No Peeking through My Windows: Conserving Privacy in Personal Drones Alem Fitwi 1 , Yu Chen 1 , Sencun Zhu 2 1 Dept. of Electrical and Computer Engineering, Binghamton University, SUNY, Binghamton, NY 13902, USA 2 Department of Computer Science and Engineering, Penn State University, University Park, PA 16802, USA Emails: {afitwi1, ychen}@binghamton.edu, sxz16@psu.edu Abstract—The drone technology has been increasingly used by many tech-savvy consumers, a number of defense companies, hobbyists and enthusiasts during the last ten years. Drones often come in various sizes and are designed for a multitude of purposes. Nowadays many people have small-sized personal drones for entertainment, filming, or transporting items from one place to another. However, personal drones lack a privacy- preserving mechanism. While in mission, drones often trespass into the personal territories of other people and capture photos or videos through windows without their knowledge and consent. They may also capture video or pictures of people walking, sitting, or doing private things within the drones’ reach in clear form without their go permission. This could potentially invade people’s personal privacy. This paper, therefore, pro- poses a lightweight privacy-preserving-by-design method that prevents drones from peeking through windows of houses and capturing people doing private things at home. It is a fast window object detection and scrambling technology built based on image enhancing, morphological transformation, segmentation and contouring processes (MASP). Besides, a chaotic scrambling technique is incorporated into it for privacy purpose. Hence, this mechanism detects window objects in every image or frame of a real-time video and masks them chaotically to protect the privacy of people. The experimental results validated that the proposed MASP method is lightweight and suitable to be employed in drones, considered as edge devices. Keywords-Privacy, Light-weight Window-Object-Detection, Personal Drones, Chaotic Scrambling, Edge Computing. I. I NTRODUCTION The world has seen a rampant advancement of unmanned aerial vehicles (UAV) technology, also known as drones, over the last decade. They come in a variety of size, sophistication, and they take many roles. Today, they have a wider range of applications in transportation, search and rescue, military, surveillance, communication relays, filming, entertainment, and monitoring [8], [10], [26], [36], [39]. As a result, the public and private sectors, and individuals have shown growing interest in using the drone technologies in a way that serves their purposes. This is likely to cause proliferation of drones in the sky and these drones are capable of garnering a lot of private information about people and places [3]. While hovering in the sky, drones can be directed to collect personal information, monitor and spy people. Furthermore, this information can be divulged into the wider cyberspace because drones are vulnerable to a range of attacks. The owners might not have full control of their drones. Drones are vulnerable to rudimentary interception and interruption attacks. A number of attacks on drones like video stealing, injection of malwares, and device hijacking have been reported since 2007 [2], [8], [13]. It was clearly demonstrated that some commercially available WiFi based drones and satellite based military-grade ones are vulnerable to basic security attacks [7], [8], [12]. This has the potential to cause the widespread of sensitive personal data garnered by the drones without the owners’ permission into the cyberspace. For this reason, people have been growing more and more paranoid about their privacy in relation to the use of drones. To some people, it feels like drones and the invasion of privacy are synonyms. In reality, personal or civilian drones have the capability to pick up virtually every information about what is happening in a certain specific scene. They can capture the images and record the footage of people in that scene without having their sanction to record them. The privacy of people is therefor at risk due to the fact that information, video or images unauthorizedly captured by drones’ cameras and sensors could be abused by the owner of the drone and attackers who manage to penetrate into the drones [7], [17], [23], [34]. As a result, there have been a number of moves to enact laws and regulations to protect privacy and ensure safety in a bunch of developed countries where drones are widely used. For instance, UK, USA, and Canada have recently tried to legislate some laws and regulations in an attempt to address the privacy issues and to ensure safety in the aftermath of some incidents [15], [18], [28], [30], [33]. However, this is not enough to preserve privacy. In fact, they have failed to address the burning privacy concerns. This paper introduces a privacy-conserving technique for personal drones to balance out their use and privacy. A privacy- preserving mechanism based on germane image processing technique is proposed where the personal drones are con- sidered as edge devices with single board computers (SBC) like Raspberry PI 3 B + . It detects window objects in images or real-time video frames and automatically scrambles the windows to prevent peeking through them in violation of the privacy rights of people inside the house. It focuses on enabling the construction of privacy-aware drone systems by design. The window-object-detection and scrambling method are proposed and designed based on a less resource-intensive and faster morphological and segmentation process (MASP) that exploits the very nature of windows. The scrambling method incorporated as part and parcel of the MASP is arXiv:1908.09935v1 [cs.CR] 26 Aug 2019