Institute of Art Design + Technology Dún Laoghaire
Ireland’s campus for the Creative Industries
On Show
 Courses
Search
Close
3D Design Modelmaking + Digital Art Animation Applied Psychology Art Character MakeUp Design Costume Design Creative Computing Creative Media Technologies Creative Music Production Design for Change Film + Television Production New Media Studies Photography Production Design User Experience Design Visual Communication Design




Shane Smyth 

BSc [Hons] Creative Media Technologies

My name is Shane Smyth. I study Creative Media Technologies at the Institute of Art, Design and Technology in Dun Laoghaire. My area of interests includes computer programming, artificial intelligence, and microcontrollers. In the last four years I have developed applications that implement Arduino Uno, Python and computer vision. I strive to create unique projects that are fun, functional and can be applied to the real world. I enjoy self-learning and keeping up to date with new technologies. I describe myself as enthusiastic, focused and optimistic.

In the first three prototypes, the aim of Breadboard2Schematic was to create a Machine Learning model that can detect electrical components in an image, and then draw the relating schematic symbol for that component. This was completed using Jupyter Notebook and Tensorflow Object Detection. In this prototype, hundreds of images were trained to optimize the performance of the model.
This is the main page of Breadboard2Schematic. This is the first page the user sees after entering the website. Once the user enters the website, a loading bar is animated at the top of the screen that indicates the machine learning model is loading. After the model is loaded, the loading bar disappears. On this screen, the user can see a navigation bar with 'Home', 'News', 'Contact' and 'About' buttons. A 'Choose File' button can be pressed for the user to select an electrical circuit image from their file manager. The aim of the website design is to convey a simplistic, approachable style because machine learning websites are a new concept that can be difficult to understand and follow.
These are the results. After selecting an image and pressing the 'predict' button on the right-hand side of the screen, detections were drawn on the image. The outcome of this image detected the first rail on the breadboard. This is used for future research to decipher if a component is connected to this rail, to be able to understand the position of each component for a circuit diagram.
This is a video showing how Breadboard2Schematic works.
Project Description

Breadboard2Schematic is a website that converts an image of a circuit into a circuit diagram using automation. This conversion is completed by using artificial intelligence. The purpose of this project is to make schematic production faster, effective, and accurate with little to no error in the process. Circuit design can be a lengthy and confusing process. These difficulties created the inspiration and idea for the project. The use of this project is for educational purposes, as well as manufacturing. In this project, a website was developed using hyper-text markup language (HTML), cascading style sheets (CSS) and JavaScript (JS). The machine learning application of the project was developed using Python with Tensorflow. In the future of this project, the aims include converting schematic diagrams, including hand-drawn, into a simulation for the user to visualize.


Download Thesis PDF

Exploring Machine Learning as a Tool to Revolutionize Electrical Circuitry

This thesis explores the use of machine learning to enhance electrical schematic diagram production. The thesis follows a procedure of planning the project, constructing three prototypes, implementing the final project and user testing. The planning section describes the methodology used in the project, the project timeline, research, and purpose of Breadboard2Schematic. The construction of the three prototypes summarizes the procedures used to develop each prototype during the year. This involved creating a machine learning application. The final project implementation chapter shows the procedure of the final project, which involved applying the machine learning application to an end-to-end built website. The user testing section involves gathering users to test the final project for feedback and further enhancements of the project in the future.


Download Thesis PDF
Creative Media Technologies
Top