2016•Unpublished venueRequires access
Pedestrian detection in the radiometric temperature based far infrared thermal images at night time
Tae-Hwan Kim, Taehoon Kim, Minseok Son, Sungho Kim, Eunryung Lee, Masoud Afrakhteh, Miryong Park
Open publisher page 1 citations
Abstract
This paper proposes novel radiometric temperature based far infrared thermal dataset and thermal infrared radiometry-aggregated channel feature (TIR-ACF). The proposed approach achieved improved detection performance by 5% in the temperature dataset for pedestrian detection. Temperature information can be building blocks beyond the limit of gray infrared data processing.