A Technical Research on UAV for Object Detection

Abstract

According to a Forbes report published in 2018, US drone industry has witnessed a dramatic commercial growth from merely $40 million in 2012 to a billion in 2017 as documented in the study conducted by MCKinsey and Company in December 2017. It is estimated by MCKinsey that by 2026 the commercial drones will annually impact $31 billion to $46 billion on country’s gross domestic product.

Drones have been widely adopted for data capturing that can be used in the fields of  defense, agriculture, emergency response and disaster management, conservation of endangered species, healthcare etc. For most of the applications of UAV, Real-time object detection is extremely crucial. In the last few years, considering the growth of drone industry and the interest of around 300 companies who are making substantial investments of time and resources in drone many technologies have been emerged for making advancement in the field of UAV focusing on object detection and recognition for UAV.

This paper summarizes a number of object detection techniques proposed till date by researchers. It reports the characteristics and requirements of UAV from object detection viewpoint. The objective of our research is to understand different architectures that are capable of detecting objects from aerial images. The main goal of this survey is to create an insight of an architecture that is accurate, fast, robust and utilizes low computation power.