Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

The analysis of travel mode choice is important in transportation planning and policy-making in order to understand and forecast travel demands Research in the field of machine learning has been exploring the use of random forest as a framework within which many traffic and transport problems can be investigated The random forest (RF) is a powerful method for constructing an ensemble of random decision trees It de-correlates the decision trees in the ensemble via randomization that leads to an improvement of forecasting and reduces the variance when averaged over the trees However, the usefulness of RF for travel mode choice behavior remains largely unexplored This paper proposes a robust random forest method to analyze travel mode choices for examining the prediction capability and model interpretability Using the travel diary data from Nanjing, China in 2013, enriched with variables on the built environment, the effects of different model parameters on the prediction performance are investigated The comparison results show that the random forest method performs significantly better in travel mode choice prediction for higher accuracy and less computation cost In addition, the proposed method estimates the relative importance of explanatory variables and how they relate to mode choices This is fundamental for a better understanding and effective modeling of people’s travel behavior

Applying a random forest method approach to model travel mode choice behavior

Floods are one of nature's most destructive disasters because of the immense damage to land, buildings, and human fatalities. It is difficult to forecast the areas that are vulnerable to flash flooding due to the dynamic and complex nature of the flash floods. Therefore, earlier identification of flash flood susceptible sites can be performed using advanced machine learning models for managing flood disasters. In this study, we applied and assessed two new hybrid ensemble models, namely Dagging and Random Subspace (RS) coupled with Artificial Neural Network (ANN), Random Forest (RF), and Support Vector Machine (SVM) which are the other three state-of-the-art machine learning models for modelling flood susceptibility maps at the Teesta River basin, the northern region of Bangladesh. The application of these models includes twelve flood influencing factors with 413 current and former flooding points, which were transferred in a GIS environment. The information gain ratio, the multicollinearity diagnostics tests were employed to determine the association between the occurrences and flood influential factors. For the validation and the comparison of these models, for the ability to predict the statistical appraisal measures such as Freidman, Wilcoxon signed-rank, and t-paired tests and Receiver Operating Characteristic Curve (ROC) were employed. The value of the Area Under the Curve (AUC) of ROC was above 0.80 for all models. For flood susceptibility modelling, the Dagging model performs superior, followed by RF, the ANN, the SVM, and the RS, then the several benchmark models. The approach and solution-oriented outcomes outlined in this paper will assist state and local authorities as well as policy makers in reducing flood-related threats and will also assist in the implementation of effective mitigation strategies to mitigate future damage.

Flood susceptibility modelling using advanced ensemble machine learning models

The World Health Organization (WHO) reported 1.25 million deaths yearly due to road traffic accidents worldwide and the number has been continuously increasing over the last few years. Nearly fifth of these accidents are caused by distracted drivers. Existing work of distracted driver detection is concerned with a small set of distractions (mostly, cell phone usage). Unreliable ad hoc methods are often used. In this paper, we present the first publicly available dataset for driver distraction identification with more distraction postures than existing alternatives. In addition, we propose a reliable deep learning-based solution that achieves a 90% accuracy. The system consists of a genetically weighted ensemble of convolutional neural networks; we show that a weighted ensemble of classifiers using a genetic algorithm yields a better classification confidence. We also study the effect of different visual elements in distraction detection by means of face and hand localizations, and skin segmentation. Finally, we present a thinned version of our ensemble that could achieve 84.64% classification accuracy and operate in a real-time environment.

/pdf/driver-distraction-identification-with-an-ensemble-of-21wy60fb3l.pdf

Driver Distraction Identification with an Ensemble of Convolutional Neural Networks

Traffic surveillance has become an important topic in intelligent transportation systems (ITSs), which is aimed at monitoring and managing traffic flow. With the progress in computer vision, video-based surveillance systems have made great advances on traffic surveillance in ITSs. However, the performance of most existing surveillance systems is susceptible to challenging complex traffic scenes (e.g., object occlusion, pose variation, and cluttered background). Moreover, existing related research is mainly on a single video sensor node, which is incapable of addressing the surveillance of traffic road networks. Accordingly, we present a review of the literature on the video-based vehicle surveillance systems in ITSs. We analyze the existing challenges in video-based surveillance systems for the vehicle and present a general architecture for video surveillance systems, i.e., the hierarchical and networked vehicle surveillance, to survey the different existing and potential techniques. Then, different methods are reviewed and discussed with respect to each module. Applications and future developments are discussed to provide future needs of ITS services.

/pdf/hierarchical-and-networked-vehicle-surveillance-in-its-a-2dq9vsx0r3.pdf

Hierarchical and Networked Vehicle Surveillance in ITS: A Survey

An efficient feature extraction approach for driving postures from a video camera, which consists of Homomorphic filtering, skin-like regions segmentation and contourlet transform (CT), was proposed. With features extracted from a driving posture dataset created at Southeast University (SEU), holdout and cross-validation experiments on driving posture classification were then conducted using random forests (RF) classifier. Compared with a number of commonly used classification methods including linear perceptron classifier, k -nearest-neighbour classifier and multilayer perceptron (MLP) classifier, the experiments results showed that the RF classifier offers the best classification performance among the four classifiers. Among the four predefined classes, that is, grasping the steering wheel, operating the shift gear, eating and talking on a cellular phone, the class of eating is the most difficult to classify. With RF classifier, the classification accuracies of eating are over 88% in holdout and cross-validation experiments, thus demonstrating the effectiveness of the proposed feature extraction method and the importance of RF classifier in automatically understanding and characterising driver%s behaviours towards human-centric driver assistance systems.

Recognition of driving postures by contourlet transform and random forests

Aiming to efficiently resolve the problem that the subtraction background model is easily contaminated by slow-moving or temporarily stopped vehicles, the Gaussian mixture model with confidence measurement (GMMCM) is proposed for vehicle detection in complex urban traffic scenes. According to the current traffic state, each pixel of background model is set a CM. Whether to update the background model and the corresponding adaptive learning rate depends on if the current pixel point is in confidence period. Using the real-world urban traffic videos, the first experiments are conducted by GMMCM, compared with three commonly used models including GMM, self-adaptive GMM (SAGMM) and local parameter learning algorithm for the GMM (LPLGMM). The first experimental results show that GMMCM excels GMM, SAGMM and LPLGMM in keeping the background model being unpolluted from slow-moving or temporarily stopped vehicles. The second experiments are conducted by GMMCM, compared with visual background extractor, sigma-delta with CM, SAGMM, LPLGMM and GMM. The average recalls of six methods are 0.899, 0.753, 0.679, 0.420, 0.447 and 0.205, and the average F-measures of six methods are 0.636, 0.612, 0.592, 0.373, 0.330 and 0.179, respectively. All experimental results show the effectiveness of the proposed GMMCM in vehicles detection of complex urban traffic scenes.

Vehicles detection in complex urban traffic scenes using Gaussian mixture model with confidence measurement

Reports of traffic accidents show that a considerable percentage of the accidents are caused by human factors. Human-centric driver assistance systems, with integrated sensing, processing and networking, aim to find solutions to this problem and other relevant issues. The key technology in such systems is the capability to automatically understand and characterize driver behaviors. In this paper, we propose a novel, efficient feature extraction approach for driving postures from a video camera, which consists of Homomorphic filter, skin-like regions segmentation, canny edge detection, connected regions detection, small connected regions deletion and spatial scale ratio calculation. With features extracted from a driving posture dataset we created at Southeast University (SEU), holdout and cross-validation experiments on driving posture classification are then conducted using Bayes classifier. Compared with a number of commonly used classification methods including naive Bayes classifier, subspace classifier, linear perception classifier and Parzen classifier, the holdout and cross-validation experiments show that the Bayes classifier offers better classification performance than the other four classifiers. Among the four predefined classes, i.e., grasping the steering wheel, operating the shift gear, eating a cake and talking on a cellular phone, the class of talking on a cellular phone is the most difficult to classify. With Bayes classifier, the classification accuracies of talking on a cellular phone are over 90 % in holdout and cross-validation experiments, which shows the effectiveness of the proposed feature extraction method and the importance of Bayes classifier in automatically understanding and characterizing driver behaviors towards human-centric driver assistance systems.

Vision-based Classification of Driving Postures by Efficient Feature Extraction and Bayesian Approach

The objective of this study is to investigate different pattern classification paradigms in the automatically understanding and characterizing driver behaviors. With features extracted from a driving posture dataset consisting of grasping the steering wheel, operating the shift lever, eating a cake and talking on a cellular phone, created at Southeast University, holdout and cross-validation experiments on driving posture classification are firstly conducted using Support Vector Machines (SVMs) with five different kernels, and then comparatively conducted with other four commonly used classification methods including linear perception classifier, k-nearest neighbor classifier, Multi-layer perception classifier, and parzen classifier. The holdout experiments show that the intersection kernel outperforms the other four kernels, and the SVMs with intersection kernel offers better classification rates and best real-time quality among the five classifiers, which shows the effectiveness of the proposed feature extraction method and the importance of SVM classifier in automatically understanding and characterizing driver behaviors towards human-centric driver assistance systems.

Classification of Driving Postures by Support Vector Machines

The identification of the make of vehicles is a challenge task. In this paper, we proposed to combine two different features, i.e., Pyramid Histogram of Oriented Gradients (PHOG) and Curve let transform, to describe vehicle images. The Curve let transform has the feature of higher time frequency resolution and high degree of directionality and anisotropy, which is particularly appropriate for images rich with edges. PHOG represents the local shape by a histogram of edge orientations computed for each image sub-region, quantized into a number of bins. Compared with previously proposed feature extraction approaches in vehicle recognition, PHOG has advantages in the extraction of discriminating information. A composite fetaure description from PHOG and Curve let Transform can further increase the accuracy of classification by taking their complementary information. We then investigate the applicability of the Random Subspace (RS) ensemble method for vehicle classification based on the combined features. A base classifier is trained with a randomly sampled subset of the original feature set and the ensemble assigns a class label by majority voting. Experimental results using more than 600 images from 21 makes show the effectiveness of the proposed approach. The composite feature is better than any single feature in the classification accuracy and the ensemble model produces better performance compared to any of the individual neural network base classifier. With moderate ensemble size 30, the Random Subspace ensembles offers a classification rate close to 96%, showing the promising potential in real applications.

Chihang Zhao

Papers

Recognition of driving postures by contourlet transform and random forests

Vehicles detection in complex urban traffic scenes using Gaussian mixture model with confidence measurement

Vision-based Classification of Driving Postures by Efficient Feature Extraction and Bayesian Approach

Classification of Driving Postures by Support Vector Machines

Classification of Vehicle Make by Combined Features and Random Subspace Ensemble