Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Analysis of Factors Contributing to the Severity of Large Truck Crashes.

In Entropy (Basel, Switzerland)

Crashes that involved large trucks often result in immense human, economic, and social losses. To prevent and mitigate severe large truck crashes, factors contributing to the severity of these crashes need to be identified before appropriate countermeasures can be explored. In this research, we applied three tree-based machine learning (ML) techniques, i.e., random forest (RF), gradient boost decision tree (GBDT), and adaptive boosting (AdaBoost), to analyze the factors contributing to the severity of large truck crashes. Besides, a mixed logit model was developed as a baseline model to compare with the factors identified by the ML models. The analysis was performed based on the crash data collected from the Texas Crash Records Information System (CRIS) from 2011 to 2015. The results of this research demonstrated that the GBDT model outperforms other ML methods in terms of its prediction accuracy and its capability in identifying more contributing factors that were also identified by the mixed logit model as significant factors. Besides, the GBDT method can effectively identify both categorical and numerical factors, and the directions and magnitudes of the impacts of the factors identified by the GBDT model are all reasonable and explainable. Among the identified factors, driving under the influence of drugs, alcohol, and fatigue are the most important factors contributing to the severity of large truck crashes. In addition, the exists of curbs and medians and lanes and shoulders with sufficient width can prevent severe large truck crashes.

Li Jinhong, Liu Jinli, Liu Pengfei, Qi Yi

2020-Oct-22

AK level crashes, AdaBoost, contributing factors, gradient boost decision tree, injury severity, large truck crash, mixed logit model, random forest

General General

An Appraisal of Incremental Learning Methods.

In Entropy (Basel, Switzerland)

As a special case of machine learning, incremental learning can acquire useful knowledge from incoming data continuously while it does not need to access the original data. It is expected to have the ability of memorization and it is regarded as one of the ultimate goals of artificial intelligence technology. However, incremental learning remains a long term challenge. Modern deep neural network models achieve outstanding performance on stationary data distributions with batch training. This restriction leads to catastrophic forgetting for incremental learning scenarios since the distribution of incoming data is unknown and has a highly different probability from the old data. Therefore, a model must be both plastic to acquire new knowledge and stable to consolidate existing knowledge. This review aims to draw a systematic review of the state of the art of incremental learning methods. Published reports are selected from Web of Science, IEEEXplore, and DBLP databases up to May 2020. Each paper is reviewed according to the types: architectural strategy, regularization strategy and rehearsal and pseudo-rehearsal strategy. We compare and discuss different methods. Moreover, the development trend and research focus are given. It is concluded that incremental learning is still a hot research area and will be for a long period. More attention should be paid to the exploration of both biological systems and computational models.

Luo Yong, Yin Liancheng, Bai Wenchao, Mao Keming

2020-Oct-22

catastrophic forgetting, incremental learning, lifelong learning

General General

Enhanced Deep Learning Architectures for Face Liveness Detection for Static and Video Sequences.

In Entropy (Basel, Switzerland)

Face liveness detection is a critical preprocessing step in face recognition for avoiding face spoofing attacks, where an impostor can impersonate a valid user for authentication. While considerable research has been recently done in improving the accuracy of face liveness detection, the best current approaches use a two-step process of first applying non-linear anisotropic diffusion to the incoming image and then using a deep network for final liveness decision. Such an approach is not viable for real-time face liveness detection. We develop two end-to-end real-time solutions where nonlinear anisotropic diffusion based on an additive operator splitting scheme is first applied to an incoming static image, which enhances the edges and surface texture, and preserves the boundary locations in the real image. The diffused image is then forwarded to a pre-trained Specialized Convolutional Neural Network (SCNN) and the Inception network version 4, which identify the complex and deep features for face liveness classification. We evaluate the performance of our integrated approach using the SCNN and Inception v4 on the Replay-Attack dataset and Replay-Mobile dataset. The entire architecture is created in such a manner that, once trained, the face liveness detection can be accomplished in real-time. We achieve promising results of 96.03% and 96.21% face liveness detection accuracy with the SCNN, and 94.77% and 95.53% accuracy with the Inception v4, on the Replay-Attack, and Replay-Mobile datasets, respectively. We also develop a novel deep architecture for face liveness detection on video frames that uses the diffusion of images followed by a deep Convolutional Neural Network (CNN) and a Long Short-Term Memory (LSTM) to classify the video sequence as real or fake. Even though the use of CNN followed by LSTM is not new, combining it with diffusion (that has proven to be the best approach for single image liveness detection) is novel. Performance evaluation of our architecture on the REPLAY-ATTACK dataset gave 98.71% test accuracy and 2.77% Half Total Error Rate (HTER), and on the REPLAY-MOBILE dataset gave 95.41% accuracy and 5.28% HTER.

Koshy Ranjana, Mahmood Ausif

2020-Oct-21

CNN-LSTM, Inception v4, Replay-Attack dataset, Replay-Mobile dataset, SCNN, diffusion, face liveness detection

General General

Salient Object Detection Techniques in Computer Vision-A Survey.

In Entropy (Basel, Switzerland)

Detection and localization of regions of images that attract immediate human visual attention is currently an intensive area of research in computer vision. The capability of automatic identification and segmentation of such salient image regions has immediate consequences for applications in the field of computer vision, computer graphics, and multimedia. A large number of salient object detection (SOD) methods have been devised to effectively mimic the capability of the human visual system to detect the salient regions in images. These methods can be broadly categorized into two categories based on their feature engineering mechanism: conventional or deep learning-based. In this survey, most of the influential advances in image-based SOD from both conventional as well as deep learning-based categories have been reviewed in detail. Relevant saliency modeling trends with key issues, core techniques, and the scope for future research work have been discussed in the context of difficulties often faced in salient object detection. Results are presented for various challenging cases for some large-scale public datasets. Different metrics considered for assessment of the performance of state-of-the-art salient object detection models are also covered. Some future directions for SOD are presented towards end.

Gupta Ashish Kumar, Seal Ayan, Prasad Mukesh, Khanna Pritee

2020-Oct-19

deep learning-based salient object detection models, saliency cues, conventional salient object detection models, salient object detection

General General

Active Learning for Node Classification: An Evaluation.

In Entropy (Basel, Switzerland)

Current breakthroughs in the field of machine learning are fueled by the deployment of deep neural network models. Deep neural networks models are notorious for their dependence on large amounts of labeled data for training them. Active learning is being used as a solution to train classification models with less labeled instances by selecting only the most informative instances for labeling. This is especially important when the labeled data are scarce or the labeling process is expensive. In this paper, we study the application of active learning on attributed graphs. In this setting, the data instances are represented as nodes of an attributed graph. Graph neural networks achieve the current state-of-the-art classification performance on attributed graphs. The performance of graph neural networks relies on the careful tuning of their hyperparameters, usually performed using a validation set, an additional set of labeled instances. In label scarce problems, it is realistic to use all labeled instances for training the model. In this setting, we perform a fair comparison of the existing active learning algorithms proposed for graph neural networks as well as other data types such as images and text. With empirical results, we demonstrate that state-of-the-art active learning algorithms designed for other data types do not perform well on graph-structured data. We study the problem within the framework of the exploration-vs.-exploitation trade-off and propose a new count-based exploration term. With empirical evidence on multiple benchmark graphs, we highlight the importance of complementing uncertainty-based active learning models with an exploration term.

Madhawa Kaushalya, Murata Tsuyoshi

2020-Oct-16

active learning, graph neural networks, graph representation learning, machine learning, node classification

Surgery Surgery

Pathological lung segmentation in chest CT images based on improved random walker.

In Computer methods and programs in biomedicine

BACKGROUND AND OBJECTIVE : Pathological lung segmentation as a pretreatment step in the diagnosis of lung diseases has been widely explored. Because of the complexity of pathological lung structures and gray blur of the border, accurate lung segmentation in clinical 3D computed tomography images is a challenging task. In view of the current situation, the work proposes a fast and accurate pathological lung segmentation method. The following contributions have been made: First, the edge weights introduce spatial information and clustering information, so that walkers can use more image information during walking. Second, a Gaussian Distribution of seed point set is established to further expand the possibility of selection between fake seed points and real seed points. Finally, the pre-parameter is calculated using original seed points, and the final results are fitted with new seed points.

METHODS : This study proposes a segmentation method based on an improved random walker algorithm. The proposed method consists of the following steps: First, a gray value is used as the sample distribution. Gaussian mixture model is used to obtain the clustering probability of an image. Thus, the spatial distance and clustering result are added as new weights, and the new edge weights are used to construct a random walker map. Second, a large number of marked points are automatically selected, and the intermediate results are obtained from the newly constructed map and retained only as pre-parameters. When new seed points are introduced, the probability value of the walker is quickly calculated from the new parameters and pre-parameters, and the final segmentation result can be obtained.

RESULTS : The proposed method was tested on 65 sets of CT cases. Quantitative evaluation with different methods confirms the high accuracy on our dataset (98.55%) and LOLA11 dataset (97.41%). Similarly, the average segmentation time (10.5s) is faster than random walker (1,332.5s).

CONCLUSIONS : The comparison of the experimental results show that the proposed method can accurately and quickly obtain pathological lung processing results. Therefore, it has potential clinical applications.

Chen Cheng, Xiao Ruoxiu, Zhang Tao, Lu Yuanyuan, Guo Xiaoyu, Wang Jiayu, Chen Hongyu, Wang Zhiliang

2020-Nov-25

Binary K-means, Gaussian mixture model, Lung segmentation, Random walker