Figure 4 - uploaded by Xiaozheng (Sean) He
Content may be subject to copyright.
Similar publications
As a single hidden layer feed-forward neural network, the extreme learning machine (ELM) has been extensively studied for its short training time and good generalization ability. Recently, with the deep learning algorithm becoming a research hotspot, some deep extreme learning machine algorithms such as multi-layer extreme learning machine (ML-ELM)...
To get higher accuracy of the CNN action recognition networks with video inputs, many methods will deepen or modify the convolutional layers of the original networks. However, this would result in a substantial increase of the parameters and cost resources. In this paper, we propose an efficient and versatile method with good transfer performance t...
Citations
... DAS and automated TSRs have been hot research topics in the last decade. As a result, multiple frameworks have been proposed for efficient TSRs in complex and resourceconstrained environments [25]. These proposed frameworks can be divided into traditional and AI&ML-based methods, respectively. ...
Recognizing and classifying traffic signs is a challenging task that can significantly improve road safety. Deep neural networks have achieved impressive results in various applications, including object identification and automatic recognition of traffic signs. These deep neural network-based traffic sign recognition systems may have limitations in practical applications due to their computational requirements and resource consumption. To address this issue, this paper presents a lightweight neural network for traffic sign recognition that achieves high accuracy and precision with fewer trainable parameters. The proposed model is trained on the German Traffic Sign Recognition Benchmark (GTSRB) and Belgium Traffic Sign (BelgiumTS) datasets. Experimental results demonstrate that the proposed model has achieved 98.41% and 92.06% accuracy on GTSRB and BelgiumTS datasets, respectively, outperforming several state-of-the-art models such as GoogleNet, AlexNet, VGG16, VGG19, MobileNetv2, and ResNetv2. Furthermore, the proposed model outperformed these methods by margins ranging from 0.1 to 4.20 percentage point on the GTSRB dataset and by margins ranging from 9.33 to 33.18 percentage point on the BelgiumTS dataset.
... Tabernik et al. [42] employ Mask RCNN to achieve good traffic sign detection performance with several improvements, such as data augmentation and hard sample mining. Besides the above TSD methods, there are also many works focusing on TSR [41,51,54]. For the sake of space, more details will not be described here. ...
As a crucial sub-task for autonomous driving and intelligent transportation, traffic sign detection attracts a lot of researchers’ attention. Although prior works have achieved promising results, it still suffers from two problems, the need of massive labeled data and the slow inference speed on high-resolution images, which are also the common problem of generic object detection. Motivated by these problems, we propose a lightweight traffic sign detection model and employ an automatic data labeling pipeline. The detection model utilizes a cascaded structure that consists of a localization module and a recognition module, achieving quite fast speed on GPU and edge devices. We discard all complex structures and operations to boost the speed of the model, making it easy for deployment. Then, we propose a two-stage automatic data labeling pipeline to reduce the cost of data labeling work. With only traffic sign template images, a synthetic dataset is constructed for generating initial pseudo labels in the first stage. In the second stage, we propose to use an image pretext model to refine the initial labels. The accuracy of the final pseudo labels is nearly 100%. We test the proposed method on TT-100K, GTSDB, and GTSRB datasets, and the results show that the model trained with the pseudo labels only has a negligible accuracy loss compared with the model trained by real labels. The proposed model’s calculation latency is around 1 ms on GPU, and the accuracy is still on par with the state-of-the-art models.
... TSR has always been a hot research topic in recent years. For this purpose, TSR is investigated to detect traffic sign region and non-traffic sign area in complex scene of images, TSR is to extract the specific features represented through traffic sign patterns [20]. The existing TSR methods are basically grouped into two categories: One is based on traditional methods, the other is related to deep learning methods. ...
Intelligent Transportation System (ITS), including unmanned vehicles, has been gradually matured despite on road. How to eliminate the interference due to various environmental factors, carry out accurate and efficient traffic sign detection and recognition, is a key technical problem. However, traditional visual object recognition mainly relies on visual feature extraction, e.g., color and edge, which has limitations. Convolutional neural network (CNN) was designed for visual object recognition based on deep learning, which has successfully overcome the shortcomings of conventional object recognition. In this paper, we implement an experiment to evaluate the performance of the latest version of YOLOv5 based on our dataset for Traffic Sign Recognition (TSR), which unfolds how the model for visual object recognition in deep learning is suitable for TSR through a comprehensive comparison with SSD (i.e., single shot multibox detector) as the objective of this paper. The experiments in this project utilize our own dataset. Pertaining to the experimental results, YOLOv5 achieves 97.70% in terms of mAP@0.5 for all classes, SSD obtains 90.14% mAP in the same term. Meanwhile, regarding recognition speed, YOLOv5 also outperforms SSD.
... Sun et al. [8] proposed the TSR method of integrating multilayer features and a kernel Extreme Learning Machine (ELM) classifier. The CNN was composed of three convolution layers, each followed by a pooling layer, and finally by a fully connected output layer. ...
Traffic sign recognition (TSR) is a crucial intelligent transport system. Nowadays, the convolutional neural network has become a vital tool in the conception of a TSR model. In this work, we propose an improved TSR algorithm for the transportation system inspired by the classical model LeNet-5. In this model, firstly, we replace the hyperbolic tangent activation function with a self-regularized non-monotonic activation function called SigmaH \(({\text{SigmaH}}(x) = x\tanh (\sqrt {\sigma \left( x \right)} ))\). SigmaH is experimentally validated on various popular benchmarks against the most suitable combinations of architectures and activation functions. Secondly, we use a convolutional block attention module, which is beneficial for extracting the most valuable features using the attention method. Finally, we combine the triplet-center loss with the Softmax activation function as a loss function to maximize the correct recognition rate. The TSR experiments are carried out based on the German Traffic Sign Recognition benchmark. The experimental results demonstrate that the improved LeNet-5 has an identification accuracy rate of 98.25%, and the average processing time per frame is 8 ms. In the meantime, the number of parameters is reduced by more than 51% compared with the classic LeNet-5 model. Our proposed model has remarkable accuracy and high training efficiency compared with other algorithms.
... In recent years, vehicle re-identi cation (Re-ID) has received more attention, and it can be broadly employed in video surveillance, intelligent traf c, and other elds [1,2]. Especially when the license plate is blocked, removed, or even forged, vehicle Re-ID will evolve into the exclusive approach for traf c control departments to nd escape vehicles. ...
... Compared with person Re-ID [4], the vehicle Re-ID is a more challenging task in the following two aspects: (1) in the case of uncontrolled natural lighting, viewing angle, low resolution, and complex background, the visual appearance of the same car under different camera viewpoint changes greatly, which shows distinct intra-class differences, as shown in Fig. 1a. (2) vehicles with the same model have similar visual appearances, such as the same color and model characteristics, leading to obvious inter-class similarity, as shown in Fig. 1b [5]. In order to solve the mentioned challenges, most of the current research adopt deep learning methods to automatically extract vehicle image features. ...
With the increasing application of surveillance cameras, vehicle re-identification (Re-ID) has attracted more attention in the field of public security. Vehicle Re-ID meets challenge attributable to the large intra-class differences caused by different views of vehicles in the traveling process and obvious inter-class similarities caused by similar appearances. Plentiful existing methods focus on local attributes by marking local locations. However, these methods require additional annotations, resulting in complex algorithms and insufferable computation time. To cope with these challenges, this paper proposes a vehicle Re-ID model based on optimized DenseNet121 with joint loss. This model applies the SE block to automatically obtain the importance of each channel feature and assign the corresponding weight to it, then features are transferred to the deep layer by adjusting the corresponding weights, which reduces the transmission of redundant information in the process of feature reuse in DenseNet121. At the same time, the proposed model leverages the complementary expression advantages of middle features of the CNN to enhance the feature expression ability. Additionally, a joint loss with focal loss and triplet loss is proposed in vehicle Re-ID to enhance the model’s ability to discriminate difficult-to-separate samples by enlarging the weight of the difficult-to-separate samples during the training process. Experimental results on the VeRi-776 dataset show that mAP and Rank-1 reach 75.5% and 94.8%, respectively. Besides, Rank-1 on small, medium and large sub-datasets of Vehicle ID dataset reach 81.3%, 78.9%, and 76.5%, respectively, which surpasses most existing vehicle Re-ID methods.
... For machine learning, the K-means algorithm is one of the simplest and popular for clustering. It does not have labels or results in data processing, so it is called unsupervised learning [10,26]. The K-means algorithm's primary idea is to group (cluster) related data sets values and recognize underlying designs. ...
For the advancement of the Internet of Things (IoT) and Next Generation Web, various applications have emerged to process structured or unstructured data. Latency, accuracy, load balancing, centralization, and others are issues on the cloud layer of transferring the IoT data. Machine learning is an emerging technology for big data analytics in IoT applications. Traditional data analyzing and processing techniques have several limitations, such as centralization and load managing in a massive amount of data. This paper introduces a Machine Learning Based Distributed Big Data Analysis Framework for Next Generation Web in IoT. We are utilizing feature extraction and data scaling at the edge layer paradigm for processing the data. Extreme Learning Machine (ELM) is adopting in the cloud layer for classification and big data analysis in IoT. The experimental evaluation demonstrates that the proposed distributed framework has a more reliable performance than the traditional framework.
... Specially, under different viewing angles, the appearances of vehicles change significantly, which leads to the instability of global features [7]. Local location features contain more stable and distinguishable information, such as windows, headlights, license plates, etc [8,9]. These features are critical for judging similar vehicles and will not change greatly with environmental changes, so the features are more robust. ...
Vehicle re-identification (ReID) aims to retrieve the target vehicle in an extensive image gallery through its appearances from various views in the cross-camera scenario. It has gradually become a core technology of intelligent transportation system. Most existing vehicle re-identification models adopt the joint learning of global and local features. However, they directly use the extracted global features, resulting in insufficient feature expression. Moreover, local features are primarily obtained through advanced annotation and complex attention mechanisms, which require additional costs. To solve this issue, a multi-feature learning model with enhanced local attention for vehicle re-identification (MFELA) is proposed in this paper. The model consists of global and local branches. The global branch utilizes both middle and highlevel semantic features of ResNet50 to enhance the global representation capability. In addition,multi-scale pooling operations are used to obtainmultiscale information.While the local branch utilizes the proposed Region Batch Dropblock (RBD), which encourages the model to learn discriminative features for different local regions and simultaneously drops corresponding same areas randomly in a batch during training to enhance the attention to local regions. Then features from both branches are combined to provide a more comprehensive and distinctive feature representation. Extensive experiments on VeRi-776 and VehicleID datasets prove that our method has excellent performance.
... Structured light technology projects a sequence of fringe images on the surface of the measurement object, and the fringe is deformed by the contour of the object. The phase calculation [5][6][7] of the collected fringe image can realize reconstruction of the three-dimensional contour of the object. However, due to the influence of instrument design, gamma nonlinear distortion [8] between the projector and camera will produce measurement phase errors, and the collected grating fringes will not be This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ...
... In current research, training generalization is the trend. In [9], abstract features are obtained for training the classifier that is more resilient to multiple types and shapes of traffic signs. In [10] the authors made one model for detecting cars in different weather condition. ...
Deep Learning is a very promising field in image classification. It leads to the automation of many real-world problems. Currently, Car seatbelt violation detection is done manually or partial manual. In this paper, an approach is proposed to make the seat belt detection process fully automated. To make the detection more accurate, sensors are set to detect the weather condition. When specific weather condition is detected, the corresponding pre-trained model is assigned the detection task. In other words, a research is conducted to check the possibility of dividing the big-sized deep-learning model-that can classify car seatbelt, into sub-models each one can detect specific weather condition. Accordingly, a single specialized model is used for each weather condition, Deep convolutional neural network (CNN) model AlexNet is used in the detection/classification process. The proposed system is sensor based AlexNet (S-AlexNet). Results support our hypothesis that "Using single model for each weather condition is better than general model that support all weather conditions". On average, previous approaches that trained single model for all weather conditions have accuracy less than 90%. The proposed S-AlexNet approach successfully reaches 90+% accuracy.
