Author: isinha

This week, my teammates and I worked on finalizing the parts we’d need for the critical components and ordered some of those to get started with tinkering with them. We also worked on finalizing how our entire design would work, while also determining backups for the different components in the design.

I worked on implementing the code for a color analysis algorithm so I could integrate it with Ishita Kumar’s code. She’s working on the image segmentation code for identifying the fruit and for separating it from its background. I would then plug in my algorithm to analyze the colors of the fruit in order to predict the rottenness of the fruit.

In the upcoming week, I plan on finalizing the code for the color analysis and integrating the code we have so far so that we can work on developing a classifier for ripe v/s rotten bananas over the next 1-1.5 weeks. For now, I’m working with Google images. As a team, I believe we should also start tinkering with the Jetson Nano to see how we can program it and also figure out how we can program the cameras to take pictures at the given intervals, to begin with.

We are behind the schedule we had planned, but that was quite an ambitious one so that even if we’re lagging, we’re still doing okay. We’re working on an updated schedule based on our actual progress. We’ll need to speed up work over the next few weeks in order to have enough time for testing.

This week, my teammates and I have been working on finalizing our final design and determining our parts list accordingly. We plan on getting this approved on Monday. Attached below is the preliminary design I drew out for our product:

Besides this, I’ve also been looking into the color analysis of bananas to write up the code for a rottenness predictor based on color.  So far, I’ve primarily researched methods for this and have tried to understand how color analysis is performed. I’ve also looked into how we can develop a rottenness predictor given a dataset of images classified as rotten, ripe, and unripe.

We are behind what we had planned in our schedule in that we haven’t yet ordered parts, but I believe that’s okay since we are waiting to confirm our parts list on Monday, after which we’ll order the parts. Besides that, in terms of working on color analysis and starting work on rottenness prediction, I believe I’m on track.

In the upcoming week, I hope my team and I can finalize the parts list and order it. I also plan on downloading Google images of bananas, making the background white if it isn’t already, and developing a dataset of rotten, ripe, and unripe bananas. I also want to work on the color analysis algorithms to train the dataset so that my team and I can then test the algorithm’s predictions before we develop our actual dataset.

This week, in our weekly team meeting with Professor Savvides and our TA, Uzair, we worked on reasoning about the feasibility of our project and what we could modify about our proposed MVP to make it better suit our needs. We realized that since bananas rot quite quickly, using a banana in the MVP would be a good idea. We also discussed ways in which we could get pictures of the fruits from all angles. Another idea that was brought up was that we could predict the rottenness of the fruit, or how long it was expected to last, and this sounded like quite an interesting idea for our final project!

Initially, we were considering only round fruits in order to be able to cover all of the sides with ease using a rotating disk. However, when we started considering bananas, we shifted our focus to find ways in which we can hold the fruit up and click pictures from all angles to check for rotting. I spent some time this week thinking about the design of our model. We could have a holder that could hold our fruits up, and then we could have a camera on a rotating stand that could click pictures of the fruit from all angles.

However, this would imply we need quite a heavy-weight stand to hoist the camera on since rotating the stand would not be easy. Thus, an alternative is that we could have the holder rotate the fruit. For this, I looked into rotating disks. To the rotating disk, we could attach a hook to hold the banana (or any other fruit), and then, as the fruit would rotate, the camera would capture pictures. When I looked up rotating disks online, I usually found ones that looked like the one below:

Such a rotating disk would’ve suited our earlier model with round fruits, but it doesn’t work for our current model. We would want rotating disks that could have a hook attached to them using which we could hang the fruit. Thus, I thought we would probably need to 3D print the rotating disk so that we can attach a hook to it, or design it such that it already has a hook built into it, from which we can hang the fruit.

We seem to be on schedule for now. Over the next week, first off, I plan on updating our project proposal presentation in order to ensure we are putting across our idea clearly. Post that, I plan on looking into image segmentation and edge/color detection algorithms so that I understand them better since we would need these algorithms not only to identify the fruit but also to be able to predict rottenness.