Skip to main content

About this book

This book constitutes the refereed proceedings of the 11th IFIP TC 12 International Conference on Intelligent Information Processing, IIP 2020, held in Hangzhou, China, in July 2020.

The 24 full papers and 5 short papers presented were carefully reviewed and selected from 36 submissions. They are organized in topical sections on machine learning; multi-agent system; recommendation system; social computing; brain computer integration; pattern recognition; and computer vision and image understanding.

Table of Contents


Machine Learning


A Salient Object Detection Algorithm Based on Region Merging and Clustering

Salient object detection has recently drawn much attention in computer vision such as image compression and object tracking. Currently, various heuristic computational models have been designed. However, extracting the salient objects with a complex background in the image is still a challenging problem. In this paper, we propose a region merging strategy to extract salient region. Firstly, boundary super-pixels are clustered to generate the initial saliency maps based on the prior knowledge that the image boundaries are mostly background. Next, adjacent regions are merged by sorting the multiple feature values of each region. Finally, we get the final saliency maps by merging adjacent or non-adjacent regions by means of the distance from the region to the image center and the boundary length of overlapping regions. The experiments demonstrate that our method performs favorably on three datasets than state-of-art.
Weiyi Wei, Yijing Yang, Wanru Wang, Xiufeng Zhao, Huifang Ma

Link-Based Cluster Ensemble Method for Improved Meta-clustering Algorithm

Ensemble clustering has become a hot research field in intelligent information processing and machine learning. Although significant progress has been made in recent years, there are still two challenging issues in the current ensemble clustering research. First of all, most ensemble clustering algorithms tend to explore similarity at the level of object but lack the ability to explore information at the level of cluster. Secondly, many ensemble clustering algorithms only focus on the direct relationship, while ignoring the indirect relationship between clusters. In order to solve these two problems, a link-based meta-clustering algorithm (L-MCLA) have been proposed in this paper. A series of experiment results demonstrate that the proposed algorithm not only produces better clustering effect but is also less influenced by different ensemble sizes.
Changlong Shao, Shifei Ding

Large-Scale Spectral Clustering with Stochastic Nyström Approximation

In spectral clustering, Nyström approximation is a powerful technique to reduce the time and space cost of matrix decomposition. However, in order to ensure the accurate approximation, a sufficient number of samples are needed. In very large datasets, the internal singular value decomposition (SVD) of Nyström will also spend a large amount of calculation and almost impossible. To solve this problem, this paper proposes a large-scale spectral clustering algorithm with stochastic Nyström approximation. This algorithm uses the stochastic low rank matrix approximation technique to decompose the sampled sub-matrix within the Nyström procedure, losing a slight of accuracy in exchange for a significant improvement of the algorithm efficiency. The performance of the proposed algorithm is tested on benchmark data sets and the clustering results demonstrate its effectiveness.
Hongjie Jia, Liangjun Wang, Heping Song

Feature Selection Algorithm Based on Multi Strategy Grey Wolf Optimizer

Feature selection is an important part of data mining, image recognition and other fields. The efficiency and accuracy of classification algorithm can be improved by selecting the best feature subset. The classical feature selection technology has some limitations, and heuristic optimization algorithm for feature selection is an alternative method to solve these limitations and find the optimal solution. In this paper, we proposed a Multi Strategy Grey Wolf Optimizer algorithm (MSGWO) based on random guidance, local search and subgroup cooperation strategies for feature selection, which solves the problem that the traditional grey wolf optimizer algorithm (GWO) is easy to fall into local optimization with a single search strategy. Among them, the random guidance strategy can make full use of the random characteristics to enhance the global search ability of the population, and the local search strategy makes grey wolf individuals make full use of the search space around the current best solution, and the subgroup cooperation strategy is very important to balance the global search and local search of the algorithm in the iterative process. MSGWO algorithm cooperates with each other in three strategies to update the location of grey wolf individuals, and enhances the global and local search ability of grey wolf individuals. Experimental results show that MSGWO can quickly find the optimal feature combination and effectively improve the performance of the classification model.
Guangyue Zhou, Kewen Li, Guoqiang Wan, Hongtu Ji

A Novel Fuzzy C-means Clustering Algorithm Based on Local Density

Fuzzy C-means (FCM) clustering algorithm is a fuzzy clustering algorithm based on objective function. FCM is the most perfect and widely used algorithm in the theory of fuzzy clustering. However, in the process of clustering, FCM algorithm needs to randomly select the initial cluster center. It is easy to generate problems such as multiple clustering iterations, low convergence speed and unstable clustering. In order to solve the above problems, a novel fuzzy C-means clustering algorithm based on local density is proposed in this paper. Firstly, we calculate the local density of all sample points. Then we select the sample points with the local maximum density as the initial cluster center at each iteration. Finally, the selected initial cluster center are combined with the traditional FCM clustering algorithm to achieve clustering. This method improved the selection of the initial cluster center. The comparative experiment shows that the improved FCM algorithm reduces the number of iterations and improves the convergence speed.
Jian-jun Liu, Jian-cong Fan

A Novel Method to Solve the Separation Problem of LDA

Linear discriminant analysis (LDA) is one of the most classical linear projection techniques for feature extraction, widely used in kinds of fields. Classical LDA is contributed to finding an optimal projection subspace that can maximize the between-class scatter and minimize the average within-class scatter of each class. However, the class separation problem always exists and classical LDA can not guarantee that the within-class scatter of each class get its minimum. In this paper, we proposed the k-classifiers method, which can reduce every within-class scatter of classes respectively and alleviate the class separation problem. This method will be applied in LDA and Norm LDA and achieve significant improvement. Extensive experiments performed on MNIST data sets demonstrate the effectiveness of k-classifiers.
Meng Zhang, Wei Li, Bo Zhang

Multi-label Classification of Short Text Based on Similarity Graph and Restart Random Walk Model

A multi-label classification method of short text based on similarity graph and restart random walk model is proposed. Firstly, the similarity graph is created by using data and labels as the node, and the weights on the edges are calculated through an external knowledge, so the initial matching degree of between the sample and the label set is obtained. After that, we build a label dependency graph with labels as vertices, and using the previous matching degree as the initial prediction value to calculate the relationship between the sample and each node until the probability distribution becomes stable. Finally, the obtained relationship vector is the label probability distribution vector of the sample predicted by the method in this paper. Experimental results show that we provides a more efficient and reliable multi-label short-text classification algorithm.
Xiaohong Li, Fanyi Yang, Yuyin Ma, Huifang Ma

Environmental Parameters Analysis and Power Prediction for Photovoltaic Power Generation Based on Ensembles of Decision Trees

Due to the influence of solar irradiation, temperature and other environmental factors, the output power of photovoltaic power generation has great randomness and randomness discontinuity. In this paper, a method for analyzing environment data related photovoltaic power generation based on ensembles of decision trees algorithm is studied. Firstly, the characteristics of environmental factors of photovoltaic power generation are analyzed by K-means clustering. And then the corresponding cluster label is assigned. Furthermore, the Radom Forests is combined to build a model. Finally, the method is validated by given data above from a real project. The results show that the proposed method can provide reference for the forecasting of photovoltaic power.
Shuai Zhang, Hongwei Dai, Aizhou Yang, Zhongzhi Shi

The Conjugate Entangled Manifold of Space–Time Induced by the Law of Unity of Contradiction

The mechanism of the contradiction between two different objects \( u \) and \( v \) is attributed to a mechanism that their opposite position information “\( x_{u} \)” and “\( x_{v} \)” of u and v are transmitted, respectively, from the initial time \( t_{0} \) , at different speeds \( \dot{x}_{u} \left( t \right) \) and \( \dot{x}_{v} \left( t \right)\,\left( \dot{x}_{v} = -\varsigma\dot{x}_{u} \left( t \right) \right)\), and is meeting at the contradiction point \( t = t_{\lambda} \) and \( x = x_{\lambda} \). Because the coordinate of contradiction point can be noted by \( z_{\lambda} \left(t_{\lambda}, x_{\lambda}\right)\) and \( z_{\lambda}^{*} \left(x_{\lambda}, t_{\lambda}\right)\) in two space time Complex Coordinates Systems which origins are \( z_{o} \left(0_{t}, 0_{x}\right)\) and \( z_{o}^{*} \left(1_{t}, 1_{t}\right)\), respectively, such that the time \( t_{\lambda}\) and the position \( x_{\lambda}\) of the contradictory points can be expressed as the sum of the complex numbers \( z_{\lambda} \left(t_{\lambda}, x_{\lambda}\right)\) and its conjugate \( \bar{z}_{\lambda} \left(t_{\lambda}, x_{\lambda}\right): t_{\lambda} = z_{\lambda} \left(t_{\lambda}, x_{\lambda}\right) + \bar{z}_{\lambda} \left(t_{\lambda}, x_{\lambda}\right) = w_{\lambda} \left(z_{\lambda}, \bar{z}_{\lambda}\right)\), and the difference of \( z_{\lambda}^{*} \left(x_{\lambda}, t_{\lambda}\right)\) and its conjugate: \( \bar z_{\lambda}^{*} \left(x_{\lambda}, t_{\lambda}\right): x_{\lambda} = z_{\lambda}^{*} \left(x_{\lambda}, t_{\lambda}\right) - \bar{z}_{\lambda}^{*} \left(x_{\lambda}, t_{\lambda}\right) = w_{\lambda}^{*} \left(z_{\lambda}^{*} , \bar{z}_{\lambda}^{*}\right) \). By synthesizing the time-space coordinate and the space-me coordinate, such their time axis \( \left[0_{t}, 1_{t}\right]\) and the space axis \( \left[1_{x}, 0_{x}\right]\) of the two complex coordinate systems are coincide with the intervals \( \left[u, v\right] \), respectively, then the contradiction point can be expressed in the synthesis Coordinate System to be a wave function: \( \psi\left(w_{\lambda}, w_{\lambda}^{*}\right) = t_{\lambda} - ix_{\lambda} = w_{\lambda} - iw_{\lambda}^{*}\). Because of the varying direction of two information “\( x_u \)” and “\( x_v \)” and their increments \( \Delta{x}_u \left( \Delta_{u}t\right) = \dot{x}_u \left( \Delta_{u}t\right) \Delta_{u}t\) and \( \Delta{x}_v \left( \Delta_{v}t\right) = \dot{x}_v \left( \Delta_{v}t\right) \Delta_{v}t\) with time t and increment \( \Delta{t} = t - 0_{t}\) are opposite each other, so the \( t_{\lambda}\) of the wave function \( \psi \)is on the time axis \( \left[0_{t}, 1_{t}\right] \) and the \( x_{\lambda} \) on the space axis \( \left[1_{x}, 0_{x}\right] \), constructed a pair of information transmission streams entangled in opposite directions appear, such that the interval [u, v] constitutes a space-time conjugate entangled manifold. The invariance of the contradiction point or wave function \( \psi\left(w_{\lambda}, w_{\lambda}^*\right) \), under the unit scale transformation of time and distance measurement, not only make all points \( z\left(t, x\right) \in \left[u, v\right] \) is contradiction point, and makes \( \lambda = \frac{1}{2} \) and \( \zeta = \frac{\lambda}{1}-{\lambda} \) It is also shown that since λ changes from 0 to \( \frac{1}{2} \) is equivalent to the integral for the on \( t_{\lambda} \) and \( x_{\lambda} \) in wave function ψ from 0 to \( \frac{1}{2} \), respectively, by it not only the inner product ψ of the ψ and the time component \( t_{\lambda} \), respectively \( \psi \left(w, w^{*}\right). t_{\lambda} \), and the outer product of ψ and the spatial component \( \psi \left(w, w^{*}\right) \wedge x_{\lambda}\) can be get, but also their sum: \( \psi \left(w, w^{*}\right) \cdot \psi_{t} + \psi\left(w, w^{*}\right) \wedge \psi_{x} \) can be gotten too.
Jiali Feng, Jingjuan Feng

Similarity Evaluation with Wikipedia Features

Wikipedia provides rich semantic features e.g., text, link, and category structure. These features can be used to compute semantic similarity (SS) between words or concepts. However, some existing Wikipedia-based SS methods either rely on a single feature or do not incorporate the underlying statistics of different features. We propose novel vector representations of Wikipedia concepts by integrating their multiple semantic features. We utilize the available statistics of these features in Wikipedia to compute their weights. These weights signify the contribution of each feature in similarity evaluation according to its level of importance. The experimental evaluation shows that our new methods obtain better results on SS datasets in comparison with state-of-the-art SS methods.
Shahbaz Wasti, Jawad Hussain, Guangjiang Huang, Yuncheng Jiang

Multi-Agent System


Adaptive Game AI-Based Dynamic Difficulty Scaling via the Symbiotic Game Agent

This work presents AdaptiveSGA, a model for implementing Dynamic Difficulty Scaling through Adaptive Game AI via the Symbiotic Game Agent framework. The use of Dynamic Difficulty Balancing in modern computer games is useful when looking to improve the entertainment value of a game. Moreover, the Symbiotic Game Agent, as a framework, provides flexibility and robustness as a design principle for game agents. The work presented here leverages both the advantages of Adaptive Game AI and Symbiotic Game Agents to implement a robust, efficient and testable model for game difficulty scaling. The model is discussed in detail and is compared to the original Symbiotic Game Agent architecture. Finally, the paper describes how it was applied in simulated soccer. Finally, experimental results, which show that Dynamic Difficulty Balancing was achieved, are briefly analyzed.
Siphesihle Philezwini Sithungu, Elizabeth Marie Ehlers

Recommendation System


Scientific Paper Recommendation Using Author’s Dual Role Citation Relationship

Vector representations learning (also known as embeddings) for users (items) are at the core of modern recommendation systems. Existing works usually map users and items to low-dimensional space to predict user preferences for items and describe pre-existing features (such as ID) of users (or items) to obtain the embedding of the user (or item). However, we argue that such methods neglect the dual role of users, side information of users and items (e.g., dual citation relationship of authors, authoritativeness of authors and papers) when recommendation is performed for scientific paper. As such, the resulting representations may be insufficient to predict optimal author citations.
In this paper, we contribute a new model named scientific paper recommendation using Author’s Dual Role Citation Relationship (ADRCR) to capture authors’ citation relationship. Our model incorporates the reference relation between author and author, the citation relationship between author and paper, and the authoritativeness of authors and papers into a unified framework. In particular, our model predicts author citation relationship in each specific class. Experiments on the DBLP dataset demonstrate that ADRCR outperforms state-of-the-art recommendation methods. Further analysis shows that modeling the author’s dual role is particularly helpful for providing recommendation for sparse users that have very few interactions.
Donglin Hu, Huifang Ma, Yuhang Liu, Xiangchun He

A Genetic Algorithm for Travel Itinerary Recommendation with Mandatory Points-of-Interest

Traveling as a very popular leisure activity enjoyed by many people all over the world. Typically, people would visit the POIs that are popular or special in a city and also have desired starting POIs (e.g., POIs that are close to their hotels) and destination POIs (e.g., POIs that are near train stations or airports). However, travelers often have limited travel time and are also unfamiliar with the wide range of Points-of-Interest (POIs) in a city, so that the itinerary planning is time-consuming and challenging. In this paper, we view this kind of itinerary planning as MandatoryTour problem, which is tourists have to construct an itinerary comprising a series of POIs of a city and including as many popular or special POIs as possible within their travel time budget. We term the most popular and special POIs as mandatory POIs in our paper. For solving the presented MandatoryTour problem, we propose a genetic algorithm GAM. We compare our approach against several baselines GA, MaxM, and GreedyM by using real-world datasets from the Yahoo Flickr Creative Commons 100 Million Dataset (YFCC100M), which include POI visits of seven touristic cities. The experimental results show that GAM achieves better recommendation performance in terms of the mandatory POIs, POIs visited, time budget (travel time and visit duration), and profit (POI popularity).
Phatpicha Yochum, Liang Chang, Tianlong Gu, Manli Zhu, Hongliang Chen

Social Computing


Stochastic Blockmodels Meets Overlapping Community Detection

It turns out that the Stochastic Blockmodel (SBM) and its variants can successfully accomplish a variety of tasks, such as discovering community structures. Note that the main limitations are inferencing high time complexity and poor scalability. Our effort is motivated by the goal of harnessing their complementary strengths to develop a scalability SBM for graphs, that also enjoys an efficient inference process and discovery interpretable communities. Unlike traditional SBM that each node is assumed to belong to just one block, we wish to use the node importance to also infer the community membership(s) of each node (as it is one of the goals of SBMs). To this end, we propose a multi-stage maximum likelihood strategy for inferring the latent parameters of adapting the Stochastic Blockmodels to Overlapping Community Detection (OCD-SBM). The intuitive properties to build the model, is more in line with the real-world network to reveal the hidden community structural characteristics. Particularly, this enables inference of not just the node’s membership into communities, but the strength of the membership in each of the communities the node belongs to. Experiments conducted on various datasets verify the effectiveness of our model.
Qiqi Zhao, Huifang Ma, Zhixin Li, Lijun Guo

Overlapping Community Detection Combining Topological Potential and Trust Value of Nodes

Aiming at the problems of existing algorithms, such as instability, neglecting interaction between nodes and neglecting attributes of node, an overlapping community discovery algorithm combining topological potential and trust value of nodes was proposed. Firstly, the importance of nodes is calculated according to topological potential and the trust value of the node, and then K core nodes are selected. In final, the final division of communities are finished by using the extended modularity and core nodes. Experimental results on LFR network datasets and three real network datasets, verify the efficiency of the proposed OCDTT algorithm.
Xiaohong Li, Weiying Kong, Weiyi Wei, Enli Fu, Huifang Ma

Brain Computer Integration


Coarse-to-Fine Classification with Phase Synchronization and Common Spatial Pattern for Motor Imagery-Based BCI

How to improve the classification accuracy is a key issue in four-class motor imagery-based brain-computer interface (MI-BCI) systems. In this paper, a method based on phase synchronization analysis and common spatial pattern (CSP) algorithm is proposed. The proposed method embodies the idea of the inverted binary tree, which transforms the multi-class problem into several binary problems. First, the phase locking value (PLV) is calculated as a feature of phase synchronization, then the calculated correlation coefficients of the phase synchronization features are used to construct two pairs of class. Subsequently, we use CSP to extract the features of each class pair and use the linear discriminant analysis (LDA) to classify the test samples and obtain coarse classification results. Finally, the two classes obtained from the coarse classification form a new class pair. We use CSP and LDA to classify the test samples and get the fine classification results. The performance of method is evaluated on BCI Competition IV dataset IIa. The average kappa coefficient of our method is ranked third among the experimental results of the first five contestants. In addition, the classification performance of several subjects is significantly improved. These results show this method is effective for multi-class motor imagery classification.
Wenfen Ling, Feipeng Xu, Qiaonan Fan, Yong Peng, Wanzeng Kong

Ballistocardiogram Artifact Removal for Concurrent EEG-fMRI Recordings Using Blind Source Separation Based on Dictionary Learning

Simultaneous recording of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) have attracted extensive attention and research owing to their high spatial and temporal resolution. However, EEG data are easily influenced by physiological causes, gradient artifact (GA) and ballistocardiogram (BCG) artifact. In this paper, a new blind source separation technique based on dictionary learning is proposed to remove BCG artifact. The dictionary is learned from original data which represents the features of clean EEG signals and BCG artifact. Then, the dictionary atoms are classified according to a list of standards. Finally, clean EEG signals are obtained from the linear combination of the modified dictionary. The proposed method, ICA, AAS, and OBS are tested and compared using simulated data and real simultaneous EEG–fMRI data. The results suggest the efficacy and advantages of the proposed method in the removal of BCG artifacts.
Yuxi Liu, Jianhai Zhang, Bohui Zhang, Wanzeng Kong

Comparison of Machine Learning and Deep Learning Approaches for Decoding Brain Computer Interface: An fNIRS Study

Recently, deep learning has gained great attention in decoding the neuro-physiological signal. However, which one (classical machine learning or deep learning) has better performance for decoding the functional near-infrared spectroscopy (fNIRS) signal is still lack of full verification. Thus, in this paper, we systematically compared the performance of many classical machine learning methods and deep learning methods in fNIRS data processing for decoding the mental arithmetic task. The classical machine learning methods such as decision tree, linear discriminant analysis (LDA), support vector machine (SVM), K-Nearest Neighbor (KNN) and ensemble methods with strict feature extraction and screening, were used for performance comparison, while the long short-term memory-fully convolutional network (LSTM-FCN) method as a representative of deep leaning methods was applied. Results showed that the classification performance of SVM was the best among the classical machine learning methods, achieving that the average accuracy of the subject-related/unrelated were 91.0% and 83.0%, respectively. Furthermore, the classification accuracy of deep learning was significantly better than that of the involved classical machine learning methods, where the accuracy of deep learning could reach 95.3% with subject-related condition and 97.1% with subject-unrelated condition, respectively. Thus, this paper has totally showed the excellent performance of LSTM-FCN as a representative of deep learning in decoding brain signal from fNIRS dataset, which has outperformed many classical machine learning methods.
Jiahao Lu, Hongjie Yan, Chunqi Chang, Nizhuan Wang

Pattern Recognition


Phase Plane Analysis of Traffic Flow Evolution Based on a Macroscopic Traffic Flow Model

In this paper, a new phase plane analysis method is proposed to study the nonlinear phenomena of traffic flow. Most of the papers describe only one or several traffic phenomena and do not analyze all of them from the perspective of system stability. Therefore, this paper studies the phase plane analysis of traffic flow phenomenon from the perspective of traffic system stability, and describes various complicated nonlinear traffic phenomena through phase plane analysis.
WenHuan Ai, Tao Xing, YuHang Su, DaWei Liu, Huifang Ma

Phase Plane Analysis of Traffic Phenomena with Different Input and Output Conditions

In this paper, the traffic flow problem is converted into a system stability problem through variable substitution and a phase plane analysis method is presented for analyzing the complex nonlinear traffic phenomena. This method matches traffic congestion with the unstable system. So these theories and methods of stability can be applied directly to solve the traffic problem. Based on an anisotropic continuum model developed by Gupta and Katiyar (GK model), this paper uses this new method to describe various nonlinear phenomena due to different input and output conditions on ramps which were rarely studied in the past. The results show that the traffic phenomena described by the new method is consistent with that described by traditional methods. Moreover, the phase plane diagram highlights the unstable traffic phenomena we are chiefly concerned about and describes the variation of density or velocity with time or sections more clearly.
WenHuan Ai, YuHang Su, Tao Xing, DaWei Liu, Huifang Ma

Bird Detection on Transmission Lines Based on DC-YOLO Model

In order to accurately detect the number of birds around the transmission line, promptly drive the birds away to ensure the normal operation of the line, a DC-YOLO model is designed. This model is based on the deep learning target detection algorithm YOLO V3 and proposes two improvements: Replacing the convolutional layer in the original network with dilated convolution to maintain a larger receptive field and higher resolution, improving the model’s accuracy for small targets; The confidence score of the detection frame is updated by calculating the scale factor, and the detection frame with a score lower than the threshold is finally removed. The NMS algorithm is optimized to improve the model’s ability to detect occluded birds. Experimental results show that the DC-YOLO model detection accuracy can reach 86.31%, which can effectively detect birds around transmission lines.
Cong Zou, Yong-quan Liang

Research on Customer Credit Scoring Model Based on Bank Credit Card

With the development of China’s economy, especially the maturity of the market economy, credit is important to the society and individuals. At present, credit system is mainly divided into two parts. Enterprise credit system is an important part of social credit system. But at the same time, as the foundation of social credit system, the establishment of the personal credit system is of great significance to reduce the cost of collecting information and improve the efficiency of loan processing. At the bank level, this paper discretizes the credit card data of a bank, selects the features by calculating Weight of Evidence and Information Value, and information divergence, then uses Logistic Regression to predict. Finally, the results of the Logistic Regression are transformed into visualized credit scores to establish a credit scoring model. It is verified that this model has a good prediction effect.
Maoguang Wang, Hang Yang

Analysis of the Stability and Solitary Waves for the Car-Following Model on Two Lanes

In this paper, Analysis of the stability and solitary waves for a car-following model on two lanes is carried out. The stability condition of the model is obtained by using the linear stability theory. We study the nonlinear characteristics of the model and obtain the solutions of Burgers equation, KDV equation, and MKDV equation, which can be used to describe density waves in three regions (i.e., stable, metastable and unstable), respectively. The analytical results show that traffic flow can be stabilized further by incorporating the effects come from the leading car of the nearest car on neighbor lane into car-following model.
WenHuan Ai, Tao Xing, YuHang Su, DaWei Liu, Huifang Ma

Queue Length Estimation Based Defence Against Data Poisoning Attack for Traffic Signal Control

With the development of intelligent transportation systems, especially in the context of the comprehensive development and popularization of big data and 5G networks, intelligent transportation signal systems have been experimented and promoted in various countries around the world. As with other big data-based systems, specific attacks pose a threat to the security of big data-based intelligent transportation system systems. Targeting system vulnerabilities, certain simple forms of attack will have a huge impact on signal planning, making Actual traffic is congested. In this article, we first show a specific attack and then add more attack points, analyze the system’s vulnerabilities, and model based on traffic waves and Bayesian predictions, so that the attack points can help the impact is weakened and the traffic can function normally. For experiments, we performed traffic simulation on the VISSIM platform to prove the impact of our attack and further verify the accuracy and effectiveness of the model.
Xu Gao, Jiqiang Liu, Yike Li, Xiaojin Wang, YingXiao Xiang, Endong Tong, Wenjia Niu, Zhen Han

A Method of Style Transfer for Chinese Painting

This paper introduces a style transfer method for traditional Chinese painting. We improved the traditional method by adding style characteristics and constraints unique to Chinese painting. By comparing Chinese painting with Western painting and natural pictures, we find that the features such as lines and textures in Chinese painting are quite different from other images. Therefore, these features are extracted and added to the original method in a restrictive manner. Finally, experiments prove that the method has a certain improvement effect on the style transfer result of Chinese painting.
Cunjian Chen

Speech Triggered Mobility Support and Privacy

Current voice assistants are offered by large IT companies such as Google, Amazon, Microsoft, Apple or Baidu. The voice assistants include numerous functionalities, which are usually executed centrally in the cloud by the providers. Nevertheless, the providers offer imprecise information on what happens to the input data of the users. Users cannot be sure whether their privacy and data are protected. The central research question is what is currently happening with the voice-based interaction between users and services, and what concepts for configurable data protection by users are conceivable in the future. In this article, we present the survey results obtained by speech assistant users. The results show, in particular, the willingness to pay for individually configurable privacy. The concept for a voice assistant with privacy-awareness is proposed and prototypically implemented.
Michael Zipperle, Marius Becherer, Achim Karduck

Computer Vision and Image Understanding


Explaining Color Evolution, Color Blindness, and Color Recognition by the Decoding Model of Color Vision

The author proposed the decoding model of color vision in 1987. International Commission on Illumination (CIE) recommended almost the same symmetric color model for color transform in 2006. For readers to understand the decoding model better, this paper first introduces the decoding model, then uses this model to explain the opponent-process, color evolution, and color blindness pictorially. Recent references on the decoding and reconstruction of colors in the visual cortex induce a new question: what is the decoding algorithm from ganglion cells to the visual cortex? This paper also explains the decoding algorithm. The decoding model is explained as a fuzzy 3–8 decoder. The fuzzy logic used is compatible with Boolean Algebra. The model first obtains the median M of three cones’ outputs B, G, and R, and then obtain three opponent signals by B, G, and R minus M respectively. This model can unify Young and Helmholtz’s tri-pigment theory and Hering’s opponent theory more naturally than the popular zone models. It is symmetrical and compatible with the popular color transform method for computer graphics. The transform from the RGB system to HSV system according to the decoding model is introduced. Several fuzzy 3–8 or 4–16 … fuzzy decoders can be used to construct a Decoding Neural Network (DNN). The decoding algorithm from ganglion cells to the visual cortex can be explained with a two-layer DNN. The reasonability of the decoding model and the potential applications of the DNN are discussed.
Chenguang Lu

A Content-Based Deep Hybrid Approach with Segmented Max-Pooling

Convolutional matrix factorization (ConvMF), which integrates convolutional neural network (CNN) into probabilistic matrix factorization (PMF), has been recently proposed to utilize the contextual information and achieve higher rating prediction accuracy of model-based collaborative filtering (CF) recommender systems. While ConvMF uses max-pooling, which may lose the feature’s location and frequency information. In order to solve this problem, a novel approach with segmented max-pooling (ConvMF-S) has been proposed in this paper. ConvMF-S can extract multiple features and keep their location and frequency information. Experiments show that the rating prediction accuracy has been improved.
Dapeng Zhang, Liu Yajun, Jiancheng Liu

Image Caption Combined with GAN Training Method

In today’s world where the number of images is huge and people cannot quickly retrieve the information they need, we urgently need a simpler and more human-friendly way of understanding images, and image captions have emerged. Image caption, as its name suggests, is to analyze and understand image information to generate natural language descriptions of specific images. In recent years, it has been widely used in image-text crossover studies, early infant education, and assisted by disadvantaged groups. And the favor of industry, has produced many excellent research results. At present, the evaluation of image caption is basically based on objective evaluation indicators such as BLUE and CIDEr. It is easy to prevent the generated caption from approaching human language expression. The introduction of GAN idea allows us to use a new method of adversarial training. To evaluate the generated caption, the evaluation module is more natural and comprehensive. Considering the requirements for image fidelity, this topic proposes a GAN-based image description. The Attention mechanism is introduced to improve image fidelity, which makes the generated caption more accurate and more close to human language expression.
Zeqin Huang, Zhongzhi Shi


Additional information

Premium Partner

    Image Credits