I'm Building Taskito
I'm Building Taskito

Taskito offers simple and intuitive tools to organize your daily tasks and to-do list. It combines features of project management software with productivity and time management tools. Try it out now!

Get it on Google Play Download on the App Store

SimpleCV - Google Summer of Code 2012

on gsoc

My proposal Automatic segregation of conveyor belt items using SimpleCV as submitted to SimpleCV has been accepted for Google Summer of Code 2012. I would like to thank Katherine Scott and Anthony Oliver for their guidance and support and providing me a great opportunity to work with them. I would also like to thank all of them who helped achieve this glory. I would like to congratulate Vijay, Victor and Seunghoon whose proposals got accepted (with SimpleCV) and all the other 1208 students.

I am very excited to kickoff my summer as soon as possible. I will be working with Anthony Oliver as he is my mentor. My project, which is based on Computer Vision, basically consists of two parts. GSoC 2012 Project Page.

Abstract: Automating the process of removal of garbage/debris (unwanted items) from conveyor line systems in factory workflows, thereby segregating wanted & unwanted items. Outliers are identified by an independent feature extraction scheme & tracked along the conveyor belt using TLD algorithm till they are removed.

  1. Identifying the unwanted items or outliers As the items move along the conveyor belt, an overhead camera will identify the items that do not belong there. For this purpose, I would develop an independent feature extraction scheme tailored for this purpose and identify the unwanted items. The location co-ordinates of these items i.e. the ROI (which will be updated in real time for each item) will be conveyed to the system.

  2. Tracking with TLD The co-ordinates i.e. the ROI of the unwanted items, as they enter the field of vision of the camera, will be obtained from closest to furthest position on the conveyor. A novel aspect of this implementation is the automation of the selection of the ROI required in TLD. I will use the ROI co-ordinates obtained above, one at a time, and set the ROI for TLD automatically.Once done, the ROI co-ordinates will be communicated to the removal mechanism viz. A robotic manipulator.

Deliverables A novel feature extraction scheme, which is both robust & real-time, tailored to the cause of identifying outliers (different objects) in an image will be developed. As the primary aim of the project, a SimpleCV port to the entire OpenTLD algorithm will be made & used. A demonstration of the final application will be provided by simulating a conveyor belt and tracking the outliers in the various objects on the same.

I will use Python and OpenCV, SimpleCV libraries. I plan to port OpenTLD to Python/SimpleCV.

I will post all my project updates here.

P.S. I hope to have an outstanding summer.

Playing around with Android UI

Articles focusing on Android UI - playing around with ViewPagers, CoordinatorLayout, meaningful motions and animations, implementing difficult customized views, etc.

Read next