RPO Feedback
Firstly, the feedback for the RPO was that overall it's too generic and the topic needs to be narrowed down further. Some of the statements will need to be backed up with more substantial resources instead of having it as a thought. Therefore, the next step is take smaller bits of information from the current research that I have and perhaps identify the problem from that particular small topic before branching it out.
Well, as I was researching the topic of AI itself, I was pretty overwhelm by the amount of potential and how huge the topic it was. But regardless, I have to move on and start with something small or a task to develope from.
Python with OpenCV
In Relation to Previous Exercise Using the concept of colored image being more valuable, to capture image with color can be in form of cameras, eyes or even art(painting). I used the OpenCV with Python to do a simple exercise with objection detection, I first followed the tutorial on basic OpenCV functions, which led me to the first completed code to detect yellow objects with a detection blob. However, what I wanted to achieve was from just detecting the object, I wanted to gather data that may be personal such as measurements of the object which led me to ChatGPT for further help in the coding process. Subsequently, I was trying out the concept from the previous exercise where RGB values are being measured and determined with a threshold that was set. I also tried color coding the dominant color pixel of the object to being normal,suspicious and dangerous. However, rather those 3 category, I believe it talk about the value of a person through the computer vision? I was thinking of what kind of data would pose a sense of threat for people on data privacy, using computer vision?
Exercise 3
Starting with Python
The first test was to get the basic detection working and creating the bounding box on the objects. I then added the detection for the measurement, however the measurement was not really accurate. I did encounter errors when getting the code to work, but with the help of ChatGPT, I was able to identify the issue efficiently with the explaination provided.
The testing was then further expanded with measuring the 3 RGB values, I then tried to implement meanings to the values showed and determine whether if the item/object is suspicious base on a certain pixel threhold that I've set to trigger the change. I then tried on the different objects that see the effects of this exercise. I also tried changing the value with more meanings that could relate to my topic, like having a certain amount of dominant pixel would determine if the data is rich or poor.
Feedback
Can be too simple if were to interpret RGB values, what is more than RGB value? The value behind the colors or the details? What kind of information can be valuable?
Cookie Collection
As I was researching for my dissertation for more readings, a sudden idea came up to me when there's all these cookies collection from the different websites. Then again, it reminded me of an article that mentioned how people are unaware of what these cookies are used for.
Exercise 4
Cookie Scraping
So the journey started with me collecting the cookie prompts from all the websites that I came across, well I wasn't really sure what to do with them for now, but somehow I thought it will be interesting collecting data from cookie collectors.
As I was going through the cookie images, I was thinking perhaps rather than just manually screenshot these cookie prompts, thus I intend to further this experiment with the help of my own scraping tool using Selenium. To also add value to this experiment, I was thinking how I could relate this to my project as well.