Understanding Thermal Face Detection: Challenges and Evaluation

Agrawal, Janhavi; Pant, Aishwarya; Dhamecha, Tejas I.; Singh, Richa; Vatsa, Mayank

doi:10.1007/978-3-319-28501-6_7

Cited by 1 publication

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of this problem, the single camera-based method, which does not require calibration between cameras, has been studied. Among the existing single camera-based methods, the authors in [ 7 , 8 , 9 ] conducted face detection using only one thermal camera. Zin et al proposed three face detection methods, among which the performance of the nighttime face detection using a multi-slit method was the highest [ 7 ].…”

Section: Related Workmentioning

confidence: 99%

“…However, if the position and angle of the camera change, the parameters need to be updated. The authors in [ 8 , 9 ] used adaboost algorithms based on various hand-crafted features. Agrawal et al [ 8 ] developed a face detection that performs a decision-level fusion of two different adaboost results by using Haar-like features and LBP features, respectively.…”

Section: Related Workmentioning

confidence: 99%

“…The authors in [ 8 , 9 ] used adaboost algorithms based on various hand-crafted features. Agrawal et al [ 8 ] developed a face detection that performs a decision-level fusion of two different adaboost results by using Haar-like features and LBP features, respectively. Both adaboosts are trained as thermal images, and the performance of the decision-level fusion that outputs only a small detection box out of two detection boxes of the two adaboosts is the highest.…”

Section: Related Workmentioning

confidence: 99%

“…The researchers in [ 7 , 8 , 9 ] used a thermal camera that is robust to illumination changes to detect faces at night. However, since a thermal camera is expensive, it is not practical to use it as a surveillance camera in various environments.…”

Section: Related Workmentioning

confidence: 99%

“…To solve this problem, a near-infrared (NIR) light camera or a thermal camera has been used in most existing nighttime face detection studies [ 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 ]. In contrast to the visible-light camera, NIR and thermal cameras use infrared (IR) wavelengths; therefore, they are robust to ambient lighting changes and low intensity of illumination.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Face Detection in Nighttime Images Using Visible-Light Camera Sensors with Two-Step Faster Region-Based Convolutional Neural Network

Cho

Baek

Kim

et al. 2018

Sensors

View full text Add to dashboard Cite

Conventional nighttime face detection studies mostly use near-infrared (NIR) light cameras or thermal cameras, which are robust to environmental illumination variation and low illumination. However, for the NIR camera, it is difficult to adjust the intensity and angle of the additional NIR illuminator according to its distance from an object. As for the thermal camera, it is expensive to use as a surveillance camera. For these reasons, we propose a nighttime face detection method based on deep learning using a single visible-light camera. In a long-distance night image, it is difficult to detect faces directly from the entire image due to noise and image blur. Therefore, we propose Two-Step Faster region-based convolutional neural network (R-CNN) based on the image preprocessed by histogram equalization (HE). As a two-step scheme, our method sequentially performs the detectors of body and face areas, and locates the face inside a limited body area. By using our two-step method, the processing time by Faster R-CNN can be reduced while maintaining the accuracy of face detection by Faster R-CNN. Using a self-constructed database called Dongguk Nighttime Face Detection database (DNFD-DB1) and an open database of Fudan University, we proved that the proposed method performs better compared to other existing face detectors. In addition, the proposed Two-Step Faster R-CNN outperformed single Faster R-CNN and our method with HE showed higher accuracies than those without our preprocessing in nighttime face detection.

show abstract

Section: Related Workmentioning

confidence: 99%