Rongfang Bie

• Professor at Beijing Normal University

Pre-trained encoders in computer vision have recently received great attention from both research and industry communities. Among others, a promising paradigm is to utilize self-supervised learning(SSL) to train image encoders with massive unlabeled samples, thereby endowing encoders with the capability to embed abundant knowledge into the feature...

Machine Learning as a Service (MLaaS) offers powerful data analytics services to clients with limited resources. However, it still raises concerns about the integrity of delegated computation and the privacy of the server's model parameters. To address these issues, zero-knowledge Machine Learning (zkML) has been suggested for computation verifiabi...

Convolutional neural network (CNN)-based methods facilitate data classification but sacrifice physical interpretability due to the complex model architecture and tight inferring integration. The interpretability requirement of our prior CNN-based golf classifier motivates us to explain the performance of the predictions and to discover the class-di...

IOTA has emerged as a promising blockchain platform specially designed for the Internet of Things (IoT). Its distributed ledger, called tangle, adopts a directed acyclic graph (DAG) structure to achieve fast transaction confirmation and high scalability. While the tangle tremendously mitigates blockchain performance concerns relative to a tradition...

Blockchain-based data storage has become an emerging paradigm, providing a fair and transparent data platform for decentralized applications. However, how to achieve secure on-chain verification for arbitrary SQL queries in such a decentralized storage remains under-explored. Due to the limitations of authenticated data structure (ADS), existing wo...

With the rapid development of Internet technology, the content of cultural education is rapidly transferred from offline to online. The transformation solves the problem of time and space limitation, yet brings new challenges such as digital copyright protection, quality certification of curriculum resources, education evaluation and certification,...

Traceability and trustiness are two critical issues in the logistics sector. Blockchain provides a potential way for logistics tracking systems due to its traits of tamper resistance. However, it is non-trivial to apply blockchain on logistics because of firstly, the binding relationship between virtue data and physical location cannot be guarantee...

Signals conveyed by sophisticated sensor equipment play a pivotal role in evaluating athletes’ exercise performance. However, some of the sports, such as modern golf, contain too many exercise phases within a short period. Consequently, these sports often face the dilemma, multi-classification with scarce data, and thus it is hard to analyze the da...

IOTA is widely used in the field of Internet of things, but it also produces a series of security problems which brings great security risks to IOTA. Starting from the perspective of parasite chain attack, we put forward the cost function of nodes successfully launching parasite chain attack. On this basis, the game relationship between nodes in IO...

As the carrier of knowledge and the basis of teaching, teaching materials affect the teaching quality of schools. However, the selection of teaching materials in colleges and universities is in a dilemma because of the variety of teaching materials and the uneven quality of teaching materials. Blockchain has gradually drawn attention from various f...

The development of online education has broken the limitations of traditional education in region and time, and promoted the reform of education. However, there are some problems in traditional online education, such as data island, lack of data sharing model and so on. This paper constructs an online education data management model based on blockc...

IOTA is a new cryptocurrency system designed for the Internet of Things based on directed an acyclic graph structure. It has the advantages of supporting high concurrency, scalability, and zero transaction fees; however, due to the particularity of the directed acyclic graph structure, IOTA faces more complex security threats than the sequence bloc...

SMEs in China always face financing constraints and hardly obtain bank loans under unsound financing system due to the information asymmetry, while thousands of SMEs have contributed greatly to Chinese economic development in the last decades. Credit reporting has been verified to be an effective way to lower information asymmetry. However, existed...

The DTN (Delay/Interrupt Tolerant Network) protocol that relies on nodes to handle network interruptions is one of the important components of the wireless sensor network (WSN) routing protocol. However, due to resource consumption, nodes may be unable to unconditionally relay data. To address this issue, several incentive mechanisms have recently...

The lack of radio spectrum resources and low resource utilization have always been existing problems. Current researchers usually use spectrum auctions to improve the utilization of radio spectrum. The existing auction mechanisms rarely consider the attributes of the radio frequency spectrums, which is likely to cause problems that it cannot meet u...

Spectrum auction is one of the most effective ways to achieve dynamic spectrum allocation in cognitive radio networks, and it provides one effective way to manage the spectrum demands of IoT devices with limited resources. Most spectrum auctions focus on protecting bidder privacy and achieving excellent social efficiency, but few tackles the verifi...

The testing information refers to the data obtained by the quality inspection center after testing steel, cement and other materials through testing machines and other instruments. The quality inspection center issues a report according to the testing result, which serves as the basis for the batch of materials to be put into the market. At present...

Inpainting represents a procedure which can restore the lost parts of an image based upon the residual information. We present an inpainting network that consists of an Encoder-Decoder pipeline and a multi-dimensional adversarial network. The Encoder-Decoder pipeline extracts features from the input image with missing area and learns these features...

In the military, police, security companies, and shooting sports, precision shooting training is of the outmost importance. In order to achieve high shooting accuracy, a lot of training is needed. As a result, trainees use a large number of cartridges and a considerable amount of time of professional trainers, which can cost a lot. Our motivation i...

We consider a practical multiunit heterogeneous spectrum market in which each buyer may request multiple channels with different bid prices at different geographical regions and each channel is associated with a reserve price indicating the desired revenue of the seller. The degree-of-freedom brought by multiunit trading and (reserve and bid) price...

Medical image analysis is motivated by the success of deep learning, where annotations are usually expensive and not easy to obtain. In this paper, we propose a deep quintuplet network CXNet-m3, where the classification of lesion type of chest x-ray images (CXRs) could benefit from easily accessible annotations like patient age, gender, identity an...

Nowadays, various online education platforms (such as MOOCs, Coursera, XuetangX and so on) not only provide a broad Internet environment for sharing multimedia learning resources, but also bring a series of challenges in digital rights management, such as the infringement of digital copyrights of multimedia learning resources, the insecurity of dig...

Featured Application This method based on deep learning may be useful in the computer-aided detection of multiple lesions on chest X-ray images. Abstract Automated detection of lung lesions on Chest X-ray images shows good performance to reduce lung cancer mortality. However, it is difficult to detect multiple lesions of single image well and trul...

Mobile ad hoc Network (MANET) is a cluster of moveable devices connected through a wireless medium to design network with rapidly changing topologies due to mobility. MANETs are applicable in variety of innovative application scenarios where smart devices exchange data among each other. In this case, security of data is the major concern to provide...

In military, police, security companies, and shooting sports, precision shooting training is of the outmost importance. In order to achieve high accuracy, trainees need to do a lot of training. Consequently, they will consume a great number of rounds (cartridges) and a considerable amount of professional coaches’ time - both could cost much. Our mo...

Chest X-ray images play an important role in diagnosing thorax diseases. They can show the complete state of the human chest, including the heart, lungs, ribs, etc. However, the diagnosis of diseases is mainly based on the location and morphology of diseased tissues, which requires our model to pay more attention to the lesion, so we choose to intr...

Crowdsourcing is a promising technology to accomplish a complex task via eliciting services from a large group of contributors. Recent observations indicate that the success of crowdsourcing has been threatened by the malicious behaviors of the contributors. In this paper, we analyze the attack problem using an iterated prisoner’s dilemma (IPD) gam...

Diagnosing multiple lesions on images is facing with challenges of incomplete and incorrect disease detection. In this paper, we propose a deep model called CXNet-m2 for the detection of multiple lesions on chest X-ray images. In our model, there is a convolutional neural network (CNN) for encoding the images, a recurrent neural network (RNN) for g...

Hundreds of thousands of images that are widely used in different fields of modern life have appeared in recent years. The process of retrieving the target images from a big database has become a meaningful problem. As one of the classical techniques of computer vision, image retrieval could effectively solve the problem. However, in most cases, hi...

Thanks to the improvement of technologies such as Internet of Things, bio-sensing and data mining, smart wearable technologies have recently received increasing attention for teenagers’ sport and health monitoring. Despite the powerful data-acquisition ability of the current wearable products on the market, they still suffer performance deficiency...

High-quality retrieval techniques can effectively retrieve target images from millions of images, and some classic techniques are widely used in different fields. As a classic image retrieval technique, deep learning shows remarkable advantages in significantly improving retrieval results. However, high-quality retrieval results highly depend on su...

Quality control in crowdsourcing is challenging due to the heterogeneous nature of the workers. The state-of-the-art solutions attempt to address the issue from the technical perspective, which may be costly because they function as an additional procedure in crowdsourcing. In this paper, an economics based idea is adopted to embed quality control...

As a classic technique in the field of computer vision, image composition has been widely used and recognized. It could composite new regions extracted from candidate images into target images, which could achieve realistic composited results. However, in the process of composition, it is difficult for traditional composition methods to select enou...

The development in single-cell technology has enabled to quantify the high throughput gene expressions of individual cell, and it became possible to discover heterogeneity at cell level. To detect heterogeneity within cell population remains challenging in presence of outliers, biological noise, and dropouts. SIMLR (single-cell interpretation via m...

In the era of big data and Internet, social network platforms, blogs, and recommender systems generate thousands of subjective information every day. The emotional content of these information may be related to books, characters, commodities, activities and so on. Analyzing and mining subjective emotional information is conducive to personal decisi...

Spectral clustering is of significance for many research areas, but high computation complexity restricted its power obviously. Even though Power Iteration Clustering(PIC) algorithm could speed up its process by random selection of the initial vector, it is still at the cost of accuracy reduction. Aiming at the problems above, we proposed one Semi-...

Inpainting refers to reconstruct the incomplete image or video via analysing their context, feature of the tailing etc. Convolutional neural network with deep learning is proved to be an effective method to achieve inpainting. However, those algorithms existed now usually have vague and blurry results with huge amount of time to train the models. T...

Detecting anomaly of chest X-ray images by advanced technologies, such as deep learning, is an urgent need to improve the work efficiency and diagnosis accuracy. Fine-tuning existing deep learning networks for medical image processing suffers from over-fitting and low transfer efficiency. To overcome such limitations, we design a hierarchical convo...

Image inpainting is an essential process of semantically filling the missing holes in a corrupt image. However, concurrent methods cannot semantically recover some self-described objects, such as a text instance. In this paper, we focus on the recovery of a missing character in a detected corrupt text instance on an image and propose a procedure to...

Three-dimensional (3D) reconstruction of a single protein molecule is essential for understanding the relationship between the structural dynamics and functions of the protein. Electron tomography (ET) provides a tool for imaging an individual particle of protein from a series of tilted angles. Individual-particle electron tomography (IPET) provide...

Image inpainting aims to fill the corrupt area semantically and recover the semantic and detailed information; however, concurrent methods suffer from convergence crash and arbitrarily paste the inference of the missing area into the corrupt context. In this paper, we study a combination of an encoder–decoder generator for image semantic inpainting...

Human facial expressions change so subtly that recognition accuracy of most traditional approaches largely depend on feature extraction. In this article, the authors employ a deep convolutional neural network (CNN) to devise a facial expression recognition system to discover deeper feature representation of facial expression. The proposed system is...

The use of smart sports equipment and body sensory systems supervising daily sports training is gradually emerging in professional and amateur sports; however, the problem of processing large amounts of data from sensors used in sport and discovering constructive knowledge is a novel topic and the focus of our research. In this article, we investig...

In recent years, smart sports equipment and body sensor systems have become popular in professional and amateur sports. One of a few remaining problems in real-time applications is the discovery of knowledge from the embedded sensors data. In sports training, such knowledge helps accelerated motor learning. The authors start with exploring the poss...

With the finish of the human genome sequencing and the great progress in molecular biology like proteomics, many established authoritative international biomedical databases are completing continually in recent years. With these opening databases, all kinds of biological molecular networks can be constructed for potential disease gene detection and...

Due to insufficiency in data collection and analysis by using the existing wearable devices in physical fitness tests for adolescents, this paper presents a machine learning based physical health evaluation model for running activity monitoring, in which a gradient boosting regression algorithm is employed to process physiological data collected fr...

Discovering the similar groups is a popular primary step in analysis of biomedical data, which cannot be identified manually. Many supervised and unsupervised machine learning and statistical approaches have been developed to solve this problem. Clustering is an unsupervised learning approach, which organizes the data into similar groups, and is us...

As one classic technique of computer vision, image retrieval could retrieve the target images from hundreds of thousands of images effectively. Furthermore, with the rapid development of deep learning, the quality of retrieval is increased obviously. However, under normal conditions, the high-quality retrieval is supported by a large number of lear...

Air quality estimation is an important and fundamental problem in environmental protection. Several efforts have been made in the past decades using expensive sensor-based or indirect methods like based on social networks; however, image-based air pollution estimation is still far from solved. This paper devises an effective convolutional neural ne...

Urban water supply network is ubiquitous and indispensable to city dwellers especially in the era of global urbanization. Preventative maintenance of water pipes, especially in urban-scale networks, thus becomes of vital importance. To achieve this goal, failure prediction that aims to pro-actively pinpoint those ‘most-risky-to-fail’ pipes becomes...

Combinatorial auctions are employed into many applications such as spectrum auctions held by the Federal Communications Commission (FCC). A crucial problem in such auctions is the lack of secure and efficiency mechanism to protect the privacy of the bidding prices and to ensure data security. To solve the problem, we propose an approach to represen...

The pursuit of high-quality modulation has become an inevitable trend along with a continuous evolution of mobile communication technologies. The filter-bank based multicarrier (FBMC) modulation, which has aroused wide concern recently for its higher spectral efficiency than orthogonal frequency division multiplexing (OFDM), is considered as a prom...

In recent years smart sport equipments have achieved great success in professional and amateur sports, as well as body sensory systems; now discovering interesting knowledge in the surge of data from those embedded sensors used in sports is necessary and the focus of our research. In this paper, we investigate golf swing data classification method...

Multi-scales data containing structures at different scales of shape and density is very common in both synthetic and real world. However, it is a big problem to cluster this kind of data accurately. Choosing an appropriate clustering number is the first step, important and not easy. In this paper, we propose a skinny method, DP-Dip, to estimate th...

With the growth of deep learning, object recognition has received increasing interests and its accuracy has been improved significantly in the past few years, However, high-quality recognition largely depends on a large number of learning instances. If the number of learning instances is reduced, it’s difficult to maintain realistic recognition acc...

Image recognition has plateaued in the last few years. According to this research field, some complicated models typically combined feature extraction and classification models effectively. Moreover, many classic models have already achieved realistic recognition. However, there are still some drawbacks of traditional methods. On the one hand, some...

Contour detection is an important and fundamental problem in computer vision which finds numerous applications. Despite significant progress has been made in the past decades, contour detection from natural images remains a challenging task due to the difficulty of clearly distinguishing between edges of objects and surrounding backgrounds. To addr...

Social networking has become part of our life in recent years, allowing users to converse and connect with people sharing similar interests in real world. However, networking via the social media suffers from serious privacy issues, and one of which is profile attribute leakage in friend discovery. While existing studies mainly focus on leveraging...

As one classic technique, object recognition could identify objects in an image effectively and it has been improved by deep learning model significantly. However, in the process of object recognition, complicated background could have negative on the feature extraction which directly reduces the quality of object recognition. Although some methods...

Location-based services (LBSs) are becoming an increasingly important component in our social and business life. All existing LBS providers support the nearest place searching via a single point of interest (POI) query. That is, in one query, a user is allowed to search for only one type of service. However, in real life, people usually need to sea...

Image retrieval is a recognition technique in the field of computer vision. In most cases, high-quality retrieval is often supported by adequate learning instances. However, in the process of learning instance selection, some useless, repeated, invalid, and even mistaken learning instances are often selected. Low-quality instances not only add to t...

In Cooperative Cognitive Radio Networks (CCRNs), primary users and secondary users are mutual beneficial through setting up cooperative transmission. But, in most of the existing work, only the SUs working as the relays can have the opportunities to access primary users’ channel, which may incur a waste of resource or a throughput reduction if the...

As a promising paradigm, mobile crowdsensing exerts the potential of widespread sensors embedded in mobile devices. The greedy nature of workers brings the problem of low-quality sensing data, which poses threats to the overall performance of a crowdsensing system. Existing works often tackle this problem with additional function components. In thi...

Crowdsourcing applications are vulnerable to malicious behavior, which poses a serious threat to their adoption and large deployment. Based on the notion that the requestor (i.e., the crowdsourcer) can block malicious behavior via leveraging the market power through the task allocation and pricing, we propose two novel frameworks based on the mecha...

Image completion is a technique for completing a missing region using a known region. With the arrival of the era of large data, an increasing number of images miss a few important regions in the process of transmission and preservation. The way to complete such missing regions effectively has thus become an important topic in recent years and has...

Clustering is an unsupervised approach to classify elements based on their similarity, and it is used to find the intrinsic patterns of data. There are enormous applications of clustering in bioinformatics, pattern recognition, and astronomy. This paper presents a clustering approach based on the idea that density wise single or multiple connected...

In a context sensing system in which a sensor-equipped mobile phone runs an unreliable context-aware application, the application can infer the user’s contexts, based on which it provides personalized services. However, the application may sell the user’s contexts to some malicious adversaries to earn extra profits, which will hinder its widespread...

Due to the complexity and volume, outsourcing ciphertexts to a cloud is deemed to be one of the most effective approaches for big data storage and access. Nevertheless, verifying the access legitimacy of a user and securely updating a ciphertext in the cloud based on a new access policy designated by the data owner are two critical challenges to ma...

In this paper, we consider the problem of cooperative spectrum sensing scheduling (C3S) in a cognitive radio network (CRN) when there exist multiple primary channels. Our work focuses on a scenario in which each secondary user (SU) has the freedom to decide whether to participate in cooperative spectrum sensing; if not, the SU becomes a free rider....

Crowdsensing has been earning increasing credits for effectively integrating the mass sensors to achieve significant tasks that one single sensor cannot imagine. However in many existing works in this field, some key information of the participants is incomplete to each other, hence causing the non-optimality result. Noticing that a potential coope...

Image composition is the process of first extracting a candidate region from a candidate image and then embedding this region into a target image. Traditional composite methods focus on reducing appearance gaps (boundary, brightness, color, and sharpness) between the candidate region and the target image. However, in the composite process, low-qual...

In this paper, we propose an extensible and flexible truthful auction framework that is individually-rational and selfcollusion resistant. By properly setting one simple parameter, this framework yields efficient auctions (like VCG), (sub)optimal auctions (like Myerson’s Optimal Mechanism (MOM)), and budget-balanced double auctions; by carefully ch...

Wireless Sensor Network (WSN) is highly vulnerable to security attacks because they consist of numerous resource-constrained devices and communicate via wireless links. It is infeasible to apply complex and resource intensive security schemes that reduce network lifetime. We have explored polynomial distribution based key establishment schemes and...

Clustering by fast search and find of density peaks (CFSFDP) is proposed to cluster the data by finding of density peaks. CFSFDP is based on two assumptions that: a cluster center is a high dense data point as compared to its surrounding neighbors, and it lies at a large distance from other cluster centers. Based on these assumptions, CFSFDP suppor...