Posted on

Team Status Report for April 16, 2022

An overview of what we did:

  • Allen and Lucky finalized the portions of their subcomponent projects
  • Tak, Allen, and Lucky are in the process of shifting to integration
    • Portiung Tak’s motion detection to the raspberry pi and camera module
    • Setting Lucky’s subcomponent with a laptop and webcam
    • Setting up the demo shelf

In terms of what we will do this upcoming week, we hope to do the following:

  • Set up the demo shelf parameters
    • Get the camera setups finalized (i.e. how far away, how high, what angles, etc.)
    • Live testing of subcomponents
    • Progress to integration and interconnectivity

 

The most significant risks that we know of thus far are:

  • Porting subcomponents to raspberry pi
    • We have already come to face difficulty in getting the motion detection onto a raspberry pi and figuring out communication
    • With delays and what not in getting the second raspberry pi, we may be forced to simply use a laptop and webcam for the object detection component and only one of the raspberry pi’s and camera’s for the motion detection
  • Latency i.e. meeting user requirements
    • After implementing the sliding window technique, we realized the algorithm could potentially take longer than intended, and may face difficulty in meeting the latency standards we set during the abstract and proposal time
  • Communication and storage in the integrated system

 

Changes To our schedule

  • We don’t’ have any significant changes to our schedule
  • If things continue to trend as planned we should be okay with some leeway for troubleshooting
Posted on

Team Status Report for April 10, 2022.

An overview of what we did:

  • Each team member presented at the interim demo, received feedback and carried that into the remaining work for our project.
  • Each member worked on individual subsystems for our pre-integration project.

In terms of what we will do this upcoming week, we hope to do the following:

  • Complete the purchases/manufacture required for our physical space and system (shelf, camera locations, items, etc)
  • Begin the process of having our scripts recognized by and running on the raspberry pi, and also test the integration by making different requests at different times.

 

The most significant risks that we know of thus far are:

  • Different designs of our system (Takshsheel’s idea for the camera) and also integration of the local machine work into one system. 
  • Completion of our subsystems which is so far on point, however, bugs that are hard to debug could possibly cause serious troubles for completion. It’s not a guarantee, but it is likely we’ll see something of that nature, and it would be critical to solve it at the earliest when that happens. 

 

Changes To our schedule

  • There are no new changes to our schedule since the interim demo.
Posted on

Lucky’s Status Report for April 10, 2022

This week I did the following tasks:

  • Presented interim demo
  • Considered solution approaches for image processing
    • Researched segmentation for image processing
      • Researched sliding window segmentation
    • Began implementation design of sliding window segmentation
      • This includes testing considerations
        • The test images must be of equal size to the image that will be taken of the aisle shelf from the fixed distance, otherwise sliding window will be either to small or too large
  • Ordered shelf to begin building for integrated testing

 

In terms of schedule, I think I am on pace. As long as the segmentation code I am in the processes of producing is finished by end of class this upcoming Monday lab session

 

What Next

  • Pick up shelf and set up camera at fixed distance that fits the shelf
    • Determine the image dimensions
    • Formulate the constraints for the sliding window
  • Complete sliding window segmentation and test
  • Proceed to integration
Posted on

Takshsheel’s Status Report for April 10, 2022.

This week, I worked on the following for our capstone project.

  • Performed an interim demo for the background subtraction system which shall be used to detect people walking in and out of frame. Worked on and mentioned a foot traffic counter during the demo, which was incomplete and buggy, which I’ve worked on to fix and as of right now, it can be inaccurate at times, but doesn’t immediately crash like it was previously.
  • Besides the demo and the counter, I drew a design that I’d like to discuss with my teammates during our meeting slot on Monday, although not time-consuming like the debugging for the counter, it is relevant as it could make a last-minute design improvement.
  • Looked into web sockets to transfer data around, since we’d like to use our raspberry pi to run this script and transfer data to our local machine, so, familiarizing myself with the process felt necessary.

In terms of personal progress, I feel like I’ve finally caught back up to where I have wanted to be since the past 2 weeks, and have a good feeling for the project itself.

By next week, I’d like for our team to have made significant progress on the integration of our system, as well as have some physical structure for our system. To make that happen, I’d like to help manufacture the shelves if needed, and set up the camera based off the present/new design that we shall discuss on Monday.

Posted on

Allen’s Status Report for April 10, 2022

  1. Performed our interim demo and demonstrated our individual subsystems as well as our updated schedule. Feedback was received and we are in the process of implementing the feedback in addition to the features we still have to do
  2. Improved upon the web application (i.e. added more views, designed more of the UI, and implemented more functionality). The web application is nearly complete (roughly 90% done)
  3. Set up the raspberry pi and camera system and familiarized myself with the tools and interface.
  4. Researched background subtraction algorithms to integrate with the raspberry pi
  5. Played around with using scripts to SSH into the raspberry pi, run commands, and return the result onto our local computer. This is essentially part of integration from the web application to the rpi/camera system.

I believe that quite a bit was done this week but we are a bit behind (because I would like to be done a week before our final demo to polish up anything). I think we can be in the thick of integration/testing by the end of this week and will reach out to the professors next Monday if things look too behind.

During this upcoming week, I plan to do the following:

  1. Continue integration with the web application and wireless transfer system.
  2. Research integration with the wireless transfer system and the object detection algorithm (Lucky’s computer)
  3. Finish the web application and hosting it on the cloud
  4. Figuring out how to set up our system and shelf/items
    1. i.e. will probably need to buy/find/make a shelf
  5. Assisting Lucky and Takshsheel with anything that they need
Posted on

Takshsheel’s Status Report for April 2, 2022.

This week I worked on the following tasks:

  • Met with our team to discuss the integration framework and subsystems performance and requirements for the same.
  • Continued to fix the buggy counter from last week. Close to done fixing, but not just there yet.
  • Built a testing framework for different cameras and videos and have tested at various distances and speeds of movement with various number of people for my counter.

My progress has been behind since last week and the buggy implementation I’ve got is still not fixed, so this is a concern and risk I’m aware of. From an integration perspective, we need a lot of work done as a team and from me, and through carnival break, I have to be working and getting a fix for my counter program.

For next week, personally I would like to have a deliverable for the counter, which was supposed to be done last week, and from a team perspective, a testable integrated software for our MVP idea.

Posted on

Team Status Report for April 2, 2022

An overview of what we did:

  • We all individually worked on subsystems we were planning on building. Takshsheel worked on his multiple people detection and counting. Our work was centered around finishing up working and testable programs and subsystems. 
  • We prepared what we would like to discuss in the interim demo. For this we met outside of class time and planned and prepared questions and demonstrations.

In terms of what we will do this upcoming week, we hope to do the following:

  • Early upcoming week, so 4/3 and 4/4 we’d like to perfect our tech demo for our subsystems.
  • After 4/4, the focus shall shift to integration and combined tests for the final project. 

 

The most significant risks that we know of thus far are:

  • The most significant risk we have is still the integration of our subsystems. We have met and planned a structure involving web sockets as well. However, the programming needs to be completed.

 

Changes To our schedule

  • There were no new changes recorded to our schedule. Takshsheel’s schedule change from last week is still in affect. 
Posted on

Allen’s Status Report for April 2, 2022

  1. Met with our group, TA, and professor to discuss the status of our project and talked about what to expect for the upcoming interim demo
  2. Added a global data set to the web application as well as a view for new users to pick items they want analyzed by the object detection algorithm

I believe that our progress is behind. With roughly 3 weeks left, I think that we have our work cut out for us. During carnival, I plan to do lots of work and I want our team to start fully integrating and testing our systems with two weeks to go to be safe.

During this upcoming week, I plan to do the following:

  1. Continue researching web sockets as that is a potential bridge that will allow our subsystems to connect with each other (for integration) (besides hosting the web application on the cloud)
  2. Continue making progress on the web application to be ready for integration next week (hopefully)
  3. Discuss what/how to present our interim demo (Sunday night)
  4. Figuring out how to set up our system and shelf/items
    1. i.e. will probably need to buy/find/make a shelf
  5. Assisting Lucky and Takshsheel with anything that they need
Posted on

Lucky’s Status Report for April 2, 2022

This week I did the following tasks:

  • Completed a testable phase 2 object detection subsystem 
  • Ran Tests with SIFT
    • Lowe’s Ratio Test (all else held constant)
      • I tweaked the lowe’s ratio by increments of 5% from 0% to 100%
    • Match Threshold Test (all else held constant)
      • I tweaked the threshold value by increments of 10 from 0 to 200
    • Combination Test (tweaked Lowe’s Ratio and the match threshold to look for the best accuracy parameters)

 

In terms of schedule, I had to miss the meeting in person on Wednesday of this week, but I worked as soon as I could that day to make up for the missed time. In addition, I was able to get more progress on object detection and did some tests. 

 

What Next

  • Prepare to present for the initial interim demo
  • Detection selection Process
    • Compared the detectors with one another and selected the detector with the best combination scores that also fell within the user latency requirements
  • Phase 3 object detection
    • Integrate my subcomponent with the other’s and the camera module
Posted on

Takshsheel’s Status Report for March 26, 2022.

This week, in addition to the common team goals, I worked on the following:

  • I worked on getting a people counter working with the use of contours along with background subtraction in openCV.
  • I also discussed with my teammates about ethical considerations that popped up in my mind after our discussion on monday, particularly to do with how our product could be used, and how much of that should be our responsibility, whether or not that is on us to fix, or is it up to the consumer of our product.
  1. In terms of an issue or risk I found, it has taken me longer than I have expected to fix a bug that I have found in my implementation of contouring, and even though I spent a fair amount of time with it, I’m currently not able to understand the bug. Therefore, I’ve decided to look beyond for outside help from faculty members in CMU for advice on how I might either do this differently, or fixing the issue I have found.

This bug I’ve mentioned here definitely puts me behind schedule. I was hoping to have the counter completed by this weekend but I’m currently still stuck with it. Luckily, for our MVP idea, we just need motion detection which is working. However, this is something I was really looking forward to having in our project, so this setback is rather detrimental.

By next week, I hope to have completed the counter, as well as help with the integration process of our wireless communication system.