YOLOv3 Algorithm with additional convolutional neural network trained for traffic sign recognition

Novak, Branislav; Ilic, Velibor; Pavkovic, Bogdan

doi:10.1109/zinc50678.2020.9161446

Cited by 25 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From Table 6, the proposed algorithm achieves a 275% improvement in recognition speed and has greater values of the recall rate, mAP, AP 50 and AP 75 , compared with the two-stage traffic sign recognition algorithm (modified Faster R-CNN [9]). In addition, the proposed algorithm achieves 3.1% increase in the recall rate, 5.4% increase in the mAP, 3.3% increase in the AP 50 and 7.4% increase in the AP 75 , compared with one of the strongest competitors in the one-stage traffic sign recognition algorithm (i.e., modified YOLOv3 [10]). Furthermore, the AP S of the proposed algorithm reaches 24.1%, which has 5.5% increase (compare with the modified YOLOv3).…”

Section: Traffic Sign Recognition Results and Analysismentioning

confidence: 97%

“…This method improves the recognition accuracy to a certain extent, while it loses some speed advantages. Similarly, Branislav et al [10] propose a traffic sign recognition method based on modified YOLOv3 which takes into account both recognition speed and recognition accuracy. However, this algorithm adopting K-means [13] clustering cannot get accurate prior anchors, resulting in its slow recognition speed and reduced recognition accuracy.…”

Section: The State-of-the-art Traffic Sign Recognition Algorithmmentioning

confidence: 99%

“…Although the two-stage traffic sign recognition method (such as modified Faster R-CNN [9]) is proposed, which can achieve high recognition accuracy, its recognition speed is very slow and cannot meet the real-time requirements. The one-stage traffic sign recognition network (such as modified YOLOv3 [10]) can achieve high recognition speed, but its recognition effect is not accurate enough.…”

mentioning

confidence: 99%

See 2 more Smart Citations

A Multiscale Recognition Method for the Optimization of Traffic Signs Using GMM and Category Quality Focal Loss

Gao

Chen

Shi

et al. 2020

Sensors

View full text Add to dashboard Cite

Effective traffic sign recognition algorithms can assist drivers or automatic driving systems in detecting and recognizing traffic signs in real-time. This paper proposes a multiscale recognition method for traffic signs based on the Gaussian Mixture Model (GMM) and Category Quality Focal Loss (CQFL) to enhance recognition speed and recognition accuracy. Specifically, GMM is utilized to cluster the prior anchors, which are in favor of reducing the clustering error. Meanwhile, considering the most common issue in supervised learning (i.e., the imbalance of data set categories), the category proportion factor is introduced into Quality Focal Loss, which is referred to as CQFL. Furthermore, a five-scale recognition network with a prior anchor allocation strategy is designed for small target objects i.e., traffic sign recognition. Combining five existing tricks, the best speed and accuracy tradeoff on our data set (40.1% mAP and 15 FPS on a single 1080Ti GPU), can be achieved. The experimental results demonstrate that the proposed method is superior to the existing mainstream algorithms, in terms of recognition accuracy and recognition speed.

show abstract

Section: Traffic Sign Recognition Results and Analysismentioning

confidence: 97%

Section: The State-of-the-art Traffic Sign Recognition Algorithmmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

A Multiscale Recognition Method for the Optimization of Traffic Signs Using GMM and Category Quality Focal Loss

Gao

Chen

Shi

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…It is known for its high detection accuracy and high detection speed, and it is widely used in vehicle detection, pedestrian detection, ship detection, garbage detection, License plate recognition, etc. [17][18][19][20][21] In addition, YOLOv3 and its improved algorithm are also often used to detect small objects [22,23]. However, there are few studies on using YOLOv3 to detect abnormal behaviors in examinations.…”

Section: Introductionmentioning

confidence: 99%

Examination of Abnormal Behavior Detection Based on Improved YOLOv3

et al. 2021

View full text Add to dashboard Cite

Examination is a way to select talents, and a perfect invigilation strategy can improve the fairness of the examination. To realize the automatic detection of abnormal behavior in the examination room, the method based on the improved YOLOv3 (The third version of the You Only Look Once algorithm) algorithm is proposed. The YOLOv3 algorithm is improved by using the K-Means algorithm, GIoUloss, focal loss, and Darknet32. In addition, the frame-alternate dual-thread method is used to optimize the detection process. The research results show that the improved YOLOv3 algorithm can improve both the detection accuracy and detection speed. The frame-alternate dual-thread method can greatly increase the detection speed. The mean Average Precision (mAP) of the improved YOLOv3 algorithm on the test set reached 88.53%, and the detection speed reached 42 Frames Per Second (FPS) in the frame-alternate dual-thread detection method. The research results provide a certain reference for automated invigilation.

show abstract

“…Assistance System (ADAS) [1,4]. It is a technology in which an AV is competent not only in making decisions, but also in making the right decisions on time [5].…”

mentioning

confidence: 99%

A Real-Time Malaysian Traffic Sign Recognition Using YOLO Algorithm

Mangshor

Paudzi

Ibrahim

et al. 2021

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

In recent years, research towards Autonomous Vehicle (AV) has grown up tremendously. The introduction of Advanced Driving Assistance System (ADAS) in AV has led researchers to explore on the key functionalities that an AV should have, proposes to create a convenience and safe environment where there is no input acquire from a human throughout the journey. One of the key functionalities is the Traffic Sign Recognition (TSR). Hence, this study aims to develop a real-time traffic sign recognition system that can recognize Malaysian traffic sign in real-time and to test the accuracy of the system. The Deep Learning technology using You Only Look Once (YOLO) algorithm is used to train the model for recognition. Five types of warning traffic sign including crossroad, crossroad right, crossroad left, school children crossing and bump from Malaysian traffic sign is used. The test results reached for real-time object detection is 96.00% accuracy on both traffic sign detection and recognition. The finding from this study is believed to be helpful as it may contribute to the ADAS development in the automotive industry for the self-driving vehicle. KeywordsTraffic sign recognition • Autonomous vehicle • YOLO 1 Introduction Nowadays, the automotive industries are vigorously exploring Autonomous Vehicle (AV) which includes the self-driving programs [1]. This technology has been growing instantaneously throughout the world [2]. The AV is a vehicle that the computers take the lead the art of self-driving program [3] using the integration of Advanced Driving

show abstract

YOLOv3 Algorithm with additional convolutional neural network trained for traffic sign recognition

Cited by 25 publications

References 9 publications

A Multiscale Recognition Method for the Optimization of Traffic Signs Using GMM and Category Quality Focal Loss

A Multiscale Recognition Method for the Optimization of Traffic Signs Using GMM and Category Quality Focal Loss

Examination of Abnormal Behavior Detection Based on Improved YOLOv3

A Real-Time Malaysian Traffic Sign Recognition Using YOLO Algorithm

Contact Info

Product

Resources

About