Cruising Review


Publication Title | Real-Time Object Detection for Unmanned Aerial Vehicles based on Cloud-based Convolutional Neural Networks

Drone Information Series

Drone search was updated real-time via Filemaker on:

Drone | Return to Search List

Search Completed | Title | Real-Time Object Detection for Unmanned Aerial Vehicles based on Cloud-based Convolutional Neural Networks
Original File Name Searched: root.pdf | Google It | Yahoo | Bing



Page Number: 001
Previous Page View | Next Page View

Text | Real-Time Object Detection for Unmanned Aerial Vehicles based on Cloud-based Convolutional Neural Networks | 001



Real-Time Object Detection for Unmanned Aerial Vehicles based on Cloud-based Convolutional Neural Networks
ˇ
Jangwon Lee, Jingya Wang, David Crandall, Selma Sabanovic ́ and Geoffrey Fox
Abstract—Real-time object detection is crucial for many applications of Unmanned Aerial Vehicles (UAVs) such as reconnaissance and surveillance, search-and-rescue, and infras- tructure inspection. In the last few years, Convolutional Neural Networks (CNNs) have emerged as a powerful class of models for recognizing image content, and are widely considered in the computer vision community to be the de facto standard approach for most problems. However, object detection based on CNNs is extremely computationally demanding, typically requiring high-end Graphics Processing Units (GPUs) that require too much power and weight, especially for a lightweight and low-cost drone. In this paper, we propose moving the com- putation to an off-board computing cloud. We apply Regions with CNNs (R-CNNs), a state-of-the-art algorithm, to detect not one or two but hundreds of object types in near real-time.
I. INTRODUCTION
In recent years, there has been increasing interest in autonomous UAVs and its applications such as reconnais- sance and surveillance, search-and-rescue, and infrastructure inspection [1, 2, 3, 4, 5]. Visual object detection is an important component in such applications of UAVs, and is critical to develop fully autonomous systems. However, the task of object detection is very challenging, and is made even more difficult by the imaging conditions aboard low- cost consumer UAVs: images are often noisy and blurred due to UAV motion, onboard cameras often have relatively low resolution, and targets are usually quite small. The task is even more difficult because of the need for near real- time performance in many UAV applications, such as when objects are used for navigation.
Many UAV studies have tried to detect and track certain types of objects such as vehicles [6, 7], people including moving pedestrians [8, 9], and landmarks for autonomous navigation and landing [10, 11] in real-time. However, there are only a few that consider detecting multiple objects [12]. despite the fact that detecting multiple target objects is obviously important for many applications of UAVs. In our view, the main reasons for this gap between application needs and technical capabilities are due to two practical but critical limitations: (1) object recognition algorithms often need to be hand-tuned to particular object and context types; (2) it is difficult to build and store a variety of target object models, especially when the objects are diverse in appearance, and (3) real-time object detection demands high computing power even to detect single objects, much less when many target objects are involved.
School of Informatics and Computing, Indiana University, Bloomington, IN 47408, USA. {leejang, wang203, djcran}@indiana.edu. This work was supported in part by The Air Force Office of Scientific Research (AFOSR) and by NVidia.
Fig. 1.
real-time using Convolutional Neural Networks running on a remote cloud.
A drone is able to detect hundreds of object categories in near
However, the first of these problems is eroding due to new breakthrough techniques in computer vision that work well on a wide variety of objects. Most of these techniques are based on “deep learning” with Convolutional Neural Networks, and have delivered striking performance increases on a range of recognition problems [13, 15, 16]. The key idea is to learn the object models from raw pixel data, instead of using hand-tuned features as in tradition recognition ap- proaches. Training these deep models typically requires large training datasets, but this problem has also been overcome by new large-scale labeled datasets like ImageNet [29]. Un- fortunately, these new techniques also require unprecedented amounts of computation; the number of parameters in an object model is typically in the millions or billions, requiring gigabytes of memory, and training and recognition using the object models requires high-end Graphics Processing Units (GPUs). Using these new techniques on low-cost, light- weight drones is thus infeasible because of the size, weight, and power requirements of these devices.
In this paper, we propose moving the computationally- demanding object recognition to a remote compute cloud, instead of trying to implement it on the drone itself, letting us take advantage of these breakthroughs in computer vision technology without paying the weight and power costs. Compute clouds, like Amazon Web Services, also have the advantage of allowing on-demand access to nearly unlim- ited compute resources. This is especially useful for drone applications where most of the processing for navigation and control can be handled onboard, but short bursts of intense computation are required when an unknown object is detected or during active object search and tracking. Using the cloud system, we are able to apply R-CNNs [13], a state-of-the-art recognition algorithm, to detect not one or two but hundreds of object types in near real-time (see Fig. 1). Of course, moving recognition to the cloud intro- duces unpredictable lag from communication latencies. We

Image | Real-Time Object Detection for Unmanned Aerial Vehicles based on Cloud-based Convolutional Neural Networks



real-time-object-detection-unmanned-aerial-vehicles-based-cloud-based-convolutional-neural-networks-001

Conrad Maldives Drone View

Learn more about becoming a digital nomad: Visit the CruisingReview.com website

YouTube Drone Video Site: Visit the YouTube website

Digital Nomad Course: Course Info

Search Contact: greg@cruisingreview.com