![](https://csdnimg.cn/release/download_crawler_static/15137471/bg1.jpg)
Using Infrared HOG-based Pedestrian Detection for Outdoor
Autonomous Searching UAV with Embedded System
Yanhua Shao
a,b
, Yanying Mei
a,b
, Hongyu Chu
a,b
, Zhiyuan Chang
a,b
, Yuxuan He
a
a
School of Information and Engineering, Southwest University of Science and Technology,
Mianyang, Sichuan, China, 621010
b
Robot Technology Used for Special Environment Key Laboratory of Sichuan Province, Southwest
University of Science and Technology, Mianyang, Sichuan, China, 621010
ABSTRACT
Pedestrian detection (PD) is an important application domain in computer vision and pattern recognition. Unmanned
Aerial Vehicles (UAVs) have become a major field of research in recent years. In this paper, an algorithm for a robust
pedestrian detection method based on the combination of the infrared HOG (IR-HOG) feature and SVM is proposed for
highly complex outdoor scenarios on the basis of airborne IR image sequences from UAV. The basic flow of our
application operation is as follows. Firstly, the thermal infrared imager (TAU2-336), which was installed on our Outdoor
Autonomous Searching (OAS) UAV, is used for taking pictures of the designated outdoor area. Secondly, image
sequences collecting and processing were accomplished by using high-performance embedded system with Samsung
ODROID-XU4 and Ubuntu as the core and operating system respectively, and IR-HOG features were extracted. Finally,
the SVM is used to train the pedestrian classifier. Experiment show that, our method shows promising results under
complex conditions including strong noise corruption, partial occlusion etc.
Keywords: HOG descriptors, SVM, infrared imaging, multiscale analysis, outdoor autonomous searching, UAV.
1. INTRODUCTION
Pedestrian detection has become one of the most active topics in computer vision, due to a wide range of promising
applications, such as visual surveillance [1], robotic [2], intelligent vehicles, etc. During last decade, the scientific
research on Unmanned Aerial Vehicles (UAVs) increased spectacularly and led to the design of multiple types of aerial
platforms, which can be used in a wide variety of scenarios [3-6]. Such as Outdoor Autonomous Searching UAV (OAS-
UAV). Numerous infrared (IR) imaging applications have been developed and deployed in fields that include the
military, medicine and industry [7, 8]. Usually, it can capture more detail information than that of the use of visible light
in complex scenes. For example, in general, detect humans in very low light condition may ineffective. But thermal
imaging can see the warm object clearly in the night [9, 10]. HOG descriptor has been presented for pedestrian detection
in CVPR 2005 and have been proven effective [11, 12].
In this paper, an algorithm for a robust pedestrian detection method based on the combination of the infrared HOG
(IR-HOG) feature and SVM is proposed for highly complex outdoor scenarios on the basis of airborne IR image
sequences from UAV. Experiment show that, our method shows promising results under complex conditions including
strong noise corruption, partial occlusion etc.
The outline of the paper is listed as follows: Section 2 introduces our tracking method in detail, and Section 3
presents the experimental results. Finally, some conclusions of our work are drawn in Section 4.
2. THE PROPOSED ALGORITHM
A diagram of our proposed system is shown in Fig. 1. Apparently, our infrared HOG-based pedestrian detection for
Outdoor Autonomous Searching UAV (OAS-UAV) is mainly composed of four steps:
1). Image preprocessing, such as Median filtering and so on, 2). Threshold segmentation, 3). HOG Descriptor
Generation and 4). SVM Classification.