Traffic lights detection and recognition based on multi-feature fusion

Wang, Wenhao; Sun, Sumei; Jiang, Mingxin; Yan, Yunyang; Chen, Xiaobing

doi:10.1007/s11042-016-4051-5

Cited by 7 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The two-stage approach allows the advantages of different algorithms in detection and recognition. Compared with the previous two-stage TLDR methods [26][27][28][29][30][31][32][33][34][35], our contribution mainly lies in:…”

Section: Discussionmentioning

confidence: 99%

“…The SVM algorithm has been widely used in previous TLDR methods [27,28,30,31]. The results suggest that SVM can be employed to recognize traffic lights.…”

Section: Recognition Methods Of Traffic Light's Shapementioning

confidence: 99%

“…First Stage: Detection Second Stage: Recognition [27] Visual selective attention model Support vector machines (SVMs) [28] HSV color-space-based SVM [29] HSV color space with maximally stable extremal region CNN [30] RGB color-space-based SVM [31] HSV color space with Otsu algorithm SVM [32] YOLOv3…”

Section: Referencesmentioning

confidence: 99%

See 2 more Smart Citations

A Lightweight Traffic Lights Detection and Recognition Method for Mobile Platform

Wang

Han

Xiang

et al. 2023

Drones

View full text Add to dashboard Cite

Traffic lights detection and recognition (TLDR) is one of the necessary abilities of multi-type intelligent mobile platforms such as drones. Although previous TLDR methods have strong robustness in their recognition results, the feasibility of deployment of these methods is limited by their large model size and high requirements of computing power. In this paper, a novel lightweight TLDR method is proposed to improve its feasibility to be deployed on mobile platforms. The proposed method is a two-stage approach. In the detection stage, a novel lightweight YOLOv5s model is constructed to locate and extract the region of interest (ROI). In the recognition stage, the HSV color space is employed along with an extended twin support vector machines (TWSVMs) model to achieve the recognition of multi-type traffic lights including the arrow shapes. The dataset, collected in naturalistic driving experiments with an instrument vehicle, is utilized to train, verify, and evaluate the proposed method. The results suggest that compared with the previous YOLOv5s-based TLDR methods, the model size of the proposed lightweight TLDR method is reduced by 73.3%, and the computing power consumption of it is reduced by 79.21%. Meanwhile, the satisfied reasoning speed and recognition robustness are also achieved. The feasibility of the proposed method to be deployed on mobile platforms is verified with the Nvidia Jetson NANO platform.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The SVM algorithm has been widely used in previous TLDR methods [27,28,30,31]. The results suggest that SVM can be employed to recognize traffic lights.…”

Section: Recognition Methods Of Traffic Light's Shapementioning

confidence: 99%

See 1 more Smart Citation

A Lightweight Traffic Lights Detection and Recognition Method for Mobile Platform

Wang

Han

Xiang

et al. 2023

Drones

View full text Add to dashboard Cite

show abstract

“…This paper proposed a weighted concatenation step before combining the feature maps from different scales. We added weighting factors to the feature maps of each scale  i , i∈ [1,4].  i were trainable parameters.…”

Section: Improved Unet Modelmentioning

confidence: 99%

“…The current research methods could be divided into color based methods, template matching methods and machine learning methods. Based on the color characteristics, the threshold value was used to segment the target area of interest [1][2][3][4][5][6][7][8]. For example, statistical color histogram and other methods were used to determine the threshold range of the signal light in a certain color space, and then the image was segmented in the corresponding color space.…”

Section: Introductionmentioning

confidence: 99%

Panel Lights Detection Based Automatically Equipment Monitoring Through Deep Learning Method

Wang¹,

Jiang²,

Yang³

et al. 2019

dtcse

View full text Add to dashboard Cite

Automatically equipment monitoring is important in automation field. It asks for the combination of image processing, pattern recognition and the corresponding computer vision technology. Since most equipment status could be read through the panel lights, this paper proposed a deep learning method to automatically monitor devices through detecting their signal lights from panel images captured by digital camera. The entire method employed Unet model to be the backbone model. The feature maps of different scales extracted by the encoder of Unet were weighted fused together to generate the final output of the detected lights. The model were trained on our equipment panel image set. The validation showed that the proposed method could detect the small panel lights accurately.

show abstract

Traffic sign detection and recognition based on pyramidal convolutional networks

Liang

Shao

Zhang

et al. 2019

Neural Comput & Applic

View full text Add to dashboard Cite

Traffic lights detection and recognition based on multi-feature fusion

Cited by 7 publications

References 11 publications

A Lightweight Traffic Lights Detection and Recognition Method for Mobile Platform

A Lightweight Traffic Lights Detection and Recognition Method for Mobile Platform

Panel Lights Detection Based Automatically Equipment Monitoring Through Deep Learning Method

Traffic sign detection and recognition based on pyramidal convolutional networks

Contact Info

Product

Resources

About