Wyniki wyszukiwania - BazTech

1

People tracking in video surveillance systems based on artificial intelligence

Nasry Abir, Ezzahout Abderrahmane, Omary Fouzia

Journal of Automation Mobile Robotics and Intelligent Systems

|

2023

|

Vol. 17, No. 1

59--68

EN

As security is one of the basic human needs, we need security systems that can prevent crimes from happen‐ ing. In general, surveillance videos are used to observe the environment and human behavior in a given location. However, surveillance videos can only be used to record images or videos, without additional information. There‐ fore, more advanced cameras are needed to obtain other additional information such as the position and move‐ ment of people. This research extracted this information from surveillance video footage using a person tracking, detection, and identification algorithm. The framework for these is based on deep learning algorithms, a popu‐ lar branch of artificial intelligence. In the field of video surveillance, person tracking is considered a challenging task. Many computer vision, machine learning, and deep learning techniques have been developed in recent years. The majority of these techniques are based on frontal view images or video sequences. In this work, we will compare some previous work related to the same topic.

2

Fusion of Depth and Thermal Imaging for People Detection

Gutfeter Weronika, Pacut Andrzej

Journal of Telecommunications and Information Technology

|

2021

|

nr 4

53--60

EN

The methodology presented in this paper covers the topic of automatic detection of humans based on two types of images that do not rely on the visible light spectrum, namely on thermal and depth images. Various scenarios are considered with the use of deep neural networks being extensions of Faster R-CNN models. Apart from detecting people, independently, with the use of depth and thermal images, we proposed two data fusion methods. The first approach is the early fusion method with a 2-channel compound input. As it turned out, its performance surpassed that of all other methods tested. However, this approach requires that the model be trained on a dataset containing both types of spatially and temporally synchronized imaging sources. If such a training environment cannot be setup or if the specified dataset is not sufficiently large, we recommend the late fusion scenario, i.e. the other approach explored in this paper. Late fusion models can be trained with single-source data. We introduce the dual-NMS method for fusing the depth and thermal imaging approaches, as its results are better than those achieved by the common NMS.

3

An innovative person detection system based on thermal imaging cameras dedicate for underground belt conveyors

Uth F., Polnik B., Kurpiel W., Kriegsch P., Baltes R., Clausen E.

Mining Science

|

2019

|

Vol. 26

275--288

EN

Within the INESI-project (Increasing Efficiency and Safety Improvement in Underground Mining Transportation Routes) long-wavelength infrared (LWIR) cameras are used for detecting persons on underground belt conveyors or within hazardous areas e.g. in front of crusher or skip vessels by the project partners KOMAG and the Institute for Advanced Mining Technologies (AMT). The test case for evaluating the performance of thermal imaging regarding these applications is the Polish Sobieski underground coal mine operated by Tauron mining company. By the development of thermal image processing algorithms, an automated detection of persons and classification of different objects was achieved. This may allow implementing smart services for person detection on underground belt conveyors as well as material characterization between coal, rock and disturbing objects on belt conveyors.

4

Autonomous person following with 3D LIDAR in outdoor environment

Bohlmann K., Beck-Greinwald A., Buck S., Marks H., Zell A.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2013

|

Vol. 7, No. 2

24--29

EN

The capability of a robot to follow autonomously a person highly enhances its usability when humans and robots collaborate. In this paper we present a system for autonomous following of a walking person in outdoor environments while avoiding static and dynamic obstacles. The principal sensor is a 3D LIDAR with a resolution of 59x29 points. We present a combination of 3D features, motion detection and tracking with a sampling Bayesian filter which results in reliable person detection for a lowresolu tion 3D-LIDAR. The method is implemented on an outdoor robot with car-like steering, which incorporates the target's path into its own path planning around local obstacles. Experiments in outdoor areas validate the approach.