Projects
A Multimodal Deep Learning Approach for Non-Destructive Fruit Quality Assessment
A multimodal deep learning regression model for non-destructive quality assessment of fig fruit is successfully developed to address the problem with regard to time consumption, human mistakes, and fruit losses. The inaccuracy and low performance of the regression model that relies only on a single image modality can be eliminated and replaceable with our proposed OM_CNN_fig regression model.
Forensic Sketch Suspect Identification System (SketchSIS)
This is an automatic photo retrieval system based on a face sketch that has very useful application as to narrow down potential suspects in criminal investigations (e.g., ours is called SketchSIS).
To identify a suspect, the system normally attempts to retrieve the most similar mugshots from the database merely based on a forensic sketch.
TECH-cREATE PriDe competition registration & jury system
The system is developed for the Jabatan Pengurusan Perniagaan ENT course on the student ENT competition projects. The system comprises two sections namely for registration & project submission and jury evaluation. The system is developed using HTML & javascript as the UI webapp, while utilising google apps script in google Sheets as the database for the system. In addition, the CSS on webapp page is used from w3school.com
SMART FERTIGATION SYSTEM WITH IR 4.0 APPLICATION (GREENHOUSE)
This is an IoT-Based fertigation system. The project focuses specifically on Ficus Carica (fig). The project aims to develop a classification model based on ANN that can predict or classify the ripening stage based on either pectin activity or fruit appearance. In addition to that, a non-destructive quality assessment of the fig using imaging technique is also investigated.
URBAN FARMING USING CABIN SYSTEM WITH IR 4.0 APPLICATION
The first phase of the project is the development of cabins for urban purposes farming. The chosen cultivation method is hydroponics using nutrient film technique (NFT) method. Vegetables to be selected as pilot models are salads and tomatoes.
The second phase is the development of a parameter monitoring system in agriculture using the Internet of things (IoT) method to increase the productivity of agricultural products at a cost-effective.
Harum manis IDENTIFICATION APPS
An android Apps is designed in such a way that it can recognize a mango called Harum Manis.