Abstract
Recently, human detection has been used in various applications. Although visible light cameras are usually employed for this purpose, human detection based on visible light cameras has limitations due to darkness, shadows, sunlight, etc. An approach using a thermal (far infrared light) camera has been studied as an alternative for human detection, however, the performance of human detection by thermal cameras is degraded in case of low temperature differences between humans and background. To overcome these drawbacks, we propose a new method for human detection by using thermal camera images. The main contribution of our research is that the thresholds for creating the binarized difference image between the input and background (reference) images can be adaptively determined based on fuzzy systems by using the information derived from the background image and difference values between background and input image. By using our method, human area can be correctly detected irrespective of the various conditions of input and background (reference) images. For the performance evaluation of the proposed method, experiments were performed with the 15 datasets captured under different weather and light conditions. In addition, the experiments with an open database were also performed. The experimental results confirm that the proposed method can robustly detect human shapes in various environments.
Highlights
With the recent development of computer vision and pattern recognition technologies, human detection has been used in various applications, including intelligent surveillance systems [1,2,3,4,5,6,7,8,9]
The condition that a human area is brighter than the surrounding areas in a thermal image is typically satisfied during night and winter, but in summer, the condition is changed, and the brightness of a human image is darker than the background during summer or on a hot day
Wepropose proposea athree three step system overviewofofthe theproposed proposedmethod method is is presented presented in step system forfor detecting humans in a thermal image: generation of a background image; obtaining detecting humans in a thermal image: generation of a background image; obtaining a a difference system with with the thebackground backgroundand andinput inputimage; image;and and detection difference image image based based on on fuzzy fuzzy system detection of of humans in the image
Summary
With the recent development of computer vision and pattern recognition technologies, human detection has been used in various applications, including intelligent surveillance systems [1,2,3,4,5,6,7,8,9]. Various conditions from images captured at different times and from different views can affect the accuracy of the results They require a significant amount of processing time to detect humans because of the need to scan the entire region of the image. In previous studies [34,35], fuzzy-based methods for background subtraction have been employed The advantage of these methods is that they can be applied to multiple conditions of images that have various object sizes. If there are motionless people located in the same positions in all the frames, these people can be factors for generating erroneous backgrounds, and degradation of performance for human detection can occur because of these erroneous backgrounds To overcome these drawbacks, we present a new approach to detect human areas in a thermal image under varying environmental conditions.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
