Robust Lane Detection From Continuous Driving Scenes Using Deep Neural Networks

Zou, Qin; Jiang, Hanwen; Dai, Qiyu; Yue, Yan; Chen, Long; Wang, Qian

doi:10.1109/tvt.2019.2949603

Cited by 276 publications

(153 citation statements)

References 59 publications

(52 reference statements)

Supporting

Mentioning

151

Contrasting

Order By: Relevance

“…The same metrics as in the binary segmentation tasks in references [ 6 , 7 , 8 , 9 , 38 , 46 , 49 , 50 , 51 ] are applied here. Pixel Accuracy (PA), Mean Pixel Accuracy (MPA), Recall, Precision, and F1 score are used to evaluate our models’ segmentation performance, and can be calculated by the following formula

…”

Section: Proposed Methodsmentioning

confidence: 99%

UnetDVH-Linear: Linear Feature Segmentation by Dilated Convolution with Vertical and Horizontal Kernels

Liao

Cao

et al. 2020

Sensors

View full text Add to dashboard Cite

Linear feature extraction is crucial for special objects in semantic segmentation networks, such as slot marking and lanes. The objects with linear characteristics have global contextual information dependency. It is very difficult to capture the complete information of these objects in semantic segmentation tasks. To improve the linear feature extraction ability of the semantic segmentation network, we propose introducing the dilated convolution with vertical and horizontal kernels (DVH) into the task of feature extraction in semantic segmentation networks. Meanwhile, we figure out the outcome if we put the different vertical and horizontal kernels on different places in the semantic segmentation networks. Our networks are trained on the basis of the SS dataset, the TuSimple lane dataset and the Massachusetts Roads dataset. These datasets consist of slot marking, lanes, and road images. The research results show that our method improves the accuracy of the slot marking segmentation of the SS dataset by 2%. Compared with other state-of-the-art methods, our UnetDVH-Linear (v1) obtains better accuracy on the TuSimple Benchmark Lane Detection Challenge with a value of 97.53%. To prove the generalization of our models, road segmentation experiments were performed on aerial images. Without data argumentation, the segmentation accuracy of our model on the Massachusetts roads dataset is 95.3%. Moreover, our models perform better than other models when training with the same loss function and experimental settings. The experiment result shows that the dilated convolution with vertical and horizontal kernels will enhance the neural network on linear feature extraction.

show abstract

…”

Section: Proposed Methodsmentioning

confidence: 99%

UnetDVH-Linear: Linear Feature Segmentation by Dilated Convolution with Vertical and Horizontal Kernels

Liao

Cao

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…The perception module in traditional approaches is replaced by a neural network, to overcome problems mentioned above. Huval [19,20] has indicated that continuous driving scenes could provide information to constrain the predicted lane boundaries. However, these works place significant demands on dataset preparation.…”

Section: Neural-network-based Methodsmentioning

confidence: 99%

SpinNet: Spinning convolutional network for lane boundary detection

et al. 2019

View full text Add to dashboard Cite

In this paper, we propose a simple but effective framework for lane boundary detection, called SpinNet. Considering that cars or pedestrians often occlude lane boundaries and that the local features of lane boundaries are not distinctive, therefore, analyzing and collecting global context information is crucial for lane boundary detection. To this end, we design a novel spinning convolution layer and a brand-new lane parameterization branch in our network to detect lane boundaries from a global perspective. To extract features in narrow strip-shaped fields, we adopt stripshaped convolutions with kernels which have 1 × n or n × 1 shape in the spinning convolution layer. To tackle the problem of that straight strip-shaped convolutions are only able to extract features in vertical or horizontal directions, we introduce the concept of feature map rotation to allow the convolutions to be applied in multiple directions so that more information can be collected concerning a whole lane boundary. Moreover, unlike most existing lane boundary detectors, which extract lane boundaries from segmentation masks, our lane boundary parameterization branch predicts a curve expression for the lane boundary for each pixel in the output feature map. And the network utilizes this information to predict the weights of the curve, to better form the final lane boundaries. Our framework is easy to implement and end-to-end trainable. Experiments show that our proposed SpinNet outperforms state-of-the-art methods.

show abstract

“…Then the training data is preprocessed by -SamSelect‖ and SCAE. In this experiment, we use the preprocessed training data to train the CDBN-based, DBN-based [45] and RNN (Recurrent Neural Network)-based [46] detectors, by comparing the detection accuracy of the three detection models, from Fig. 9, we can see that our detection method can achieve the highest detection accuracy among the three different methods.…”

Section: Figure 8 the Detection Performance With Different Time Obsementioning

confidence: 99%

Real-Time Intrusion Detection in Wireless Network: A Deep Learning-Based Intelligent Mechanism

Yang

Yin

et al. 2020

IEEE Access

View full text Add to dashboard Cite

With the development of the wireless network techniques, the number of cyber-attack increases significantly, which has seriously threat the security of Wireless Local Area Network (WLAN). The traditional intrusion detection technology is a prevalent area of study for numerous years, but it may not have a good detection performance in a real-time way. Therefore, it is urgent to design a detection mechanism to detect the attacks timely. In this paper, we exploit a CDBN (Conditional Deep Belief Network)-based intrusion detection mechanism to recognize the attack features and perform the wireless network intrusion detection in real time. To avoid the impact of the imbalanced dataset and the data redundancy on the detection accuracy, a windowbased instance selection algorithm-SamSelect‖ is adopted to undersample the majority class data samples, and a Stacked Contractive Auto-Encoder (SCAE) algorithm is proposed to reduce the dimension of the data samples. By doing so, our proposed mechanism can effectively detect the potential attack and achieve high accuracy. The experiment results show that CDBN can be effectively combined with-SamSelect‖ and SCAE, and the proposed mechanism has a high detection speed and accuracy, with the average detection time 1.14 ms and the detection accuracy 0.974.

show abstract

Robust Lane Detection From Continuous Driving Scenes Using Deep Neural Networks

Cited by 276 publications

References 59 publications

UnetDVH-Linear: Linear Feature Segmentation by Dilated Convolution with Vertical and Horizontal Kernels

UnetDVH-Linear: Linear Feature Segmentation by Dilated Convolution with Vertical and Horizontal Kernels

SpinNet: Spinning convolutional network for lane boundary detection

Real-Time Intrusion Detection in Wireless Network: A Deep Learning-Based Intelligent Mechanism

Contact Info

Product

Resources

About