Traffic light recognition in varying illumination using deep learning and saliency map

John, Vijay; Yoneda, Keisuke; Qi, Bin; Liu, Zheng; Mita, Seiichi

doi:10.1109/itsc.2014.6958056

Cited by 105 publications

(74 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some prior art utilizes digital maps and GPS information to improve the efficiency and accuracy of detection [16,17]. However, prior information is not always accessible and is not in principle necessary (humans make no use of such data).…”

Section: Traffic Light and Traffic Sign Detection And Classificationmentioning

confidence: 99%

“…John et al [17,18] showed that CNNs are effective as classifiers of traffic lights, but they used traditional methods for saliency map generation. In 2016, Zhu et al [5] published the Tsinghua-Tencent 100K dataset for traffic sign benchmark, and developed an end-to-end CNN for both detection and classification.…”

Section: Traffic Light and Traffic Sign Detection And Classificationmentioning

confidence: 99%

See 1 more Smart Citation

Traffic signal detection and classification in street views using an attention model

Zhang

et al. 2018

Comp. Visual Media

View full text Add to dashboard Cite

Detecting small objects is a challenging task. We focus on a special case: the detection and classification of traffic signals in street views. We present a novel framework that utilizes a visual attention model to make detection more efficient, without loss of accuracy, and which generalizes. The attention model is designed to generate a small set of candidate regions at a suitable scale so that small targets can be better located and classified. In order to evaluate our method in the context of traffic signal detection, we have built a traffic light benchmark with over 15,000 traffic light instances, based on Tencent street view panoramas. We have tested our method both on the dataset we have built and the Tsinghua-Tencent 100K (TT100K) traffic sign benchmark. Experiments show that our method has superior detection performance and is quicker than the general faster RCNN object detection framework on both datasets. It is competitive with state-of-theart specialist traffic sign detectors on TT100K, but is an order of magnitude faster. To show generality, we tested it on the LISA dataset without tuning, and obtained an average precision in excess of 90%.

show abstract

Section: Traffic Light and Traffic Sign Detection And Classificationmentioning

confidence: 99%

Section: Traffic Light and Traffic Sign Detection And Classificationmentioning

confidence: 99%

Traffic signal detection and classification in street views using an attention model

Zhang

et al. 2018

Comp. Visual Media

View full text Add to dashboard Cite

show abstract

“…As previously mentioned, the color of traffic lights can be changed due to different lighting situations. Different color spaces have also been investigated to more robustly detect the traffic lights (John et al, 2014) and multiple exposure images are tested to rigorously detect traffic lights in dark and bright environments (Jang et al, 2014). Diaz-Cabrera et al (Diaz-Cabrera et al, 2015) claim that color segmentation using fuzzy clustering can improve the traffic light detection results.…”

Section: Previous Workmentioning

confidence: 99%

Traffic Light Detection Using Conic Section Geometry

Hosseinyalmdary

Yılmaz

2016

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Traffic lights detection and their state recognition is a crucial task that autonomous vehicles must reliably fulfill. Despite scientific endeavors, it still is an open problem due to the variations of traffic lights and their perception in image form. Unlike previous studies, this paper investigates the use of inaccurate and publicly available GIS databases such as OpenStreetMap. In addition, we are the first to exploit conic section geometry to improve the shape cue of the traffic lights in images. Conic section also enables us to estimate the pose of the traffic lights with respect to the camera. Our approach can detect multiple traffic lights in the scene, it also is able to detect the traffic lights in the absence of prior knowledge, and detect the traffics lights as far as 70 meters. The proposed approach has been evaluated for different scenarios and the results show that the use of stereo cameras significantly improves the accuracy of the traffic lights detection and pose estimation.

show abstract

“…In addition, although deep learning-based methods often achieve better prediction performance when compared to traditional machine learningbased methods, it is considered as a back-box approach because of being not interpretable. The saliency map [36] was introduced to visualize image features in classification task at first, now it plays an important role in various practical applications right from video surveillance [37] to traffic light detection [38]. This strategy can help evaluate the degree of genomics features such as aberration attributes to the prediction of drug response.…”

Section: Introductionmentioning

confidence: 99%

Graph convolutional networks for drug response prediction

Nguyen¹,

Nguyen

Le³

2020

Preprint

View full text Add to dashboard Cite

Background: Drug response prediction is an important problem in computational personalized medicine. Many machine learning-, especially deep learning-, based methods have been proposed for this task. However, these methods often represented the drugs as strings, which are not a natural way to depict molecules. Also, interpretation has not been considered thoroughly in these methods. Methods: In this study, we propose a novel method, GraphDRP, based on graph convolutional network for the problem. In GraphDRP, drugs are represented in molecular graphs directly capturing the bonds among atoms, meanwhile cell lines are depicted as binary vectors of genomic aberrations. Representative features of drugs and cell lines are learned by convolution layers, then combined to represent for each drug-cell line pair. Finally, the response value of each drug-cell line pair is predicted by a fully-connected neural network. Four variants of graph convolutional networks are used for learning the features of drugs. Results: We find that GraphDRP outperforms tCNN in all performance measures for all experiments. Also, through saliency maps of the resulting GraphDRP models, we discover the contribution of the genomic aberrations to the responses. Conclusion: Representing drugs as graphs are able to improve the performance of drug response prediction. Data and source code can be downloaded at https://github.com/hauldhut/GraphDRP.

show abstract

Traffic light recognition in varying illumination using deep learning and saliency map

Cited by 105 publications

References 8 publications

Traffic signal detection and classification in street views using an attention model

Traffic signal detection and classification in street views using an attention model

Traffic Light Detection Using Conic Section Geometry

Graph convolutional networks for drug response prediction

Contact Info

Product

Resources

About