Road Sign Recognition: Using Machine Learning to Spot Road Signs
Abstract:
Bibliography/Citations:
“Teachable Machine Startup Guide.” Google, Google, https://teachablemachine.withgoogle.com/faq#Getting-started.
“What Is Traffic-Sign Recognition?” J.D. Power, https://www.jdpower.com/cars/shopping-guides/what-is-traffic-sign-recognition.
“Facial Recognition in Smartphones.” Wikipedia, Wikimedia Foundation, 15 Feb. 2023, https://en.wikipedia.org/wiki/Face_ID#:~:text=Face%20ID%20is%20based%20on,module%20that%20reads%20the%20pattern.
Barla, Nilesh. “Self-Driving Cars With Convolutional Neural Networks (CNN).” Neptune.ai, 25 Jan. 2023, https://neptune.ai/blog/self-driving-cars-with-convolutional-neural-networks-cnn.
DriversEd.com. “Helpful Driving Information: Signs.” Driving Information | DriversEd.com, https://driversed.com/driving-information/signs-signals-and-markings/signs/.
Additional Project Information
Research Plan:
I want to test whether machine learning tools and programs could be used to identify road signs when driving. This would help people because this kind of technology could be used in self-driving cars to make driving safer for the passengers.
I hypothesize that the machine learning tool Teachable Machine can be used to create a program that would allow a computerized object to identify road signs. Other information that supports my hypothesis is that many people use such technology everyday for things like facial recognition on their phones and devices. This technology works the same way, just with road signs and not people. I know that this hypothesis is testable because it uses only a computer, phone, and notebook, and all variables of concern can be controlled.
I plan to input images into Teachable Machine where the program will assess the images and evaluate whether or not the signs are the same or apply to what the machine has been taught. After that, I will test the model’s output by putting as many different pictures of the signs as I can in different environments to see if it can be differentiated. To assess the resulting data, I will make graphs and use the provided "confusion matrix" from Teachable Machine which shows how many of the images were categorized correctly, and how many were not.
Bibliography:
“Teachable Machine Startup Guide.” Google, Google, https://teachablemachine.withgoogle.com/faq#Getting-started.
“What Is Traffic-Sign Recognition?” J.D. Power, https://www.jdpower.com/cars/shopping-guides/what-is-traffic-sign-recognition.
“Facial Recognition in Smartphones.” Wikipedia, Wikimedia Foundation, 15 Feb. 2023, https://en.wikipedia.org/wiki/Face_ID#:~:text=Face%20ID%20is%20based%20on,module%20that%20reads%20the%20pattern.
Barla, Nilesh. “Self-Driving Cars With Convolutional Neural Networks (CNN).” Neptune.ai, 25 Jan. 2023, https://neptune.ai/blog/self-driving-cars-with-convolutional-neural-networks-cnn.
DriversEd.com. “Helpful Driving Information: Signs.” Driving Information | DriversEd.com, https://driversed.com/driving-information/signs-signals-and-markings/signs/.
Questions and Answers
1. What was the major objective of your project and what was your plan to achieve it?
My major objective was to see whether machine learning programs were capable of being trained to recognize certain signs. I wanted to look at the impact of this kind of technology in another type of technology, such as a self-driving car.
a. Was that goal the result of any specific situation, experience, or problem you encountered?
It was not a result of one of my experiences, but I have always been curious about how computer science can be implemented in technologies like self-driving cars and the impact it would have on its users.
b. Were you trying to solve a problem, answer a question, or test a hypothesis?
I was trying to test a hypothesis. I believed that technology like the computer program Teachable Machine could be used to create a model that could recognize road signs.
2. What were the major tasks you had to perform in order to complete your project?
Some major tasks I had to perform were planning my experiment, doing my experiment, and then analyzing and graphing the data.
3. What is new or novel about your project?
a. Is there some aspect of your project's objective, or how you achieved it that you haven't done before?
I have never used an external program to create an experiment, so in doing this experiment, I learned a lot about Teachable Machine and how to use a resource like this.
b. Is your project's objective, or the way you implemented it, different from anything you have seen?
The way I implemented the project and its real-world application are different from what I've seen. I've seen some people try similar projects, but never with the intention or intent of learning its real-world applications. In this way, my connections are different than they are for other people.
c. If you believe your work to be unique in some way, what research have you done to confirm that it is?
I have looked at how other people have done similar experiments, and some of them are very similar to mine, considering that there is only one way to really use the program. However, as I stated I believe my real-world situations are different than they may be for others.
4. What was the most challenging part of completing your project?
The most challenging part of completing my experiment was doing the actual experiment. It required a focused time without distractions, and it required me moving in and out of my surrounding environment and required a lot of data collection.
a. What problems did you encounter, and how did you overcome them?
I had a hard time choosing my project. I was conflicted on whether to do a project with human participants, or whether to choose a project that I could fit well within my time span. I eventually chose to do the project which fit better within my time span, and it yielded a better result than the other experiment would have in the time that we were given to prepare.
b. What did you learn from overcoming these problems?
I learned the importance of time management from this problem, and how important it is to work on something consistently.
5. If you were going to do this project again, are there any things you would you do differently the next time?
If I were to do this project again, I would have figured other ways to train a model and used different programs to see if they would give different or the same results, and if they gave different results, learn more about why. This would also create a good basis for different answers and may allow for me to have more accurate and thoroughly checked data.
6. Did working on this project give you any ideas for other projects?
Working on this project gave me ideas for other variations of the same project, mostly with improvements like the ones I listed in the previous question.
7. How did COVID-19 affect the completion of your project?
In this case, COVID-19 did not affect the completion of my project, since it had to do with computers and only involved me orchestrating the experiment.