Publications

Development of a system for the validation of industrial parts pose on conveyor belts with a focus on simulation and advanced experimentation (Master's Thesis)

15 feb. 2024
Link to PDF
- Spanish
Description:

This project focuses on developing a system that uses reference and current images from a 2D camera to monitor and analyze parts on a conveyor belt in an industrial environment. Graphical user interfaces, communication interfaces, and software are implemented for integrating sensor data, running preprocessing algorithms, motion detection, feature extraction and comparison, as well as object recognition using classical computer vision techniques. A practical solution is proposed, developed under the Qt framework, supported by a robust simulator that facilitates the experimentation and refinement of algorithms in simulated 3D industrial environments. The simulator is presented as a versatile tool that can support the research of various computer vision and machine learning techniques; developed using the OpenGL standard and the C++ programming language with an object-oriented approach, it serves as a powerful and versatile tool for 3D emulation of industrial spaces. The implemented algorithm, using the OpenCV library, acts as an analytical classifier to determine the validity of a part's pose on a production line. Two data sources are used: a static collection of images (reference parts) and pre-calculated feature vectors, and on the other hand, frames from a 2D camera (parts to be validated). The system employs quasi-intelligent algorithmic entities to process images and other signals, perform numerical analysis, feature recognition, and other tasks, generating a final response based on quantitative estimates, votes, weightings, and empirical thresholds. The combination of multiple algorithms proves effective in handling variations and complex deformations. An experimental phase is highlighted, both in simulation and reality, where various environments and conditions are set up to measure the system’s ability to classify valid and invalid parts, as well as its accuracy in adverse scenarios. Finally, after testing the system using statistical repeatability analysis to evaluate its robustness under various cases, the results are analyzed, and it is concluded that the system as a whole, along with the implemented algorithms, proves effective, culminating in a comprehensive solution that successfully addresses the challenge of monitoring and analyzing the pose of parts on a production line.

Integration of an educational omnidirectional autonomous mobile robot in ROS (Bachelor's Thesis)

19 jul. 2022
Link to PDF
- Spanish
Description:

This document is part of a larger work within the context of an analytical and descriptive approach to mobile robotics and the open-source software framework ROS. It takes advantage of the availability of an educational, omnidirectional commercial mobile robot to integrate it, from scratch, into this metasystem, using the URDF markup language format, creating and developing various plugins, packages, and nodes to carry out the simulation and real implementation of the robot’s hardware components. Among other things, it includes the study, under this specific framework, of embedded devices such as SBCs (Raspberry Pi) or Arduino controllers. Initially, this work introduces several highly relevant concepts in a notably qualitative manner, belonging to specific fields of study that are fundamental to understanding the overall scope of the project. Given that much of the work is centered around the ROS environment, it also intensively explores various other aspects related to the mobile robot to expand the range of possibilities that can be abstracted based on this proposed work. Later, the process followed to successfully integrate the robot model into the ROS platform (using Gazebo as the simulation environment) is analyzed, along with successfully controlling the movement of its motors or actuators, intercommunicating the microcomputer and microcontroller boards of the robotic system, and incorporating the necessary sensor and node structure to achieve a mobile robot that autonomously performs basic navigation and mapping tasks (both in simulation and in reality), among others. Finally, several short- and long-term proposals are briefly presented and discussed, which could serve as extensions, expansions, or future opportunities for this project.

Preprints

Eventually, I research a few exciting topics in image processing and computer vision.

Physics Meets Pixels: PDE Models in Image Processing

11 dic. 2024
Link to preprint
- arxiv (Cornell Tech)
Abstract:

Partial Differential Equations (PDEs) have long been recognized as powerful tools for image processing and analysis, providing a framework to model and exploit structural and geometric properties inherent in visual data. Over the years, numerous PDE-based models have been developed and refined, inspired by natural analogies between physical phenomena and image spaces. These methods have proven highly effective in a wide range of applications, including denoising, deblurring, sharpening, inpainting, feature extraction, and others. This work provides a theoretical and computational exploration of both fundamental and innovative PDE models applied to image processing, accompanied by extensive numerical experimentation and objective and subjective analysis. Building upon well-established techniques, we introduce novel physical-based PDE models specifically designed for various image processing tasks. These models incorporate mathematical principles and approaches that, to the best of our knowledge, have not been previously applied in this domain, showcasing their potential to address challenges beyond the capabilities of traditional and existing PDE methods. By formulating and solving these mathematical models, we demonstrate their effectiveness in advancing image processing tasks while retaining a rigorous connection to their theoretical underpinnings. This work seeks to bridge foundational concepts and cutting-edge innovations, contributing to the evolution of PDE methodologies in digital image processing and related interdisciplinary fields. Link to code.

A Hybrid Frameworkfor Statistical Feature Selection and Image-Based Noise-Defect Detection

11 dic. 2024
Link to preprint
- arxiv (Cornell Tech)
Abstract:

In industrial imaging, accurately detecting and distinguishing surface defects from noise is critical and challenging, particularly in complex environments with noisy data. This paper presents a hybrid framework that integrates both statistical feature selection and classification techniques to improve defect detection accuracy while minimizing false positives. The motivation of the system is based on the generation of scalar scores that represent the likelihood that a region of interest (ROI) is classified as a defect or noise. We present around 55 distinguished features that are extracted from industrial images, which are then analyzed using statistical methods such as Fisher separation, chi-squared test, and variance analysis. These techniques identify the most discriminative features, focusing on maximizing the separation between true defects and noise. Fisher's criterion ensures robust, real-time performance for automated systems. This statistical framework opens up multiple avenues for application, functioning as a standalone assessment module or as an a posteriori enhancement to machine learning classifiers. The framework can be implemented as a black-box module that applies to existing classifiers, providing an adaptable layer of quality control and optimizing predictions by leveraging intuitive feature extraction strategies, emphasizing the rationale behind feature significance and the statistical rigor of feature selection. By integrating these methods with flexible machine learning applications, the proposed framework improves detection accuracy and reduces false positives and misclassifications, especially in complex, noisy environments.

Other

A few papers and some more informal, short-format essays.

Introductory, Descriptive, and Practical Essay on Neural Networks

05 jun. 2023
Link to PDF
- Spanish
Description:

This document encompasses several words addressing matters related to the art of creating the quality of discernment, namely, Artificial Intelligence. More specifically, it focuses on one of the most significant foreign contributions to this field: artificial neural networks (ANNs), which are currently highly popular due to their most important functional characteristic—the ability to learn automatically. This capability does not aim to formally create knowledge but rather emulate it through training or learning. Juxtaposed with the results attached to this document, which provide various experiments using MATLAB tools—particularly studying convolutional (artificial) neural networks—it is intended for the reader to accurately grasp and absorb numerous concepts related to these topics and many others that constitute, as a broad ensemble, artificial intelligent systems. All of this is undertaken while keeping in mind the question that, at the time, brilliantly ignited the beginning of this field of study: Can machines think beyond what we can instruct them? A question “whose ramifications continue to be explored to this day.” Based on final master's course assignment report.

Processor vs Controller

01 sep. 2021
Link to PDF
- Spanish
Description:

This document aims to concisely and conclusively outline all the differences between two fundamental elements of electronics and computing: microprocessors and microcontrollers. The title of this report omits the prefixes "micro-" (from the Greek mikrós [small], typically related to devices on a micro or smaller scale) from both terms to emphasize that it is the functions of these components that often confuse those new to these fields of study when trying to differentiate between them. After reviewing some important historical aspects and briefly summarizing the definitions of both, their differences are cited and clarified, concluding with a general framework that introduces basic and common applications of these devices, as well as promoting an understanding of their internal structure and utilization.

The embedded questions

01 aug. 2021
Link to PDF
- Spanish
Description:

This document is a curious and brief formalization of a series of questions originally intended to be “launched” expressly in job interviews for potential embedded systems developers and programmers (“computerized systems deliberately built for a specific application”). These questions and the concepts they suggest belong to the collective imagination of everyone dedicated to this field of study and research. However, the specific merits (in the case discussed in this report) go to certain contributors and developers in the domain of the website www.eevblog.com (Electronics Engineering Video Blog), founded by Australian electronic engineer David L. Jones. With a total of 12 interesting and varied "questions," the reader will delve into a wide range of concepts, mostly related to fields of knowledge such as analog and digital electronics (devices, microcontrollers, microprocessors, etc.), high-level programming (C and C++ languages, etc.), communications (standards for transferring information between different ICs), or control engineering. Both during the process of understanding each question and while assimilating its possible answers and observations, one can expand and better comprehend the “visible spectrum” of this fascinating topic that encompasses embedded systems in general.

The metal-oxide-semiconductor field-effect transistor (MOSFET)

12 jul. 2021
Link to PDF
- Spanish
Description:

This document aims to gather a significant amount of relevant information, of a divulgative, educational, and investigative nature, about a device that has been extremely important for modern electronics for more than fifty years. This device, which holds a central place as the title of the report, is the MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor). Various aspects will be addressed, such as its history and predecessors (the scientific and technological path that has been followed over the years to reach the actual component), its structure and operation (from a rather general, qualitative, and practical perspective, but still attempting to briefly delve into its characteristics and properties that distinguish its physical behavior from that of other devices), its position within a global classification of different types of transistors, common applications, and other general and specific aspects that will prove to be of great value in helping the reader gain a global, clear, and concise understanding of this marvelous electronic device.