Ebook: Information Technology and Intelligent Transportation Systems

Intelligent transport systems, from basic management systems to more application-oriented systems, vary in the technologies they apply. Information technologies, including wireless communication, are important in intelligent transportation systems, as are computational technologies: floating car data/floating cellular data, sensing technologies, and video vehicle detection. Theoretical and application technologies, such as emergency vehicle notification systems, automatic road enforcement and collision avoidance systems, as well as some cooperative systems are also used in intelligent transportation systems.

This book presents papers selected from the 128 submissions in the field of information technology and intelligent transportation systems received from 5 countries. In December 2019 ChangâĂŹan University organized a round-table meeting to discuss and score the technical merits of each selected paper, of which 23 are included in this book. The meeting was also co-sponsored by XiâĂŹan University of Technology, Northwestern Polytechnical University, CAS, Shaanxi Sirui Advanced Materials Co., LTD and Special Aircraft Engineering Research Institute.

We are grateful to all contributors for their efforts in preparing their manuscripts in a timely manner. We would also like to express our appreciation to the reviewers for their support, and gratefully acknowledge their help in bringing out this volume on time.

Lakhmi C. Jain

Xiangmo Zhao

Valentina Emilia Balas

Fuqian Shi

Combinatorial test (CT) is a commonly used method to conduct comprehensive and efficient test. To further improve test efficiency, based on CT, this paper proposes an improved combinatorial test (ICT) method for model-in-the-loop (MIL) test of the autonomous parallel parking system (APPS). Different from CT, in the process to generate test cases, the ICT takes not only combination coverage but also importance degree into consideration. When selecting parameter value of the test case, the ICT tends to select the value with higher importance degree, on the premise of the combination coverage and the number of test cases remaining unchanged. Among the different values of the same parameter, these more likely to lead to bad parking performance are assigned higher importance degrees using Analytic Hierarchy Process (AHP). The experiment results show that, overall, for different values of the same parameter, the values with higher importance degree appear more frequently in the ICT test suite. Besides, compared with CT, the ICT further improves the test efficiency.

Urban rail transit accounts for an increasing proportion of the travel structure and becomes the backbone of the city. At the same time, in order to make full use of the efficient and convenient rail transit, how other modes of transportation such as buses, cars, and bicycles form an effective multimodal transport mode has become a challenge for current traffic planners, managers and transportation travel participants. At this stage, the rapid development of urban shared bicycles provides favorable conditions for commuters to use bicycles to connect rail transit. This paper establishes the distance attenuation model and the non-aggregate price sensitivity measurement method under the influence of winter traffic restriction measures by studying the significant passenger flow using the bicycle to connect the rail transit of the Xi’an metro line 2 terminal Weiqu South Station. The RP and SP traffic survey data were used to analyze the impact of the restricted measures on the spatio-temporal threshold determination of the bicycle-connected rail transit, which provided a reference for the planners to formulate relevant plans for bicycle-connected rail transit.

In this paper, per capita area is used to describe service level. An AnyLogic simulation system is formulated to study on subway design parameters, which influence the service level of transfer stations. For stations in the case, it is suggested that the platform width in different transfer stations should be different. According to the simulation, the platform width of “One”, “Cross” and “T” type transfer stations should be widened by 3.6m, 1.2m and 2.4m respectively on the basis of the original models. As a result, the service level of the platform at the most crowded moment can be upgraded from B to A, and the increase of the minimum per capita area of the platform is 34.9%, 15.8% and 21.2%. Besides, the width of stair should be widened by 0.4m, 0.8m and 0.2m, so that the service level of the platform can reach the peak, and the average per capita area of the platform can reach 1.754 m2/p (square meter per person), 2.000 m2/p, 2.178 m2/p. Finally, if conditions permit, the stairs should be placed at both ends of the platform.

Visual Traffic Surveillance has become a significant framework in vision computing with the advent of intelligent transport system in recent years. The Automatic Vehicle Surveillance system is gaining growing attention to handle traffic congestion in urban roadways. The primary task of such system is to detect any moving object in the video sequence, robustly track and finally classify into different categories. Motion Segmentation plays an important role among the methods of vehicle detection and are briefly discussed in this study. After detection, the surveillance system should be capable of tracking and classifying the vehicles. There are several conventional tracking techniques for vehicle tracking which are explained with its related researches, advantages and limitations in this study. The surveillance becomes more challenging due to the presence of occlusion, variance in illumination, unusual orientation of vehicles etc. Vehicle detection and tracking has an emerging application in real time passenger management system, security issues and accident prevention in highways. To prevent accidents, abnormal behaviour of the vehicles should be detected in advance. For the achievement of the unusual behaviour detection, a comprehensive analysis of different methods of anomaly detection are highlighted in this survey. The challenges, future scope with notable application domains of the system are located and a dominant bibliography is also included.

This study conducts investigations on the basic data of urban rail transit station energy consumption, analyzes the composition of urban rail transit station energy consumption, grasps the energy consumption differences of different types of stations, and establishes a tree structure evaluation system of energy consumption quality for metro stations. The system reflects the energy quality level of urban rail transit stations. Several stations of the Beijing Subway were taken as cases to analyze the energy quality characteristics of different types of stations under the evaluation system.

Turn-back ability of the intermediate station is a key factor to determine the carrying capacity of the rail line in the multi-routing mode of urban rail transit. In this paper, we analyze the calculation principle of the turn-back ability of the intermediate station, and the calculation model of the minimum turn-back interval time is established. Taking 2.5 minutes as the train headway time, analyze the limiting factors of the intermediate turn-back station under the condition of the ratio of the short routing train and long routing train is 1:1, and use the simulation method to measure the relevant influencing factors. Studies have shown that the line-front turning-back stations with single cross line has restrictions on the capacity of the rail line, while the line-behind station reentry has no limit. When the interval is less than 2.5 minutes, the line-behind station with double turn-back line can be used to meet the turn-back ability requirement. In addition, the speed of arrival the station, the length of train, the position of enter the station and the train dwelling time at station have a certain impact on the turn-back ability. The train dwelling time at station and the speed of arrival the station has a great influence on the turn-back ability. When speed arriving at the station is less than 60 km/h, the increase of speed improves turn-back ability obviously.

The traditional way of providing security is not enough for providing authentication over a large population. In this era of digital advancements and artificial intelligence, biometric security systems are transforming such security problems across the world. The characteristics of the iris pattern are extraordinarily unique. Hence it is extremely favored among other biometric modalities. In the iris image based biometric system, each image of the iris pattern is transformed into a set of distinct features by the process called feature extraction. It is one of the key steps in any recognition system. In this paper, statistical features are extracted from different domains like the histogram of intensity level, local binary pattern (LBP), histogram of oriented gradients (HoG), Eigenspace and moments of the iris image. A comparison of these different feature extraction methods for iris biometric system is discussed using minimum distance classifier. The experimental results show moment based feature extraction performs better than other feature extraction methods.

Emotion recognition is one of the better way to recognize state of mind of humans, there are several emotions such as Happy, Neutral, Disgust, Sad, Anger, Surprise, and Fear which are known as universal emotions. In this paper we are trying to extract different facial features using some well-known techniques such as Local Binary Patterns (LBP), Histogram of Gradients (HOG), Scale Invariant Feature Transform (SIFT) and Speeded-up Robust Features(SURF). These techniques are tested on the most popular Japanese Female Facial Expression (JAFFE) database. These Database have frontal view face images.

Now-a-days augmented reality (AR) has become an interactive and collaborative tool for educational applications. AR makes the teaching and learning process more effective and allows the user to access the virtual objects in the real world. Although AR technology enhances the educational outcome, psychological, and pedagogical aspect for real-time usage in the classroom, the most important problem is the lighting conditions on which the output clarity is dependent. They can be affected by reflections from overhead lights and the light from the windows. Omni directional lighting can be used to avoid this problem. If the marker is moved out of the camera view, then the recognition will fail. Hence, we are using markers with large black and white regions that are low-frequency patterns. An augmented reality application for mathematics and geometry in school level education system is discussed in this paper.

Skin malignancy is a catastrophic health problems witnessed in Europeans and western area of the world because of the changes in the ozone layer. Ultraviolet (UV) rays common threats for the human health. Scientists have studied on Computer-Aided Diagnosis (CAD) scheme to ease interpretation detection of melanoma. There are several variations of features of the lesions and different Artificial Intelligence (AI) based design participates in an essential role for building CAD system. This study has refined skin lesions with diffusion and dull razor technique. Lesion images have taken for color-based shape and texture feature extraction. Scientists have found new fused color features are effective for melanoma and nevus classification. It has discovered details of 2000 images from ISIC (database archive) helped to build improved feature set. These features were analyzed through various 12 machine learning models as highest accuracy of 93.9%. Proposed Deep Neural Network (DNN) has reached 95.8% accuracy within few epochs. This model was assessed specific limits which were discussed in the results section. In future this exercise will motivate investigators to experience with color features and its variations with other AI based models.

The truck escape ramp is a type of traffic emergency facility for out-of-control trucks on long downhill slopes. As the rigid pebbles were randomly placed in arrester beds, it is difficult to simulate truck tires rolling onto the randomly shaped pebbles. Coupled with an overlapping method, the randomly shaped pebble DEM models were built. Based on the side view of a truck tire, the tire DEM model was built using a close packing agglomerate method. Next, the process of trucks running onto the truck escape ramp was simulated, and the signals from the truck speed and travel distance were recorded. Simulation results are consistent with the test results. The built tire-pebble DEM model could be used for truck escape ramp safety predication and the pavement of the arrester beds.

In pathological practice, a substantial number of procedures are followed to detect and analyze the disease in humans. Usually, pathologists examine the suspicion to diseases in various examination levels ranging from tissues to organs to discover the cause and the stage of the disease. The proposed study aims to investigate the endoscopy recorded digital pictures of Gastric Polyps (GP). This study aims to implement a Computer based Disease Examination Tool (CDET) to analyze the abnormal regions in the stomach. The proposed work comprises a threshold process based on the Brain-Strom-Optimization-Algorithm and Kapur’s Function (BSOA+KF) to augment the polyp fragment and the segmentation based on the Active-Contour (AC) to mine the polyp segment. The performance of implemented technique is checked using the benchmark GP endoscopy images of CVC-ClinicDB dataset. The performance of the proposed CDET is confirmed based on a relative assesment with the Ground-Truth images existing in the considered database. Further, the performance of the AC segmentation is validated with Chan-Vese (LAC) and Seed-Region-Growing (SRG) segmentation techniques. The results of this study confirms that, AC segmentation technique offers better performance values compared to LAC and SRG.

Bird sound classification based on their vocalization has become a significant research field nowadays. Acoustic sound produced by the birds is very rich and used to detect their species. In earlier days ornithologist used to detect the bird species, but this manual recognition is costly and requires huge amount of time. With the advancement of machine learning and deep learning, classification of syllables has become more significant. Methods of automatic sound recognition consist of different stages, such as preprocessing of the input audio file, segmentation of syllables, feature extraction followed by classification. In this study the models used for audio classification are concisely reviewed. Identification becomes more challenging due to a huge similarity between different species. However, noise reduction from the audio files is possible using several machine learning models. Deep learning techniques are also an emerging field in the classification domain, which is discussed in this review. Using these models, it is possible for researchers to detect species or even individual bird from their vocalization which is more time efficient. This paper aims to deliver a review summary, and present guidelines for utilizing the broadly used machine learning techniques in order to identify the challenges as well as future research directions of bird song recognition systems.

In the era of Internet of things and big data, a substantial problem in data transmission, processing, and storage is high data volume. In the case of harmonics, due to the high frequency contents, and the requirement by the Nyquist sampling theorem, short length data becomes impractical for efficient data transmission. Compressive sensing (CS) is a technique that comes to solve this problem deploying the sparsity of the signal to measure the signal with considerably fewer number of samples than the conventional methods. While most of CS techniques are via random sampling matrices, deterministic CS uses well-known matrices and deploy their characteristics. Deterministic CS by chirp codes is deterministic CS where this paper gives a detailed analysis to its technique and its implementation to power signal. The analysis includes the effect on deviations of the parameters of the harmonics, inter-harmonics, sub-harmonics.

With the rapid development of safety science, traditional accident management has become risk management. Based on the analysis of cases, this paper puts forward the method of establishing a risk assessment model, which can effectively reduce the risk of navigation, and constructs the relative risk model based on multi-factors and the Bayesian Risk consequence model under stochastic information. Before a ship goes to complex waters, it can quantitatively analyze the factors such as accident frequency, risk degree and accident consequence, and construct a risk assessment model for navigation accidents, to effectively reduce the probability of navigation accidents.

Medical images aligning is essential due to the existing angular rotation and translation of objects inside the medical images which leads to complex post-processing. Accordingly, medical image registration becomes one of the key tools for aligning images. In image registration, to the images’ frames and the objects are transformed by aligning each image with its corresponding reference one. Demons registration is one of the most efficient non-rigid image registration techniques. In this paper, the proposed system optimized the parameters of Demons registration using the firefly optimization algorithm in retinal images. Several metrics were measured for evaluating the proposed model, including the mean square error, joint entropy, and mutual information calculation between the original and registered images.

Aiming at the shortcomings of long queues and low efficiency of vehicle queues in the process of plane mobile stereo garage parking space allocation. By analyzing the queueing process of the vehicle and the time characteristics of the vehicle entering and leaving the garage, we propose a parking space allocation method based on the probability characteristics of the vehicle entering and leaving the garage. The parking space allocation simulation program under the probability of near allocation and outbound storage is written separately, and the average waiting time of the customer, the average waiting queue length, the average service time and the idle probability of the carrier are used as the evaluation indexes of the service efficiency of the stereo garage, and the two schemes are compared and analyzed. The simulation results show that the parking space allocation with the probability of entering and leaving the garage is better than the nearest allocation principle. This method can effectively improve the service efficiency of the stereo garage.

Composite guidance mode can realize better aircraft control than single guidance mode. However, trajectory wreck and attitude change are inevitable during handover from midcourse guidance to terminal guidance, leading to unpredicted control failure. How to realize smooth handoff is of crucial importance to robust control. Focusing on the problem, an effective method using linear factor to realize smooth transition between the midcourse and terminal guidance for the helicopter-borne aircrafts is proposed in this paper. In accordance with the strict constraint conditions of smooth handover, the application scope of the proposed method is analyzed. The effectiveness of the proposed method has been verified on simulated experiments.

According to the personnel evacuation characteristics of highway tunnel, simulation was conducted to analyze the evacuation pattern of different crowds under the mixed behavior mode by using a renovated cellular automaton model (CA). Research on intelligent evacuation system was based on the basic principle of safety evacuation, the installation location of intelligent evacuation direction system, the tunnel space structure and typical fire scenarios, with comprehensive consideration of factors such as path length, exit width, population density and distribution of evacuated people. The effects of visual induction, auditory induction and dual induction on the evacuation process of intelligent evacuation guidance system were studied by changing the range of guidance signals to population evacuation, and a method for intelligent dynamic identification evacuation path based on multi-parameter was obtained. The results of the simulation show that the intelligent evacuation guidance system can offer a dynamic evacuation route via the real-time control of the guidance signals, such as the sound and light indicators, and instruct the people under the fire to choose the most feasible behavior pattern so as to enhance the efficiency of evacuation. Under the different behavior patterns, it would be possible to effectively reduce the evacuation time via the dual induction mechanism of the sound and the light if a crowd manages to choose the appropriate number, location and direction of the induction signals, and enlarge the impact range of those signals. In addition, based on the intelligent dynamic identification algorithm, the evacuation efficiency can be expected to be raised by controlling the working status of induction signals to provide people with dynamic evacuation route.

Currently most of the traffic camera self-calibration algorithms are performed based on vanishing point, however, there would be a tending to infinity ill-condition of vanishing point in certain angle of view. To overcome this problem, firstly, we establish a typical complementary calibration models, then we get vanishing points through the diamond space created by the vehicle trajectories and body edges, meanwhile detect geometrical markings in road active area, finally, in order to avoid the “shock” effect of the vanishing point, adopt the optimization method to optimize the parameter space of the self-calibration model by using redundant information to make the model more accurate. The experimental results on real traffic images demonstrate the effectiveness and practicability of our self-calibration method, and it is especially suitable for PTZ cameras that constantly changing angles of view.

The analysis of plantar pressure imaging does not perform well preprocessing, dimensionality reduction and feature calculation; this makes the research of foot comfort by statistical methods have the defects of linearization and poor robustness. The intelligent analysis technology to achieve the extraction of the plantar functional area, providing a streamlined and content-rich data set for the study of plantar pressure comfort is very effective and feasible. Different from the existing local-based segmentation technique of the plantar pressure image, the bottom pressure image mean shifting segmentation model segments the plantar pressure image from a global perspective to obtain more accurate segmentation results; by using pixel precision, average pixel precision, uniform cross-section, frequency-to-weight ratio, segmentation accuracy, over-segmentation rate, under-segmentation rate and Dice coefficient for contrast segmentation, the proposed mean shift local de-dimensionality morphological segmentation performs higher effectiveness.

Integrated Modular Avionics (IMA) system realizes the residence of various avionics functions on standardized hardware and integration also brings some problems such as resource defect spread and functional error cross-linking. To make sure the health of the aircraft, it is essential to carry out a comprehensive health assessment of the IMA system. This paper proposes an IMA system health assessment method based on random forest. First, build the IMA system health assessment framework and transform the mapping that affects the health of the IMA system and the health status of the IMA system into the classification problem in supervised learning. Then give a health assessment method based on random forest, and through incremental learning, the new data generated during the operation of the IMA system can improve the evaluation effect in real-time. Finally, through the system health dataset to realize a comprehensive evaluation of the health status of the IMA system. The results show that the method can perform a comprehensive health assessment with high accuracy for the IMA system.

This paper is based on naturalistic driving data that collected by Intelligent Vehicle Test and Evaluation Center of China Automotive Engineering Research Institute. In order to extract the typical scenario from naturalistic driving data, driver behavior characteristics need to be researched. For massive naturalistic driving data, a method of extraction of driving behaviors is needed. This paper focuses on automatic extraction lane changing behavior from naturalistic driving data. Analyzed the original methods and the data in our project, an algorithm which automatically extract lane change data is proposed based on distance between vehicle and lane markings. The results show that, our algorithm can well obtain lane change behaviors from straight roads in highway and urban road, and support research on typical traffic scenario in China.