrasi - April 2024
Publisher : Pusat Penelitian dan Pengabdian Masyarakat Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/ji.v16i1.4598

Abstract

Image Processing is a technique of processing images with the input of an image and producing an image as well. One of the functions of Image Processing that the author wants to apply is the detection of an object from a still image or a moving image. In this application, the object to be identified by the author will be applied to the PBL sorting machine in the form of cans and bottles. In designing this system, the You Only Look Once (YOLO) method is used and several libraries. YOLO is an algorithm for detecting an object using an artificial neural network (ANN) from an image where this network divides the image into several regions and predicts each bounding box and probability for each region of the image. The author also uses a webcam to detect the object and a Servo Motor as a sorter on the PBL Sorting Machine. The result of this final project is that the system can detect objects cans and bottles properly and produce precise accuracy and is able to move the sorter based on the output data from the detection results.
Klasifikasi Wajah Manusia Menggunakan Multi Layer Perceptron Jefiza, Adlian; Diono, Diono; Putra, Irwanto Zarma; Budiana, Budiana; Nur Suciningtyas, Ika Karlina Laila; Siregar, Lindawani; Puspita, Widya Rika; Harlan, Fandy Bestario; Assegaf, Iqchan; Marpaung, Roy Hitmen
JURNAL INTEGRASI Vol 15 No 2 (2023): Jurnal Integrasi - Oktober 2023
Publisher : Pusat Penelitian dan Pengabdian Masyarakat Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/ji.v15i2.4843

Abstract

The problem of data security at a time when it is needed in the world of technology. The use of biometrics as data security is very necessary. This study aims to detect human biometrics using the Kinect sensor. The biometric that is detected is the face. The face image is captured by the Kinect sensor. For data feature extraction using Gray Level Co-Occurrence Matrix (GLCM. The parameters used are Contrast, Energy, Homogenity, and Correlation. The data obtained will be classified using Multi Layer Perceptron. Face classification is based on race. There are 3 races studied namely Indonesian, Chinese and African Native Races. The total data used are 100 photos of faces. The classification results show an accuracy of 86.7% using Multi Layer Perceptron
Prototype Sistem Elevator Menggunakan Motor Stepper Berbasis Atmega 16 Wicaksono, Muhammad Jaka Wimbang; Diono, Diono; Sani, Abdullah; Dzulfiqar, Mohamad Alif; Budiana, Budiana; Aryeni, Illa; Oktani, Dessy; Kamarudin, Kamarudin; Futra, Asrizal Deri; Darmoyono, Aditya Gautama; Mahdaliza, Rahmi; Hasnira, Hasnira; Maulidiah, Hana Mutialif; Gusnam, Mu'thiana
JURNAL INTEGRASI Vol 15 No 2 (2023): Jurnal Integrasi - Oktober 2023
Publisher : Pusat Penelitian dan Pengabdian Masyarakat Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/ji.v15i2.6058

Abstract

Elevator are one of the most important transportation these days. Elevator is a link in a tall building that has many floors. The role of the elevator which is always used by the public requires a high level of precision. In this study, researchers made a prototype elevator system using a stepper motor and obtained test results with a maximum error rate of 2% on the 10 cm elevator input using a ruler as a testing tool. The largest comparison of testing result using a ruler and using a rotary encoder is 0.2 cm or 2 mm at the 40 cm elevator input. The precision of the elevator prototype still can be improved by using half step mode when controlling the stepper motor.