Lane Image Detection Based on Convolution Neural Network Multi-Task Learning

Li, Junfeng; Zhang, Dehai; Ma, Yu; Liu, Qing

doi:10.3390/electronics10192356

Cited by 10 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model was developed for multi-task learning. Another version of the method is CZF-VPGNet which can be easily implemented in the embedded devices without affecting accuracy [ 33 ]. Chen et al designed a Spatio-temporal attention module (STAM) to integrate into a VGG-16-based FCN network for predicting lanes from multiple consecutive input frames.…”

Section: Literature Reviewmentioning

confidence: 99%

LLDNet: A Lightweight Lane Detection Approach for Autonomous Cars Using Deep Learning

Khan

Haque

Hasan

et al. 2022

Sensors

View full text Add to dashboard Cite

Lane detection plays a vital role in making the idea of the autonomous car a reality. Traditional lane detection methods need extensive hand-crafted features and post-processing techniques, which make the models specific feature-oriented, and susceptible to instability for the variations on road scenes. In recent years, Deep Learning (DL) models, especially Convolutional Neural Network (CNN) models have been proposed and utilized to perform pixel-level lane segmentation. However, most of the methods focus on achieving high accuracy while considering structured roads and good weather conditions and do not put emphasis on testing their models on defected roads, especially ones with blurry lane lines, no lane lines, and cracked pavements, which are predominant in the real world. Moreover, many of these CNN-based models have complex structures and require high-end systems to operate, which makes them quite unsuitable for being implemented in embedded devices. Considering these shortcomings, in this paper, we have introduced a novel CNN model named LLDNet based on an encoder–decoder architecture that is lightweight and has been tested in adverse weather as well as road conditions. A channel attention and spatial attention module are integrated into the designed architecture to refine the feature maps for achieving outstanding results with a lower number of parameters. We have used a hybrid dataset to train our model, which was created by combining two separate datasets, and have compared the model with a few state-of-the-art encoder–decoder architectures. Numerical results on the utilized dataset show that our model surpasses the compared methods in terms of dice coefficient, IoU, and the size of the models. Moreover, we carried out extensive experiments on the videos of different roads in Bangladesh. The visualization results exhibit that our model can detect the lanes accurately in both structured and defected roads and adverse weather conditions. Experimental results elicit that our designed method is capable of detecting lanes accurately and is ready for practical implementation.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

LLDNet: A Lightweight Lane Detection Approach for Autonomous Cars Using Deep Learning

Khan

Haque

Hasan

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Therefore, to design the backbone network reasonably first is necessary. Currently, there are many researches in this field, and some scholars have proposed different network types, including DenseNet and VGG16, which have differences in specific structure and applicability [ 8 , 9 ]. Although ResNet network can obtain more semantic features, it is easy to lose some effective features by adopting additive method in feature fusion, affecting the effect of the application.…”

Section: Multiscale Detection Of Cultural Relics Based On Feature Pyramidmentioning

confidence: 99%

Museum Relic Image Detection and Recognition Based on Deep Learning

Wang

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

To improve the accuracy of museum cultural relic image recognition, the DenseNet and ResNet are selected as the backbone neural networks for detection and recognition. In view of the small target problem in cultural relics, the feature pyramid is introduced in this paper to improve the DenseNet method. The accuracy of target detection is improved through multiscale feature extraction and fusion. At the same time, aiming the problem of weak robustness and feature extraction of cultural relic images, the attention mechanism is proposed to improve ResNet. Therefore, this network can pay attention to the key of feature areas in the image. Finally, the aforementioned methods are verified by experiments. The results show that compared with the YOLOv3 and other algorithms, the accuracy of the improved ResNet proposed in this experiment is above 90%. Furthermore, the number of missed and erroneous detection is the lowest, which are 171 and 134, respectively. The identified mAP indicator accuracy can reach 86%, which also exceeds SVD-Net and DenseNet. It can be seen that the constructed method can effectively detect and recognize the museum cultural relic images.

show abstract

“…LkLaneNet [ 9 ] proposes a novel multi-lane detection method called Large Kernel Lane Network to detect multiple lanes accurately and efficiently in challenging scenarios. ZF-VPGNet [ 10 ] builds a multi-task learning network consisting of multi-label classification, grid box regression, and object mask. This work obtains the correct results and achieves high accuracy and robustness.…”

Section: Introductionmentioning

confidence: 99%

QuantLaneNet: A 640-FPS and 34-GOPS/W FPGA-Based CNN Accelerator for Lane Detection

Khai

Pham

2023

Sensors

View full text Add to dashboard Cite

Lane detection is one of the most fundamental problems in the rapidly developing field of autonomous vehicles. With the dramatic growth of deep learning in recent years, many models have achieved a high accuracy for this task. However, most existing deep-learning methods for lane detection face two main problems. First, most early studies usually follow a segmentation approach, which requires much post-processing to extract the necessary geometric information about the lane lines. Second, many models fail to reach real-time speed due to the high complexity of model architecture. To offer a solution to these problems, this paper proposes a lightweight convolutional neural network that requires only two small arrays for minimum post-processing, instead of segmentation maps for the task of lane detection. This proposed network utilizes a simple lane representation format for its output. The proposed model can achieve 93.53% accuracy on the TuSimple dataset. A hardware accelerator is proposed and implemented on the Virtex-7 VC707 FPGA platform to optimize processing time and power consumption. Several techniques, including data quantization to reduce data width down to 8-bit, exploring various loop-unrolling strategies for different convolution layers, and pipelined computation across layers, are optimized in the proposed hardware accelerator architecture. This implementation can process at 640 FPS while consuming only 10.309 W, equating to a computation throughput of 345.6 GOPS and energy efficiency of 33.52 GOPS/W.

show abstract

Lane Image Detection Based on Convolution Neural Network Multi-Task Learning

Cited by 10 publications

References 22 publications

LLDNet: A Lightweight Lane Detection Approach for Autonomous Cars Using Deep Learning

LLDNet: A Lightweight Lane Detection Approach for Autonomous Cars Using Deep Learning

Museum Relic Image Detection and Recognition Based on Deep Learning

QuantLaneNet: A 640-FPS and 34-GOPS/W FPGA-Based CNN Accelerator for Lane Detection

Contact Info

Product

Resources

About