Jane You - Academia.edu (original) (raw)

Papers by Jane You

IEEE transactions on pattern analysis and machine intelligence, 2021

This paper studies instance-dependent Positive and Unlabeled (PU) classification, where whether a... more This paper studies instance-dependent Positive and Unlabeled (PU) classification, where whether a positive example will be labeled (indicated by s) is not only related to the class label y, but also depends on the observation x. Therefore, the labeling probability on positive examples is not uniform as previous works assumed, but is biased to some simple or critical data points. To depict the above dependency relationship, a graphical model is built in this paper which further leads to a maximization problem on the induced likelihood function regarding P(s,y|x). By utilizing the well-known EM and Adam optimization techniques, the labeling probability of any positive example P(s=1|y=1,x) as well as the classifier induced by P(y|x) can be acquired. Theoretically, we prove that the critical solution always exists, and is locally unique for linear model if some sufficient conditions are met. Moreover, we upper bound the generalization error for both linear logistic and non-linear networ...

Lecture Notes in Computer Science, 2018

This paper targets the problem of image set-based face verification and identification. Unlike tr... more This paper targets the problem of image set-based face verification and identification. Unlike traditional single media (an image or video) setting, we encounter a set of heterogeneous contents containing orderless images and videos. The importance of each image is usually considered either equal or based on their independent quality assessment. How to model the relationship of orderless images within a set remains a challenge. We address this problem by formulating it as a Markov Decision Process (MDP) in the latent space. Specifically, we first present a dependency-aware attention control (DAC) network, which resorts to actor-critic reinforcement learning for sequential attention decision of each image embedding to fully exploit the rich correlation cues among the unordered images. Moreover, we introduce its sample-efficient variant with off-policy experience replay to speed up the learning process. The pose-guided representation scheme can further boost the performance at the extremes of the pose variation.

Journal of Biomedical Science and Engineering

Recently, biological technology and computer science are of great importance in medical applicati... more Recently, biological technology and computer science are of great importance in medical applications. Since one's breath biomarkers have been proved to be related with diseases, it is possible to detect diseases by analysis of breath samples captured by e-noses. In this paper, a novel medical e-nose system specific to disease diagnosis was used to collect a large-scale breath dataset. Methods for signal processing, feature extracting as well as feature & sensor selection were discussed for detecting diseases on respiratory, metabolic and digestive system. Sequential forward selection is used to select the best combination of sensors and features. The experimental results showed that the proposed system was able to well distinguish healthy samples and samples with different diseases. The results also showed the most significant sensors and features for different tasks, which meets the relationship between diseases and breath biomarkers. By selecting best combination of different sensors and features for different tasks, the e-nose system is shown to be helpful and effective for diseases diagnosis on respiratory, metabolic and digestive system.

We present a new approach to content-based image retrieval by addressing three primary issues: im... more We present a new approach to content-based image retrieval by addressing three primary issues: image indexing, similarity measure, and search methods. The proposed algorithms include: an image data warehousing structure for dynamic image indexing; a statistically based feature selection procedure to form flexible similarity measures in terms of the dominant image features; and a feature component code to facilitate query processing and guide the search for the best matching. The experimental results demonstrate the feasibility and effectiveness of the proposed method.

Pattern Recognition, 2015

Effective and efficient texture feature extraction and classification is an important problem in ... more Effective and efficient texture feature extraction and classification is an important problem in image understanding and recognition. Recently, texton learning based texture classification approaches have been widely studied, where the textons are usually learned via K-means clustering or sparse coding methods. However, the K-means clustering is too coarse to characterize the complex feature space of textures, while sparse texton learning/encoding is time-consuming due to the l 0-norm or l 1-norm minimization. Moreover, these methods mostly compute the texton histogram as the statistical features for classification, which may not be effective enough. This paper presents an effective and efficient texton learning and encoding scheme for texture classification. First, a regularized least square based texton learning method is developed to learn the dictionary of textons class by class. Second, a fast two-step l 2-norm texton encoding method is proposed to code the input texture feature over the concatenated dictionary of all classes. Third, two types of histogram features are defined and computed from the texton encoding outputs: coding coefficients and coding residuals. Finally, the two histogram features are combined for classification via a nearest subspace classifier. Experimental results on the CUReT, KTH TIPS and UIUC datasets demonstrated that the proposed method is very promising, especially when the number of available training samples is limited.

International Computer Science Conference, 1995

This paper presents the development of a real-time system for recognition of textured objects. In... more This paper presents the development of a real-time system for recognition of textured objects. In contrast to current approaches which mostly rely on specialized multiprocessor architectures for fast processing, we use a distributed network architecture to support parallelism and attain real-time performance. In this paper, a new approach to linage matching is proposed as the basis of object localization and

PLoS ONE, 2014

Using statistical textons for texture classification has shown great success recently. The maxima... more Using statistical textons for texture classification has shown great success recently. The maximal response 8 (Statistical_MR8), image patch (Statistical_Joint) and locally invariant fractal (Statistical_Fractal) are typical statistical texton algorithms and state-of-the-art texture classification methods. However, there are two limitations when using these methods. First, it needs a training stage to build a texton library, thus the recognition accuracy will be highly depended on the training samples; second, during feature extraction, local feature is assigned to a texton by searching for the nearest texton in the whole library, which is time consuming when the library size is big and the dimension of feature is high. To address the above two issues, in this paper, three binary texton counterpart methods were proposed, Binary_MR8, Binary_Joint, and Binary_Fractal. These methods do not require any training step but encode local feature into binary representation directly. The experimental results on the CUReT, UIUC and KTH-TIPS databases show that binary texton could get sound results with fast feature extraction, especially when the image size is not big and the quality of image is not poor.

PLoS ONE, 2013

In some large-scale face recognition task, such as driver license identification and law enforcem... more In some large-scale face recognition task, such as driver license identification and law enforcement, the training set only contains one image per person. This situation is referred to as one sample problem. Because many face recognition techniques implicitly assume that several (at least two) images per person are available for training, they cannot deal with the one sample problem. This paper investigates principal component analysis (PCA), Fisher linear discriminant analysis (LDA), and locality preserving projections (LPP) and shows why they cannot perform well in one sample problem. After that, this paper presents four reasons that make one sample problem itself difficult: the small sample size problem; the lack of representative samples; the underestimated intra-class variation; and the overestimated inter-class variation. Based on the analysis, this paper proposes to enlarge the training set based on the inter-class relationship. This paper also extends LDA and LPP to extract features from the enlarged training set. The experimental results show the effectiveness of the proposed method.

Pattern Recognition, 2012

In an imbalanced dataset, the positive and negative classes can be quite different in both size a... more In an imbalanced dataset, the positive and negative classes can be quite different in both size and distribution. This degrades the performance of many feature extraction methods and classifiers. This paper proposes a method for extracting minimum positive and maximum negative features (in terms of absolute value) for imbalanced binary classification. This paper develops two models to yield the feature extractors. Model 1 first generates a set of candidate extractors that can minimize the positive features to be zero, and then chooses the ones among these candidates that can maximize the negative features. Model 2 first generates a set of candidate extractors that can maximize the negative features, and then chooses the ones that can minimize the positive features. Compared with the traditional feature extraction methods and classifiers, the proposed models are less likely affected by the imbalance of the dataset. Experimental results show that these models can perform well when the positive class and negative class are imbalanced in both size and distribution.

In this paper, some results concerning the k-truck problem are produced. First, the algorithms an... more In this paper, some results concerning the k-truck problem are produced. First, the algorithms and their complexity concerning the off-line k-truck problem are discussed. Following that, a lower bound of competitive ratio for the on-line k-truck problem is given. Based on the Position Maintaining Strategy (PMS), we get some new results which are slightly better than those of [1] for general cases. We also use the Partial-Greedy Algorithm (PG) to solve this problem on a special line. Finally, we extend the concepts of the on-line k-truck problem to obtain a new variant: Deeper On-line k-Truck Problem (DTP).

… Intelligence in Image and …, 2007

With the rapid advances in computing and electronic imaging technology, there has been increasing... more With the rapid advances in computing and electronic imaging technology, there has been increasing interest in developing computer aided medical diagnosis systems to improve the medical service for the public. Images of ocular fundus provide crucial observable features for diagnosing many kinds of pathologies such as diabetes, hypertension, and arteriosclerosis. A computer-aided retinal image analysis system can help eye specialists to screen larger populations and produce better evaluation of treatment and more effective clinical study. This paper is focused on the immediate needs for clinical studies on diabetic patients. Our system includes multiple feature extraction, robust retinal vessel segmentation, hierarchical change detection and classification. The output throughout this system will assist doctors to speed up screening large populations for abnormal cases, and facilitate evaluation of treatment for clinical study. I. INTRODUCTION ITH the fast advances in computing technology and computer industry, multimedia data such as digital signal, image, document, audio, graphics, and video have become widely used in different areas. The aim of the development of automatic medical diagnosis systems for medical applications is to provide storage, processing, and communication services required by the medical community effectively and reliably. Reliable and accurate medical diagnosis requires knowledge of changes in different clinical symptoms due to health degeneration and disease deterioration. One of the main critical issues of such systems is the handling of multimedia medical information in a uniform way to analyze medical data accurately and diagnose different diseases reliably. Image processing techniques offer the means to acquire digital information, at different scales, quickly and efficiently. This paper is focused on the immediate needs for clinical studies on diabetic patients. To tackle key issues in image understanding, we propose to investigate, design, analyze, implement and evaluate new algorithms for feature extraction, segmentation, region representation and classification. The proposed system includes extracting multiple image features via wavelet transforms; segmenting

IEEE Transactions on Fuzzy Systems, 2003

This paper presents an agent-based Web-mining approach to Internet shopping. We propose a fuzzy n... more This paper presents an agent-based Web-mining approach to Internet shopping. We propose a fuzzy neural network to tackle the uncertainties in practical shopping activities, such as consumer preferences, product specification, product selection, price negotiation, purchase, delivery, after-sales service and evaluation. The fuzzy neural network provides an automatic and autonomous product classification and selection scheme to support fuzzy decision making by

Full rank principal component analysis (FR-PCA) is a special form of principal component analysis... more Full rank principal component analysis (FR-PCA) is a special form of principal component analysis (PCA) which retains all nonzero components of PCA. Generally speaking, it is hard to estimate how the accuracy of a classi¯er will change after data are compressed by PCA. However, this paper reveals an interesting fact that the transformation by FR-PCA does not change the accuracy of many well-known classi¯cation algorithms. It predicates that people can safely use FR-PCA as a preprocessing tool to compress high-dimensional data without deteriorating the accuracies of these classi¯ers. The main contribution of the paper is that it theoretically proves that the transformation by FR-PCA does not change accuracies of the k nearest neighbor, the minimum distance, support vector machine, large margin linear projection, and maximum scatter di®erence classi¯ers. In addition, through extensive experimental studies conducted on several benchmark face image databases, this paper demonstrates that FR-PCA can greatly promote the e±ciencies of above-mentioned¯ve classi¯cation algorithms in appearance-based face recognition.

Existing approaches to holistic appearance based face recognition require a high dimensional feat... more Existing approaches to holistic appearance based face recognition require a high dimensional feature space to attain fruitful performance. We have proposed a relatively low feature dimensional scheme to deal with the face recognition problem. We use the aggregated responses of 2D Gabor filters to represent face images. We have investigated the effect of "duplicate" images and the effect of facial expressions. Our results show that the proposed method is more robust than the PCA-based method under varying facial expressions, especially in recognizing "duplicate" images.

This paper presents an efficient and effective approach to personal identification by parallel bi... more This paper presents an efficient and effective approach to personal identification by parallel biometrics computing using mobile agents. To overcome the limitations of the existing password-based authentication services on the Internet, we integrate multiple personal features including fingerprints, palmprints, hand geometry and face into a hierarchical structure for fast and reliable personal identification and verification. To increase the speed and flexibility of the process, we use mobile agents as a navigational tool for parallel implementation in a distributed environment, which includes hierarchical biometric feature extraction, multiple feature integration, dynamic biometric data indexing and guided search. To solve the problems associated with bottlenecks and platform dependence, we apply a four-layered structural model and a three-dimensional operational model to achieve high performance. Instead of applying predefined task scheduling schemes to allocate the computing resources, we introduce a new on-line competitive algorithm to guide the dynamic allocation of mobile agents with greater flexibility. The experimental results demonstrate the feasibility and the potential of the proposed method.

Automated line detection is a classical image processing topic with many applications such as roa... more Automated line detection is a classical image processing topic with many applications such as road detection in remote images and vessel detection in medical images. Many traditional line detectors, such as Gabor filter, the second order derivative of Gaussian and Radon transform will response not only to lines but also to edges, e.g. they will give high responses to the edges of bright lines or blobs when only dark lines are required. To reduce false detections when extracting only dark (or bright) lines, in this paper we propose a line detector by using the first derivative of Gaussian. It can detect dark lines without much false detection on blobs or bright lines. Meanwhile, the proposed method can estimate line width simultaneously. Experiments on various images are performed to test the proposed algorithm.

Extracting local rotation invariant features is a popular method for the classification of rotati... more Extracting local rotation invariant features is a popular method for the classification of rotation invariant texture. To address the issue of local rotation invariance, many algorithms based on anisotropic features were proposed. Usually a dominant orientation is found out first, and then anisotropic feature is extracted by this orientation. To validate whether local dominant orientation is necessary for the classification of rotation invariant texture, in this paper, two isotropic statistical texton based methods are proposed. These two methods are the counterparts of two state-of-the-art anisotropic texton based methods: maximum response 8 (MR8) and gray value image patch. Experimental results on three public databases show that local dominant orientation plays an important role when the training set is less; when training samples are enough, local dominant orientation may not be necessary.