One Laptop Per Child (OLPC) is an organization dedicated to create educational opportunities for the world's poorest children by providing each child with a rugged, low-cost, low-power, connected laptop (XO) with content and software designed for collaborative, joyful, self-empowered learning. As community engineers associated with this unique proposition, we have constantly evolved our skills to align ourselves with the mission statement and develop tools for educational purposes.

What it does

Our solution offers an Artificial Intelligence-based object detection system that utilizes blockchain solutions for sorting information obtained from a variety of cameras. With just a cell phone, users are offered a serverless solution that can detect objects in real time and more object types for better accuracy.

How we built it

Humans at the mere sight of an image are capable of identifying what objects are there in an image and develop all possible relations between objects in that image. The human detection system when mimicked is capable of generating new solutions in different domains. The current image recognition softwares use classifiers and run feature sets across the image to detect objects in the image, this does lead to a great deal of accuracy. Some of the best ways to get involved with object detection is using cognitive services from Microsoft Vision, TensorFlow [1] and Yolo [2]. The proposed working solution uses a python based wrapper and openCV [3] to feed real time data into the neural network and take intelligent actions from there on.

By running a set of same images on cognitive systems Microsoft Vision, TensorFlow and Yolo, a comparative analysis is discussed and based on which certain conclusions are drawn. Factors contributing to selection primarily include speed and accuracy under various stress factors which include distance, light, network connectivity and cost. The stress conditions are taken into consideration after a series of observations and discussions with farmers where the problem is dominant.

With the solution aimed to serve vehicles belonging to citizens of different income groups, it is important to keep the solution cost effective. Cognitive support is the backbone of this system. It is a must to choose cognitive support which comes at minimal cost and satisfies the trade off between cost and accuracy. Despite the fact, providing extremely high accuracies consistently, cognitive support offered by Microsoft comes at a price which makes it not suitable to use in the module, while TensorFlow and Yolo are free to use. Apart from that, we find Microsoft cognitive support requires the continuous presence of an active internet connection which again comes at a cost and thus, further adding to t he amount required for buying and maintaining the product. With this, the module let go of Microsoft Cognitive support despite of its amazing accuracy and consistent predictions. To perform comparative analysis between TensorFlow and Yolo, we run these free cognitive services on a same video[4]. The video is trimmed such that it spans for a duration of 20 seconds. It is important to note that the video strictly contains buffalos / cows. So any other object predicted can be treated as a wrong prediction and will be treated as 0% accurate on cow/ buffalo.

Using cognitive service of Yolo, which plots the prediction number vs accuracy for the first 100 predictions, we see that the number of sharp drops are less as compared to TensorFlow and the detection is consistent over a long period of time despite consistency slightly being on the lower side. It is also evident that in the trade off between consistency and accuracy Yolo performs better by providing the required consistency over a longer duration. The application of a Convolutional Neural Network over a predicted set of boxes instead of the whole image where the confidence of finding an object in the image is greater than a set threshold makes Yolo superior and the system believes it would be the most apt cognitive service to fit in the module.

Alarm Clock Screencast -


  1. Mart´ın Abadi et. al. TensorFlow: Large-scale machine learning on heterogeneous systems, 2015. Software available from
  2. Joseph Redmon et. al. You only look once: Unified, real-time object detection. CoRR, abs/1506.02640, 2015.
  3. Culjak Ivan et. al. A brief introduction to opencv. In MIPRO, 2012 proceedings of the 35th international convention, pages 1725–1730. IEEE, 2012.
  4. Zhu Liya and Yong Shaohong. The effect of corporate social responsibility on employees. In 6th International Conference on Information Management and Industrial Engineering, volume 1, pages 268–271. IEEE, 2013.

Challenges we ran into

Our solution supports the notions that users should “share by default” and be able to “explore, express, debug, and critique.” Thus our solution puts an emphasis on “activities” rather than “applications.” The foundation will focus on solving the challenges that are relevant to these aspects of the interface, namely:

To make it “simple” to share user activities. This will require an architecture that allows discovery of activities.

To create versions of software that run on multiple operating systems and on multiple hardware platforms. It should be “simple” to install application everywhere. Specifically, it means packaging for every distribution and every virtual machine—removing hardware-related dependencies wherever possible.

To make it “simple” to write software activities. This necessitates stable APIs and example code that uses these APIs.

To make software activities even more secure. Our principal user community is comprised of commoners; they must be protected from malware, phishing, botnets, etc.

Accomplishments that we're proud of

Special Award Presented to SEETA, 24th Global Contest, South Korea The award was presented to SEETA on behalf of its remarkable results at 24th Global Software Contest hosted by IPAK and NIPA, South Korea.

Publications on open source spreadsheet platform:

"SocialCalc: A Spreadsheet Activity for Computer Supported Collaborative Learning", Manu Sheel Gupta, K.S. Preeti, Vijit Singh, Proceedings of the 2010 Conference on Frontiers in Education: Computer Science and Computer Engineering, FECS 2010, Las Vegas, Nevada, U.S.A., CSREA Press 2010, ISBN 1-60132-143-0, pp. 304-309 URL -

"Implementation of Private Cloud Computing using Integration of JavaScript and Python", K.S. Preeti, Vijit Singh, Manu Sheel Gupta, The Python Papers Monograph, The PyCon Asia Pacific 2010, Singapore Management University Download URL -

"Spreadsheet on Cloud - Framework for Learning and Health Management System", K.S. Preeti, Vijit Singh, Sushant Bhatia, Ekansh Preet Singh, Manu Sheel Gupta, Proceedings of the EuSpRIG Conference "Spreadsheet Governance - Policy and Practice" ISBN : 978-0-9566256-9-4

Deployment in Uruguay - Plan Ceibal OLPC Deployment Project, Uruguay -

Community work with FCC (Federal Communication Commission) -

Training Video by Dan Bricklin -

Download - SocialCalc on Sugar page at Sugar Activities Catalogue -

What we learned: Our solution is useful only to the extent it is used by the user community. Thus, we are working with users around the world to focus on these key challenges:

To make our solution freely and readily available to users everywhere

To explore and share best practices

To provide a forum for discussion and support for technology for learning

To provide mechanism for evaluation and dissemination of results.

What's next for Security11X

Our solution is here to support community innovation, security, and enterprise. We would like to help community members start projects that help sustain and grow the technology and communities:

To provide local and regional technical and pedagogical support.

To create new AI activities.

To provide localization and internationalization of software, content, and documentation.

To provide integration and customization services.

Built With

Share this project: