This paper studies the composition in portrait paintings and develops an algorithm to improve the composition of portrait photographs based on example portrait paintings. A study of portrait paintings shows that the placement of the face and the figure is pose-related. Based on this observation, this paper develops an algorithm to improve the composition of a portrait photograph by learning the placement of the face and the figure from an example portrait painting. This example portrait painting is selected based on the similarity of its figure pose to that of the input photograph. This similarity measure is modeled as a graph matching problem. Finally, space cropping is performed using an optimization function to assign a similar location for each body part of the figure in the photograph with that of the figure in the example portrait painting. The experimental results demonstrate the effectiveness of the proposed method. A user study shows that the proposed pose-based composition improvement is preferred more than rule-based methods and learning-based methods.

This paper addresses the aesthetic enhancement of the vignetting effect in photographs to follow such effect organization in example paintings. The example painting is selected based on a comparison of the contextual geometry of the painting and the photograph. Then an algorithm is developed to transfer the lightness weighting extracted from an example painting to a photograph to create the painter-style vignetting effect. Experiments show that the proposed algorithm can successfully transfer the vignetting effect from an example painting to a photograph. Meanwhile, the vignetting effect is more naturally presented with regard to aesthetic composition comparing with popular software tools and camera models.

This paper studies the composition of portrait paintings and develops an algorithm to improve the composition of portrait photographs. The study of portrait paintings shows that placement of the face and the figure in portrait paintings is pose-related. Based on this observation, this paper develops an algorithm to improve the composition of a portrait photograph by learning the placement of the face and the figure from an example portrait painting. The example portrait painting is selected based on the similarity of its figure pose to that of the input photograph. This similarity measure is modeled as a graph matching problem. Finally, space cropping is performed using an optimization function. Experimental results and a user study demonstrate that the proposed pose-based improvement is preferred more than rule-based methods.

In this paper, we present a novel method based on online target-specific metric learning and coherent dynamics estimation for tracklet (track fragment) association by network flow optimization in long-term multi-person tracking. Our proposed framework aims to exploit appearance and motion cues to prevent identity switches during tracking and to recover missed detections. Furthermore, target-specific metrics (appearance cue) and motion dynamics (motion cue) are proposed to be learned and estimated online, i.e., during the tracking process. Our approach is effective even when such cues fail to identify or follow the target due to occlusions or object-to-object interactions. We also propose to learn the weights of these two tracking cues to handle the difficult situations, such as severe occlusions and object-to-object interactions effectively. Our method has been validated on several public datasets and the experimental results show that it outperforms several state-of-the-art tracking methods.

Sparse representation has been popular in visual tracking recently for its robustness and accuracy. However, for most conventional sparse coding based trackers, the target candidates are considered independently between consecutive frames. This paper shows that the temporal correlation of these frames can be exploited to improve the performance of tracking and makes the tracker more robust to noise. Furthermore, to improve the tracking speed, we revisit a more efficient method for ℓ1norm problem, marginal regression, which can solve the sparse coding problem more efficiently. Consequently we can realize real-time tracking based on the temporal smooth sparse representation. Extensive experiments have been done to demonstrate the effectiveness and efficiency of our method.

In this paper, we propose an approach to learn hierarchical features for visual object tracking. First, we offline learn features robust to diverse motion patterns from auxiliary video sequences. The hierarchical features are learned via a two-layer convolutional neural network. Embedding the temporal slowness constraint in the stacked architecture makes the learned features robust to complicated motion transformations, which is important for visual object tracking. Then, given a target video sequence, we propose a domain adaptation module to online adapt the pre-learned features according to the specific target object. The adaptation is conducted in both layers of the deep feature learning module so as to include appearance information of the specific target object. As a result, the learned hierarchical features can be robust to both complicated motion transformations and appearance changes of target objects. We integrate our feature learning algorithm into three tracking methods. Experimental results demonstrate that significant improvement can be achieved using our learned hierarchical features, especially on video sequences with complicated motion transformations.

This paper presents a novel introduction of online target-specific metric learning in track fragment (tracklet) association by network flow optimization for long-term multi-person tracking. Different from other network flow formulation, each node in our network represents a tracklet, and each edge represents the likelihood of neighboring tracklets belonging to the same trajectory as measured by our proposed affinity score. In our method, target-specific similarity metrics are learned, which give rise to the appearance-based models used in the tracklet affinity estimation. Trajectory-based tracklets are refined by using the learned metrics to account for appearance consistency and to identify reliable tracklets. The metrics are then re-learned using reliable tracklets for computing tracklet affinity scores. Long-term trajectories are then obtained through network flow optimization. Occlusions and missed detections are handled by a trajectory completion step. Our method is effective for long-term tracking even when the targets are spatially close or completely occluded by others. We validate our proposed framework on several public datasets and show that it outperforms several state of art methods.

Pedestrian detection is one of the most important task for video analytics of an intelligent surveillance system. In this paper, we propose a framework to improve the detection performance of a generic pedestrian detector for highly crowded scenes. The generic offline-trained pedestrian detectors usually cannot handle the problem of detecting pedestrians in highly crowded scenes due to the severe mutual occlusions of the pedestrians. In our approach, we firstly enhance the head detection and suppress the detections of other body parts in the deformable part-based model because the heads of pedestrians less likely to be occluded in highly crowded scenes. Then we propose to utilize multiple-instance dictionary learning to refine the previous detection responses. Compared to other related work, our approach builds a data-adaptive dictionary (codebook) for the heads of pedestrians, hence it can better handle the problem of detecting pedestrians in highly crowded scenes. The experiments on three datasets containing video clips of crowded scenes demonstrated the effectiveness of our proposed approach, significantly improving the state-of-the-art detector.

Model-free object tracking is still challenging because of the limited prior knowledge and the unexpected variation of the target object. In this paper, we propose a feature learning algorithm for model-free multiple object tracking. First, we pre-learn generic features invariant to diverse motion transformations from auxiliary video data by using a deep network of anto-encoder. Then, we adapt the pre-learned features according to multiple target objects respectively in a multi-task learning manner. We treat the feature adaptation for each target object as one single task. We simultaneously learn the common feature shared by all target objects and the individual feature of each object. Experimental results demonstrate that our feature learning algorithm can significantly improve multiple object tracking performance.

Abstract This paper proposes an approach to enhance the regional contrasts in snap‐shot style portrait photographs by using pre‐modern portrait paintings as aesthetic exemplars. The example portrait painting is selected based on a comparison of the existing contrast properties of the painting with those of the photograph. The contrast organization in the selected example painting is transferred to the photograph by mapping the inter‐ and intra‐regional contrasts of the regions, such as the face and skin areas of the foreground figure, the non‐face/skin part of the foreground and the background region. A piecewise non‐linear transformation curve is used to achieve the contrast mapping. Finally, the transition boundary between regions is smoothed to achieve the final results. The experimental results and user study demonstrate that, by using this proposed approach, the visual appeal of the portrait photographs is effectively improved, and the face and the figure become more salient.

The atmospheric perspective effect is a physical phenomenon relating to the effect that atmosphere has on distant objects, causing them to be lighter and less distinct. The exaggeration of this effect by artists in 2D images increases the illusion of depth thereby making the image more interesting. This paper addresses the enhancement of the atmospheric perspective effect in landscape photographs by the manipulation of depth-aware lightness and saturation contrast values. The form of this manipulation follows the organization of such contrasts in landscape paintings. The objective of this manipulation is based on a statistical study which has clearly shown that the saturation and lightness contrasts inter- and intra- depth planes in paintings are more purposefully organized than those in photographs. This contrast organization in paintings respects the existing contrast relationships within a natural scene guided by the atmospheric perspective effect and also exaggerates them sufficiently with a view to improving the visual appeal of the painting and the illusion of depth within it. The depth-aware lightness and saturation contrasts revealed in landscape paintings guide the mapping of such contrasts in photographs. This contrast mapping is formulated as an optimization problem that simultaneously considers the desired inter-contrast, intra-contrast, and specified gradient constraints. Experimental results demonstrate that by using this proposed approach, both the visual appeal and the illusion of depth in the photographs are effectively improved.

In this paper, we propose a unified energy minimization model for segmentation of non-smooth image structures, e.g., textures, based on Mumford-Shah functional and linear patch model. We consider that image patches of a non-smooth image structure can be modeled by a patch subspace, and image patches of different non-smooth image structures belong to different patch subspaces, which leads to a computational framework for segmentation of non-smooth image structures. Motivated by the Mumford-Shah model, we show that this segmentation framework is equivalent to minimizing a piecewise linear patch reconstruction energy. We also prove that the error of segmentation is bounded by the error of the linear patch reconstruction, meaning that improving the linear patch reconstruction for each region leads to reduction of the segmentation error. In addition, we derive an algorithm for the linear patch reconstruction with proven global optimality and linear rate of convergence. The segmentation in our method is achieved by minimizing a single energy functional without requiring predefined features. Hence, compared with the previous methods that require predefined texture features, our method can be more suitable for handling general textures in unsupervised segmentation. As a by-product, our method also produces a dictionary of optimized orthonormal descriptors for each segmented region. We mainly evaluate our method on the Brodatz textures. The experiments validate our theoretical claims and show the clear superior performance of our methods over other related methods for segmentation of the textures.

This paper addresses the aesthetic enhancement of contrasts in saturation and luminance in a landscape photograph to follow the organization of such contrasts in landscape paintings. Different from many existing work on similar topics that mainly emulate the surface characteristics of a painting, this paper presents a technique that transfer the contrast organization in landscape paintings inter and intra four depth layers onto landscape photographs. The contrasts in saturation and luminance revealed in landscape paintings along the scene depth provide the references to guide the mapping of such contrasts in photographs. A novel inter and intra depth layer luminance and saturation contrast mapping algorithm using gradient histogram matching is developed. Experimental results demonstrate the effectiveness of the proposed method.

In clinical diagnosis, a grade indicating the severity of nuclear cataract is often manually assigned by a trained ophthalmologist to a patient after comparing the lens' opacity severity in his/her slit-lamp images with a set of standard photos. This grading scheme is often subjective and time-consuming. In this paper, a novel computer-aided diagnosis method via ranking is proposed to facilitate nuclear cataract grading following conventional clinical decision-making process. The grade of nuclear cataract in a slit-lamp image is predicted using its neighboring labeled images in a ranked image list, which is achieved using a learned ranking function. This ranking function is learned via direct optimization on a newly proposed approximation to a ranking evaluation measure. Our proposed method has been evaluated by a large dataset composed of 1000 different cases, which are collected from an ongoing clinical population-based study. Both experimental results and comparison with several existing methods demonstrate the benefit of grading via ranking by our proposed method.

Due to the lack of explicit spatial consideration, existing epitome model may fail for image recognition and target detection, which directly motivates us to propose the so-called spatialized epitome in this paper. Extended from the original graphical model of epitome, the spatialized epitome provides a general framework to integrate both appearance and spatial arrangement of patches in the image to achieve a more precise likelihood representation for image(s) and eliminate ambiguities in image reconstruction and recognition. From the extended graphical model of epitome, an EM learning procedure is derived under the framework of variational approximation. The learning procedure can generate an optimized summary of the image appearance with spatial distribution of the similar patches. From the spatialized epitome, we present a principled way of inferring the probability of a new input image under the learnt model and thereby enabling image recognition and target detection. We show how the incorporation of spatial information enhances the epitome's ability for discrimination on several vision tasks, e.g., misalignment/cross-pose face recognition and vehicle detection with a few training samples.

This paper addresses the problem of rigid and nonrigid shape registration and recovery in the presence of shape deformation, missing parts and/or overlapping of multiple shapes. A novel shape evolution approach based on truncated warping transformation formulated in an Energy-Minimization-Curve-Evolution framework is proposed to solve this problem. We deterministically model the rigid and nonrigid shape deformation/registration as curve evolution by a warping function mapping. We also derive the curve evolution equation of warping to minimize functional energies. Hence, by selecting a prior shape as the initial curve for the curve evolution, the shape evolution for registration is within the shape space generated by the warping transformation of the prior shape. Based on the Fourier shape contour spectrum, local shape contour distortions that result in significant visual impact is considered to be largely contained in the changes of the high frequency components. Thus, the smoothing of the warping function by truncation is performed to recover the true shape. We adopted the Chan-Vese model and a truncated warping function to obtain our algorithm for shape registration and recovery. Experiments validated our model and algorithm quantitatively.

Partial Differential Equation(PDE) based curve evolution can stop far away from the stationary solution of the corresponding Euler-Lagrange(EL) equation, which is considered as a Pseudo-Stationary-Phenomenon(PSP)problem in this paper. The consequence of this is early termination of evolution of the deformable curve towards desired object boundary. In this paper, we attempted to provide a solution to this problem. Our contribution is twofold. First, from the theoretical point of view, the PSP problem in PDE based curve evolution is studied and a new geometric flow called the Equilibrium Flow(EF) is proposed to address the problem. We prove that by implementing the EF and the original curve evolution flow alternatively, the stationary solution to the corresponding EL equation can be attained in general. Second, we apply our idea to a general snakes model. We show that there is the PSP problem with a general snakes model. We then solve the PSP problem by applying the derived general solution to the snake model. Experimental results on both synthetic and real images are reported in this paper to demonstrate the PSP problem and illustrate our proposed method to overcome this problem. 1

For a finite set of points lying on a lower dimensional manifold embedded in a high-dimensional data space, algorithms have been developed to study the manifold structure. However, many algorithms will fail if data are noisy. We propose a method based on Gaussian process latent variable models for manifold denoising with the following advantages: (1), it is probabilistic, which naturally handles noise and missing data; (2), it works well for very high dimensional data with small sample size; (3), it can recover the low-dimensional submanifolds corrupted by high-dimensional noise; and (4), it deals well with multimodal manifolds.

Superresolution reconstruction (SR) is a technique for reconstructing a high resolution image from several low resolution ones. Recently, some researchers study the relationship of SR error and magnification factor, and give out the efficient and forbidden magnification range in their published literature. Based on their work, we describe SR error bound in more explicit way and reveal the essential factor affecting SR performance, if infinite low resolution (LR) observations are given. For SR implementation, smooth regularization serves as an efficient tool for producing stable SR solution, but this operator degrades SR performance and produces overly smooth SR image at certain magnifications. To show its mechanism, we augment SR model by integrating regularization operator into it, and then present the curve of SR error bound vs magnification factors. It is proven that the forbidden magnification range no longer exists when the regularization is involved.

Model selection in kernel linear discriminant analysis (KLDA) refers to the selection of appropriate parameters of a kernel function and the regularizer. By following the principle of maximum information preservation, this paper formulates the model selection problem as a problem of selecting an optimal kernel-induced space in which different classes are maximally separated from each other. A scatter-matrix-based criterion is developed to measure the "goodness" of a kernel-induced space, and the kernel parameters are tuned by maximizing this criterion. This criterion is computationally efficient and is differentiable with respect to the kernel parameters. Compared with the leave-one-out (LOO) or k-fold cross validation (CV), the proposed approach can achieve a faster model selection, especially when the number of training samples is large or when many kernel parameters need to be tuned. To tune the regularization parameter in the KLDA, our criterion is used together with the method proposed by Saadi (2004). Experiments on benchmark data sets verify the effectiveness of this model selection approach.

Practical applications call for efficient model selection criteria for multiclass support vector machine (SVM) classification. To solve this problem, this paper develops two model selection criteria by combining or redefining the radius-margin bound used in binary SVMs. The combination is justified by linking the test error rate of a multiclass SVM with that of a set of binary SVMs. The redefinition, which is relatively heuristic, is inspired by the conceptual relationship between the radius-margin bound and the class separability measure. Hence, the two criteria are developed from the perspective of model selection rather than a generalization of the radius-margin bound for multiclass SVMs. As demonstrated by extensive experimental study, the minimization of these two criteria achieves good model selection on most data sets. Compared with the k-fold cross validation which is often regarded as a benchmark, these two criteria give rise to comparable performance with much less computational overhead, particularly when a large number of model parameters are to be optimized.

A new statistical shape prior model is proposed in this paper which is incorporated into geodesic active contours for robust object detection. The object shapes that undergo nonlinear deformable changes are assumed to lie in a low dimensional feature subspace and form clusters after a nonlinear mapping. They are approximated by a probabilistic density model to explore the structure of data distribution. The obtained probability is treated as a shape energy term and is incorporated into geodesic active contour equation to constrain the further curve evolution process. This shape prior model is based on a more sophisticated statistical learning of the training data distribution and thus is more robust in presence of occlusions and cluttered background. Experiments demonstrate its promising detection performance for the intended tasks.

In this paper, a novel offline supervised learning method is proposed to map low-level visual features to high-level semantic concepts for region-based image retrieval. The contributions of this paper lie in three folds. (1) For each semantic concept, a set of low-level tokens are extracted from the segmented regions of training images. Those tokens capture the representative information for describing the semantic meaning of that concept; (2) a set of posteriors are generated based on the low-level tokens through pairwise classification, which denote the probabilities of images belonging to the semantic concepts. The posteriors are treated as high-level features that connect images with high-level semantic concepts. Long-term relevance feedback learning is incorporated to provide the supervisory information needed in the above offline learning process, including the concept information and the relevant training set for each concept; (3) an integrated algorithm is implemented to combine two kinds of information for retrieval: the information from the offline feature-to-concept mapping process and the high-level semantic information from the long-term learned memory. Experimental evaluation on 10,000 images proves the effectiveness of our method.

When detecting abnormalities in colonoscopic images, the location, shape and size of the abnormal regions in the image are unknown and vary across images. It is difficult to determine the appropriate patch-size for patch-based approach. So multi-size patches are used simultaneously to represent the image regions and an ensemble is constructed in which each classifier handles one patch size. The combination of classifiers trained using multiple-size patches can recognize abnormal regions more effectively than only using single-size patches. The classification of the image patches can be performed using a discriminative binary support vector machine (SVM) or a recognition-based one-class SVM. Integration of the two types of SVMs is expected to further improve abnormal region detection. Experimental results show the good performance of our proposed ensemble.

The performance of SVM-based image retrieval is often constrained by the scarcity of training samples. The total number of image samples labeled by users in a retrieval session is very limited, and this small number of labeled samples cannot effectively represent the true distributions of positive and negative image classes, especially for the negative image class. This paper proposes a novel approach to deal with this problem. Instead of treating it as a problem, the mere existence of the small number of labeled images and their desired distribution in the kernel space is considered as prior knowledge from image retrieval to aid the design of the kernel used by SVMs. This is achieved by maximizing a criterion, such as one based on scatter matrices, through gradient-based search methods, incurring very little computational overhead to real-time retrieval process. Experimental results on two benchmark image databases demonstrate the improved retrieval performance by the dynamically designed kernel and hence the effectiveness of the proposed approach for SVM based image retrieval

In this paper, a novel framework is proposed to deliver a fast, robust, and generally applicable SVM-based image retrieval for large databases. A quick test scheme is developed, and on-line kernel learning is employed to realize it after analyzing the relationship between them. Then an upper bound on maximum test scope is derived to speed up testing further. Also, the general applicability is well maintained because this framework does not need a kernel function and index structure to be pre-defined. Taking the advantages of this framework, more sophisticated SVM can be used to improve retrieval performance while keeping short response time. Experimental results on large image databases verify the effectiveness and efficiency of the proposed framework.

A criterion is proposed to optimize the kernel parameters in Kernel-based Biased Discriminant Analysis (KBDA) for image retrieval. Kernel parameter optimization is performed by optimizing the kernel space such that the positive images are well clustered while the negative ones are pushed far away from the positives. The proposed criterion measures the goodness of a kernel space, and the optimal kernel parameter set is obtained by maximizing this criterion. Retrieval experiments on two benchmark image databases demonstrate the effectiveness of proposed criterion for KBDA to achieve the best possible performance at the cost of a small fractional computational overhead.

In this paper, a method of harmonics extraction from Higher Order Statistics (HOS) is developed for texture decomposition. We show that the diagonal slice of the fourth-order cumulants is proportional to the autocorrelation of a related noiseless sinusoidal signal with identical frequencies. We propose to use this fourth-order cumulants slice to estimate a power spectrum from which the harmonic frequencies can be easily extracted. Hence, a texture can be decomposed into deterministic components and indeterministic components as in a unified texture model through a Wold-like decomposition procedure. The simulation and experimental results demonstrated that this method is effective for texture decomposition and it performs better than traditional lower order statistics based decomposition methods.

The performance of image retrieval with SVM active learning is known to be poor when started with few labeled images only. In this paper, the problem is solved by incorporating the unlabelled images into the bootstrapping of the learning process. In this work, the initial SVM classifier is trained with the few labeled images and the unlabelled images randomly selected from the image database. Both theoretical analysis and experimental results show that by incorporating unlabelled images in the bootstrapping, the efficiency of SVM active learning can be improved, and thus improves the overall retrieval performance.

Image retrieval with relevance feedback suffers from the small sample problem. Recently, SVM active learning has been proposed to tackle this problem, showing promising results. However, a small but sufficient number of initially labelled samples are still required to ensure subsequent efficient active learning and good retrieval performance. In the existing method, the user is asked to label more images before active learning starts. In this paper, a method of embedding Euclidean search into SVM active learning is proposed. With the help of Euclidean search, the adverse effect on retrieval performance due to lack of initially labelled samples can be reduced. Experimental results demonstrate the improvement by the proposed method, especially when the number of initially labelled samples is small.

This paper presents a texture analysis method using a newly developed linear compositional texture model. In this model, an image texture is considered to be a linear composition of both structural and random components. By using a Wold-like texture decomposition, an image texture can be decomposed into two orthogonal fields, namely, the deterministic field and the purely indeterministic field. By the concept of linear composition of the two components, we propose the composition ratio of the two components to be presented by the proportion of their spectral energies. The two components are also individually represented by using the multi-channel filtering model with Gabor wavelet and the Gaussian Markov random field model, respectively. The linear compositional texture model thus extends the representation and analysis capacity of the model to deal with a wider variety of image textures.

By considering a texture being composed of two orthogonal components in a unified texture model, the deterministic component and the indeterministic component, a method of harmonics extraction from a Higher Order Statistics (HOS) based spectral decomposition is developed. The method estimates the power spectrum based on the diagonal slice of the fourth-order cumulants From this spectrum, the harmonic frequencies can be easily extracted even for noisy images. The simulation and experimental results indicate that this method is effective for texture decomposition and performs better than the traditional lower order statistics based decomposition methods.

Support vector machines (SVM) have been recently proposed as a new technique for pattern recognition. SVM with a binary tree recognition strategy are used to tackle the face recognition problem. We illustrate the potential of SVM on the Cambridge ORL face database, which consists of 400 images of 40 individuals, containing quite a high degree of variability in expression, pose, and facial details. We also present the recognition experiment on a larger face database of 1079 images of 137 individuals. We compare the SVM-based recognition with the standard eigenface approach using the nearest center classification (NCC) criterion.

This paper proposed a new adaptive texture model applicable to a wide variety of texture types. In this model, the texture field is assumed to be a realization of a regular homogeneous random field and can be decomposed into two components which are indeterministic and deterministic components as in the Wold (1954) texture model. These two components are represented by Gaussian Markov random field (GMRF) model and multichannel filtering model based on the Gabor function (Gabor model), respectively. According to the different ratio of composition from each component in the texture model, an adaptive factor was proposed in the new model. The results of a classification experiment on 81 Brodatz texture images are presented which demonstrated that the new model can better represent a wide variety of textures.

This paper addresses the generic issue of texture image analysis using local spectra features that are based on joint space/spatial-frequency (S/SF) analysis methods. The two-dimensional (2D) Wigner distribution (WD) and its discrete implementation pseudo-Wigner distribution (PWD) are discussed. A set of new local spectral features are calculated from the peaks of the 2D PWD. To assess the feasibility of these features for discriminating different textures by a stable set of feature values, they are applied to texture segmentation using a standard fuzzy c-means clustering algorithm. The obtained results show that PWD allows us to extract the intrinsic features of the texture image, and that using the proposed local spectra features yields satisfactory texture segmentation results.

A new method to extract the minutiae from a gray-level fingerprint image is presented in this paper. The proposed method determines the minutiae by examining the relationship between the ridges and furrows. The minutiae change the relationship between the ridges and furrows, so the minutiae can be detected by finding the relationship change. By following the ridges and furrows along the local direction, the relationship change can be extracted. Experimental results demonstrate that the method developed based on the new approach is insensitive to noise and hence extracts minutiae better.

This paper presents an adaptive texture model for texture classification. In this model, a texture is considered containing both structural and stochastic components. These two components are indeterministic and deterministic parts as in the Wold texture model that are represented by Gaussian Markov random field (GMRF) model and multichannel filtering model based on Gabor function (Gabor model), respectively. According to the different ratio of composition from each component in the texture model, an adaptive factor was proposed for the new adaptive model. Experiments demonstrated that the new adaptive model can better represent a wide variety of textures and hence can lead to better classification results.

This paper addresses the generic issue of textured image analysis using local spectra features that are based on space/spatial frequency analysis methods. The 2D Wigner-distribution and its discrete implementation pseudo-Wigner-distribution (PWD) are discussed. A set of new local spectral features are derived from a simple decorrelation procedure (principal component analysis) of the PWD. In order to assess the feasibility of the features for characterizing local texture properties, texture segmentation experiments were carried out using these features with the help of the fuzzy-c mean clustering algorithm. The segmentation results show that PWD allows one to extract the intrinsic features of texture image regions, and that using the proposed local spectral features yields satisfactory texture segmentation results.

In this paper, an unsupervised optimal fuzzy clustering (UOFC) algorithm is proposed for the classification of images. The main advantage of UOFC is that it can deal with clusters of arbitrary shape, as well as the ability to improve the convergence rate and optimality of the algorithm. In addition, the cluster validity criterion, which combines the properties of the fuzzy membership degrees and cluster geometrical properties, is used in the UOFC algorithm to evaluate the goodness of clustering. The UOFC algorithm is evaluated by simulation data as well as Brodatz textures, represented by the Gabor features.

A novel pattern classification method, called the nearest feature line (NFL), has been proposed by Li (1999). The NFL provides a better alternative to the popular nearest neighbor (NN) classifier when multiple prototypes per class are available. It has been shown to achieve consistently better performance than the NN in terms of the error rate with simulated data as well as real application data. This paper gives a theoretical justification of the NFL. The main result is a proof that the NFL can achieve lower probabilistic error than the NN when the number of available prototypes for each class is finite and the dimension of a feature space is high. A simulation experiment shows that the NFL produces considerably lower error rate than the NN.

A method, the nearest feature line (NFL) method, is used in image classification and retrieval and its performance is evaluated and compared with other methods by extensive experiments. The NFL method is demonstrated to make efficient use of knowledge about multiple prototypes of a class to represent that class.

A hierarchical representation consisting of two level linear combinations (LC) is proposed for face recognition. At the first level, a face image is represented as a linear combination (LC) of a set of basis vectors, i.e. eigenfaces. Thereby a face image corresponds to a feature vector (prototype) in the eigenface space. Normally several such prototypes are available for a face class, each representing the face under a particular condition such as in viewpoint, illumination, and so on. We propose to use the second level LC, that of the prototypes belonging to the same face class, to treat the prototypes coherently. The purpose is to improve face recognition under a new condition not captured by the prototypes by using a linear combination of them. A new distance measure called nearest LC (NLC) is proposed as opposed to the NN. Experiments show that our method yields significantly better results than the one level eigenface methods.

A constrained optimization method, called the Lagrange-Hopfield (LH) method, is presented for solving Markov random field (MRF) based Bayesian image estimation problems for restoration and segmentation. The method combines the augmented Lagrangian multiplier technique with the Hopfield network to solve a constrained optimization problem into which the original Bayesian estimation problem is reformulated. The LH method effectively overcomes instabilities that are inherent in the penalty method (e.g. Hopfield network) or the Lagrange multiplier method in constrained optimization. An additional advantage of the LH method is its suitability for neural-like analog implementation. Experimental results are presented which show that LH yields good quality solutions at reasonable computational costs.