|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
Object detection on targets in dark scenario are identified using external illuminated light sources or using thermal scanning of the target, where the existing work produces less precision, low quality, or taking more time for detection of targets in dark scenario. In order to get around these problems, an Infrared light source is used to find targets in the dark scenario. The random forest machine learning algorithm is implemented using the visual studio code programming tool, which relies on the principle of decision trees. Multiple trees are constructed by the random forest algorithm, with each tree based on a subset of features from the same training data source. The infrared light sources with webcam interfaced in the model obtained an accuracy level of 96.3795, error rate precision as 72.56 and error rate falseness as 29.1675. Thus, this paper is proposed for Object detection to find effective targets in an anonymous way under dark scenario.
Keyword
Infrared, Random Forest, Object detection, Dark scenario
PDF Download (click here)
|
|
|
|
|