It is of great significance for safe driving to study drivers' eye movement and driving operation behavior when they encounter other road users violating traffic rules. The underlying reason is that ...most drivers are unable to process the unexpected visual stimulation, which is more likely to lead to driving accidents, especially in a hybrid situation. In this study, a driving simulator is used to design driving scenarios and study the driving performance of drivers with different driving experiences when other road users violate traffic rules. The experimental results show that some novice drivers ignore the position of their own vehicle when they encounter traffic violations which will lead to the collision with other road users. Moreover, some novice drivers can only perform one of the operations between steering and braking to avoid collision in these emergent situations. They cannot reasonably combine braking and steering to complete emergency driving operations like experienced drivers. Finally, when the driving difficulty increases, experienced drivers spend less time looking and more time scanning their surroundings to ensure that they can cope with the more complex driving environment while novice drivers do the opposite. This study reveals the difference between novices and experienced drivers, which paves a useful reference for the future advanced driving assistance system.
Electroencephalogram (EEG) has been widely utilized in emotion recognition due to its high temporal resolution and reliability. However, the individual differences and non-stationary characteristics ...of EEG, along with the complexity and variability of emotions, pose challenges in generalizing emotion recognition models across subjects. In this paper, an end-to-end framework is proposed to improve the performance of cross-subject emotion recognition. A novel evolutionary programming (EP)-based optimization strategy with neural network (NN) as the base classifier termed NN ensemble with EP (EPNNE) is designed for cross-subject emotion recognition. The effectiveness of the proposed method is evaluated on the publicly available DEAP, FACED, SEED, and SEED-IV datasets. Numerical results demonstrate that the proposed method is superior to state-of-the-art cross-subject emotion recognition methods. The proposed end-to-end framework for cross-subject emotion recognition aids biomedical researchers in effectively assessing individual emotional states, thereby enabling efficient treatment and interventions.
Optical character recognition (OCR) technology is promoting the process of automation in daily information recording and inspection of industrial meters. However, bad installation scenes and ...long-time running cause industrial meters to age to varying degrees, which increases the difficulty of OCR. Besides, in practice, operators typically use handheld cameras to extract tag information from industrial meters. During the shooting process, extreme lighting, free shooting angles, and shooting distance also cause many difficulties for OCR. Considering such difficulties, an end-to-end recognition architecture is developed to obtain a better OCR performance. The proposed architecture can quickly extract structured information from normal or skewed text images. A novel yolov5_adaloss with a specific penalty factor is designed to alleviate the influence of illumination, installation scene, age, skew, and distance on the classification accuracy of Tags. For images with large skew angles, a mathematical method is proposed to calculate the text inclination angle for better OCR performance. The contribution of this work is twofold. Firstly, the architecture proposed is lightweight, which only needs a low computing cost and a short inference time. Secondly, as a practical application-oriented architecture, this work does not require much training, labeling, and fine-tuning, which is easy to generalize to other structured text recognition tasks. Experiments show that the method proposed in this paper can achieve excellent performance in meter tag recognition tasks on actual industrial images of meters.
Multiple-View Stereo (MVS) aims to reconstruct the dense 3D representations of scenes. MVS has potential applications in the fields of autonomous driving (unstructured environment construction) and ...robotic navigation (visual-inertial navigation). To mitigate the error of depth estimation in low-textured or occluded regions, this work proposes a two-stage multi-view stereo network for fast and accurate depth estimation. The improvements of this work over the state of the art are as follows: 1) Sparse costs are constructed to jointly predict the initial depth map and surface normal by cost regularization, which proves that the surface normals can be estimated in this way with low memory consumption. 2) A new edge refinement block is developed to refine the coarse surface normal to obtain a fine-grained surface normal map. 3) Instead of using the general variance-based metric to equally aggregate cost, a new content-adaptive cost aggregation mechanism based on the similarity of the neighboring surface normal is designed for reliable cost aggregation. To the best of our knowledge, the proposed work is the first trainable network that leverages surface normal as guidance to capture neighboring pixel-visibility, which is an effective supplement to existing depth/normal estimation frameworks. Experimental results indicate that our method can not only achieve accurate depth estimation for scene perception but also make no concession to the real-time performance and limited memory bottleblock. Multiple-view stereo (MVS) aims to reconstruct the dense 3D representations of scenes. It is widely used in the fields of industrial measurement, autonomous driving, and robotic navigation. To mitigate the error of depth estimation in challenging scenarios, this work proposes a two-stage multi-view stereo network for fast and accurate depth estimation. Our method is the first trainable network that leverages surface normal as pixel-visibility guidance to aggregate reliable cost, which could achieve accurate depth estimation and provide the perception ability for the robot. The proposed method has great potential in the fields of 3D reconstruction, industrial measurement, and robotic navigation to estimate real-time and accurate depth with limited memory consumption.
The application of automation technology and artificial intelligence technology has promoted the improvement of the business capabilities of enterprises in industrial scenarios. Compared with the ...improvement or innovation of the business process, in recent years, part of academic research and practical applications has also shifted their attention from a single point of business intelligence perspective to a comprehensive intelligent upgrade of the industrial system. To the best of our knowledge, however, there is little research on the concept and model of the industrial intelligent system (IIS). To make up for the lack, this article presents the concept and reference model of IIS by analyzing the intelligentization requirement of the industrial systems. Different from academic research on general intelligent system capabilities, the reference model given emphasizes factors that need to be considered when implementing IIS in the industry. By analyzing the reference model, knowledge as the core driving force of IIS is recognized. Then, the four main forms of knowledge in IIS, as well as the role and key technologies of knowledge in different stages of IIS, are discussed in detail. In addition, several important potential applications of IIS are pointed out in this article.
Dynamic, accurate, and predictive conflict detection is crucial for ensuring collision-free movement and efficient route planning for Automated Guided Vehicles (AGVs). Timely alarms based on ...detection enable AGVs to efficiently plan collision-free routes. However, existing conflict detection methods typically do not consider the above elements simultaneously, either making assumptions that reduce the accuracy of applications, such as assuming constant vehicle speed (value of velocity), or focusing on real-time physical collisions but ignoring the future vehicle moving state. In this paper, we propose a dynamic AGV conflict detection scheme with time windows to detect possible AGV conflicts timely. The proposed scheme fully considers the accuracy, dynamics, and predictability requirements of detection, which encompasses the following three components: preliminary conflict detection, potential motion trajectory generation, and conflict detection and alarm. In particular, for conflict detection and alarm where potential conflicts trigger corresponding alarm reminders, we have observed that the AGV conflict occurrence conditions within a specific time window can be mathematically expressed as a binary quadratic inequality, then conflict alarms between two AGVs can be efficiently calculated. Our proposed scheme could help dynamically and accurately detect potential AGV conflicts within the planned time window, and the speed fluctuation of the AGV is also fully considered. Experiment results on several sets of simulation scenarios prove the applicability and effectiveness of our scheme.
With the deepening exploration of Industry 4.0, smart factories are gradually replacing traditional factories with rapid momentum. In smart factories, a large number of digitally networked devices ...are deployed in a less-populated or even unmanned environment. Data security and fast access have become particularly important due to the automation and intelligence of production. As the environment of smart factories becomes increasingly complex, meeting the requirements for rapid authentication has become increasingly difficult for traditional authentication systems. In this study, a lightweight blockchain-based radio-frequency identification (RFID) identity authentication mechanism is proposed for smart factories represented by the medical device manufacturing industry by integrating blockchain and RFID technologies. Through bitwise operations, cyclic shift operation, and hash arithmetic, the proposed mechanism cannot only guarantee security between the RFID reader and the electronic tag but also requires less communication and storage to complete authentication. Thus, this mechanism is suitable for the environment of medical device manufacturing factories with a high-load operation of equipment. It helps further research on the data security of smart factories.
The rapid developments of mobile robotics and autonomous navigation over the years are largely empowered by public datasets for testing and upgrading, such as sensor odometry and SLAM tasks. ...Impressive demos and benchmark scores have arisen, which may suggest the maturity of existing navigation techniques. However, these results are primarily based on moderate structured scenario testing. When transitioning to challenging unstructured environments, especially in GNSS-denied, texture-monotonous, and dense-vegetated natural fields, their performance can hardly sustain at a high level and requires further validation and improvement. To bridge this gap, we build a novel robot navigation dataset in a luxuriant botanic garden of more than 48000 m 2 . Comprehensive sensors are used, including Gray and RGB stereo cameras, spinning and MEMS 3D LiDARs, and low-cost and industrial-grade IMUs, all of which are well calibrated and hardware-synchronized. An all-terrain wheeled robot is employed for data collection, traversing through thick woods, riversides, narrow trails, bridges, and grasslands, which are scarce in previous resources. This yields 33 short and long sequences, forming 17.1 km trajectories in total. Excitedly, both highly-accurate ego-motions and 3D map ground truth are provided, along with fine-annotated vision semantics. We firmly believe that our dataset can advance robot navigation and sensor fusion research to a higher level.
Imbalanced data is a major challenge in classification tasks. Most classification algorithms tend to be biased toward the samples in the majority class but fail to classify the samples in the ...minority class. Recently, ensemble learning, as a promising method, has been rapidly developed in solving highly imbalanced classification. However, the design of the base classifier for the ensemble is still an open question because the optimization problem of the base classifier is gradientless. In this study, the evolutionary algorithm (EA) technique is adopted to solve a wide range of optimization design problems in highly imbalanced classification without gradient information. A novel EA-based classifier optimization design method is proposed to optimize the design of multiple base classifiers automatically for the ensemble. In particular, an EA method with a neural network (NN) as the base classifier termed NN ensemble with EA (NNEAE) is developed for highly imbalanced classification. To verify the performance of NNEAE, extensive experiments are designed for testing. Results illustrate that NNEAE outperforms other compared methods.
Driver identification has been popular in the field of driving behavior analysis, which has a broad range of applications in anti-thief, driving style recognition, insurance strategy, and fleet ...management. However, most studies to date have only researched driver identification without a robust verification stage. This paper addresses driver identification and verification through a deep learning (DL) approach using psychological behavioral data, i.e., vehicle control operation data and eye movement data collected from a driving simulator and an eye tracker, respectively. We design an architecture that analyzes the segmentation windows of three-second data to capture unique driving characteristics and then differentiate drivers on that basis. The proposed model includes a fully convolutional network (FCN) and a squeeze-and-excitation (SE) block. Experimental results were obtained from 24 human participants driving in 12 different scenarios. The proposed driver identification system achieves an accuracy of 99.60% out of 15 drivers. To tackle driver verification, we combine the proposed architecture and a Siamese neural network, and then map all behavioral data into two embedding layers for similarity computation. The identification system achieves significant performance with average precision of 96.91%, recall of 95.80%, F1 score of 96.29%, and accuracy of 96.39%, respectively. Importantly, we scale out the verification system to imposter detection and achieve an average verification accuracy of 90.91%. These results imply the invariable characteristics from human factors rather than other traditional resources, which provides a superior solution for driving behavior authentication systems.
