Physical Sciences and Mathematics Commons^™

Irrelevant Pixels Are Everywhere: Find And Exclude Them For More Efficient Computer Vision, Caleb Tung, Abhinav Goel, Xiao Hu, Nick Eliopoulos, Emmanuel Amobi, George K. Thiruvathukal, Vipin Chaudhary, Yung-Hisang Lu

Computer Science: Faculty Publications and Other Works

Computer vision is often performed using Convolutional Neural Networks (CNNs). CNNs are compute-intensive and challenging to deploy on power-constrained systems such as mobile and Internet-of-Things (IoT) devices. CNNs are compute-intensive because they indiscriminately compute many features on all pixels of the input image. We observe that, given a computer vision task, images often contain pixels that are irrelevant to the task. For example, if the task is looking for cars, pixels in the sky are not very useful. Therefore, we propose that a CNN be modified to only operate on relevant pixels to save computation and energy. We propose a …

Modular Neural Networks For Low-Power Image Classification On Embedded Devices, Abhinav Goel, Sara Aghajanzadeh, Caleb Tung, Shuo-Han Chen, George K. Thiruvathukal, Yung-Hisang Lu

Computer Science: Faculty Publications and Other Works

Embedded devices are generally small, battery-powered computers with limited hardware resources. It is difficult to run deep neural networks (DNNs) on these devices, because DNNs perform millions of operations and consume significant amounts of energy. Prior research has shown that a considerable number of a DNN’s memory accesses and computation are redundant when performing tasks like image classification. To reduce this redundancy and thereby reduce the energy consumption of DNNs, we introduce the Modular Neural Network Tree architecture. Instead of using one large DNN for the classifier, this architecture uses multiple smaller DNNs (called modules) to progressively classify images …

A Framework Architecture For Student Learning In Distributed Embedded Systems, William L. Honig, Konstantin Läufer, George K. Thiruvathukal

Computer Science: Faculty Publications and Other Works

Academic courses focused on individual microcomputers or client/server applications are no longer sufficient for students to develop knowledge in embedded systems. Current and near-term industrial systems employ multiple interacting components and new network and security approaches; hence, academic preparation requires teaching students to develop realistic projects comparable to these real-world products. However, the complexity, breadth, and technical variations of these real-world products are difficult to reproduce in the classroom. This paper outlines preliminary work on a framework architecture suitable for academic teaching of modern embedded systems including the Internet of Things. It defines four layers, two of which are at …

Intelligent Systems Development In A Non Engineering Curriculum, Emily A. Brand, William L. Honig, Matthew Wojtowicz

Computer Science: Faculty Publications and Other Works

Much of computer system development today is programming in the large - systems of millions of lines of code distributed across servers and the web. At the same time, microcontrollers have also become pervasive in everyday products, economical to manufacture, and represent a different level of learning about system development. Real world systems at this level require integrated development of custom hardware and software.

How can academic institutions give students a view of this other extreme - programming on small microcontrollers with specialized hardware? Full scale system development including custom hardware and software is expensive, beyond the range of any …

Putting A Slug To Work, Konstantin Läufer, George K. Thiruvathukal, Ryohei Nishimura, Carlos Ramirez Martinez-Eiroa

Computer Science: Faculty Publications and Other Works

In this article, the authors explore various uses of inexpensive embedded Linux devices such as the Linksys NSLU2 ("slug"). Embedded computing is a topic of growing interest. Although novel architectures such as cell processors, graphics processors (GPUs), and FPGAs are growing in popularity, conventional microproessor designs such as Intel's Xscale (ARM) and Atom pack a punch in a small footprint, not to mention being widely supported by commodity operating system and development tools. We're convinced that this entire space is a tool worth keeping in the scientific programmer's and software developer's toolchests.