Remote detection of idling cars using infrared imaging and deep networks

• Main Authors: Bastan, Muhammet, Yap, Kim-Hui, Chau, Lap-Pui
• Other Authors: School of Electrical and Electronic Engineering
• Format: Article
• Language: English
• Published: 2020
• Subjects: Engineering::Electrical and electronic engineering; Infrared Image; Car Detection
• Online Access: https://hdl.handle.net/10356/144919
• Institution: Nanyang Technological University

Idling vehicles waste energy and pollute the environment through exhaust emission. In some countries, idling a vehicle formore than a predefined duration is prohibited and automatic idling vehicle detection is desirable for law enforcement. Wepropose the first automatic system to detect idling cars, using infrared (IR) imaging and deep networks. We rely on thedifferences in spatio-temporal heat signatures of idling and stopped cars and monitor the car temperature with a long-wavelength IR camera. We formulate the idling car detection problem as spatio-temporal event detection in IR imagesequences and employ deep networks for spatio-temporal modeling. We collected the first IR image sequence dataset foridling car detection. First, we detect the cars in each IR image using a convolutional neural network, which is pre-trainedon regular RGB images and fine-tuned on IR images for higher accuracy. Then, we track the detected cars over time toidentify the cars that are parked. Finally, we use the 3D spatio-temporal IR image volume of each parked car as input toconvolutional and recurrent networks to classify them as idling or not. We carried out an extensive empirical evaluation oftemporal and spatio-temporal modeling approaches with various convolutional and recurrent architectures. We presentpromising experimental results on our IR image sequence dataset.