Marwin B. Alejo
mbalejo.cpe@gmail.com; marwin.alejo@eee.upd.edu.ph; mbalejo@national-u.edu.ph 

University of the Philippines, Diliman, Quezon City
National Graduate School of Engineering (NGSE)
Electrical and Electronics Engineering Institute (EEEI)
Digital Signal Processing Lab
September 2020 - in progress (completed all units except dissertation)
General Weighted Average as of January 2023 (24 units out of 36): 1.2188*

Relevant Courses:
[EE274] Digital Signal Processing I
[EE374] Digital Signal Processing II
[EE290] Advances in Digital Signal Processing
[EE298] Deep Learning
[CS284] Machine Learning
[CS282] Computer Vision
[CS295] Reinforcement Learning
[EE214] Statistics for EE (graduate-level)

*UP Grading System (pages 26 and 37 of Academic Information)

[course code] PhDEEE courses description

Niches: Applied DSP, Machine Learning, Computer Vision.

Technological Institute of the Philippines, Quezon City
November 2015 - March 2018
General Weighted Average: 1.43**

**Similar to UP Grading System

Thesis (Equivalent to Capstone): Ear-based Biometric Recognition System through Computer Vision and Learning Approach 

Limited ear dataset yields to the adaption of domain adaptive deep learning or transfer learning in the development of ear biometric recognition. Ear recognition is a variation of biometrics that is becoming popular in various areas of research due to the advantages of ears towards human identity recognition. In this paper, handpicked CNN architectures: AlexNet, GoogLeNet, Inception-v3, Inception-ResNet-v2, ResNet-18, ResNet-50, SqueezeNet, ShuffleNet, and MobileNet-v2 are explored and compared for use in an unconstrained ear biometric recognition. 250 unconstrained ear images are collected and acquired from the web through web crawlers and are preprocessed with basic image processing methods including the use of contrast-limited adaptive histogram equalization for ear image quality improvement. Each CNN architecture is analyzed structurally and are fine-tuned to satisfy the requirements of ear recognition. Earlier layers of CNN architectures are used as feature extractors. Last 2-3 layers of each CNN architectures are fine-tuned thus, are replaced with layers of the same kind for ear recognition models to classify 10 classes of ears instead of 1000. 80 percent of acquired unconstrained ear images is used for training and the remaining 20 percent is reserved for testing and validation. Results of each architectures are compared in terms of their training time, training and validation outputs as such learned features and losses, and test results in terms of above-95% accuracy confidence. Above all the used architectures, ResNet, AlexNet, and GoogleNet achieved an accuracy confidence of 97-100% and is best for use in unconstrained ear biometric recognition while ShuffleNet, despite of achieving approximately 90%, shows promising result for use in mobile version of unconstrained ear biometric recognition.

Niches: Applied DSP, Machine Learning, Computer Vision.

Technological Institute of the Philippines, Quezon City
June 2010 - April 2015
General Weighted Average: 2.13**

**Similar to UP Grading System

Capstone: Chicken Egg Grade Classification System

A machine vision system for grading table chicken eggs according to their standard sizes through their geometric parameters instead of the traditional weighing means. In this project, we only used image processing techniques (given that ML are not that prevalent for actual applications in 2014-2015) to determine the geometric parameters (in pixels) of the eggs and apply traditional fuzzy logic in determining the classification of the eggs based on their geometric features. The algorithm of this project was implemented on a Raspberry Pi (RPi) B+ (version 1) while the whole hardware of the system was controlled by a DIY Arduino that is in-sync with the RPi board. Overall, the grading results of the developed machine vision system achieved a comparable result with the traditional weight-based grading of chicken eggs. 

Niches: Robotics and Embedded Systems, Machine Vision, Image Processing.