International Journal of Pattern Recognition and Artificial Intelligence

In this paper, we propose a novel algorithm for feature selection and model detection using Student's t-distribution based on the variational Bayesian (VB) approach. First, our method is based on the Student's t-mixture model (SMM) which has heavier tail than the Gaussian distribution and is therefore less sensitive to small numbers of data points and consequent precision-estimates of the components number. Second, the number of components, the local feature saliency and the parameters of the mixture model are simultaneously estimated by Bayesian variational learning. Experimental results using synthetic and real data demonstrate the improved robustness of our approach.

Considering it is difficult to guarantee that at least one continuous complete gait cycle is captured in real applications, we address the multi-view gait recognition problem with short probe sequences. With unified multi-view population hidden markov models (umvpHMMs), the gait pattern is represented as fixed-length multi-view stances. By incorporating the multi-stance dynamics, the well-known view transformation model (VTM) is extended into a multi-linear projection model in a four-order tensor space, so that a view-independent stance-independent identity vector (VSIV) can be extracted. The main advantage is that the proposed VSIV is stable for each subject regardless of the camera location or the sequence length. Experiments show that our algorithm achieves encouraging performance for cross-view gait recognition even with short probe sequences.

The present paper describes a novel approach to performing feature extraction and classification in possibly layered circular structures, as seen in two-dimensional cutting planes of three-dimensional tube-shaped objects. The algorithm can therefore be used to analyze histological specimens of blood vessels as well as intravascular ultrasound (IVUS) datasets. The approach uses a radial signal-based extraction of textural features in combination with methods of machine learning to integrate a priori domain knowledge. The algorithm in principle solves a two-dimensional classification problem that is reduced to parallel viable time series analysis. A multiscale approach hereby determines a feature vector for each analysis using either a Wavelet-transform (WT) or a S-transform (ST). The classification is done by methods of machine learning — here support vector machines. A modified marching squares algorithm extracts the polygonal segments for the two-dimensional classification. The accuracy is above 80% even in datasets with a considerable quantity of artifacts, while the mean accuracy is above 90%. The benefit of the approach therefore mainly lies in its robustness, efficient calculation, and the integration of domain knowledge.

The recursive deterministic perceptron (RDP) is a generalization of the single layer perceptron neural network. This neural network can separate, in a deterministic manner, any classification problem (linearly separable or not). It relies on the principle that in any nonlinearly separable (NLS) two-class classification problem, a linearly separable (LS) subset of one or more points belonging to one of the two classes can always be found. Small network topologies can be obtained when the LS subsets are of maximum cardinality. This is referred to as the problem of maximum separability and has been proven to be NP-Complete. Evolutionary computing techniques are applied to handle this problem in a more efficient way than the standard approaches in terms of complexity. These techniques enhance the RDP training in terms of speed of conversion and level of generalization. They provide an alternative to tackle large classification problems which is otherwise not feasible with the algorithmic versions of the RDP training methods.

This paper presents an effective method for writer recognition from offline handwritten documents by extending the idea of codebook-based writer recognition. The study is based on generating two codebooks, primary and secondary. The writer of a handwritten document is characterized by the probability of occurrence of codebook patterns in his/her writing. The main contribution of this study is the proposition of a secondary codebook to capture information on connecting strokes in addition to the main strokes. The writing is divided into small windows and, for each window, four small adjacent windows are considered. The patterns in the main and adjacent windows are clustered separately to generate two codebooks. The proposed method evaluated on 650 writers of the IAM database reports an identification rate of 96% and validates the idea that complementing the primary codebook with a secondary codebook serves to enhance the recognition rates.

This paper presents a scheme of improvement on the local Gaussian distribution fitting energy (LGDF) model in terms of robustness to initialization and noise. The LGDF energy is redefined as a weighted energy integral. The weights are defined based on local entropy deriving from a gray level distribution of local image, which enables the proposed model to be robust to the initialization. Experimental results prove that the proposed model is more robustness to noise than the original LGDF model, local binary fitting (LBF) model and local image fitting (LIF) model.

Nowadays, motion attention model is a controversial topic in the biological computer vision area. The computational attention model can be decomposed into a set of features via predefined channels. Here we designed a bio-inspired vision attention model, and added the rarity measurement onto it. The priority of rarity is emphasized under the assumption of weighting effect upon the features logic fusion. At this stage, a final saliency map at each frame is adjusted by the spatiotemporal and rarity values. By doing this, the process of mimicking human vision attention becomes more realistic and logical to the real circumstance. The experiments are conducted on the benchmark dataset of static images and video sequences. We simulated the attention shift based on several dataset. Most importantly, our dynamic scenes are mostly selected from the objects moving on the highway and dynamic scenes. The former one can be developed on the detection of car collision and will be a useful tool for further application in robotics. We also conduct experiment on the other video clips to prove the rationality of rarity factor and feature cues fusion methods. Finally, the evaluation results indicate our visual attention model outperforms several state-of-the-art motion attention models.

The analysis of tennis data from broadcast video in order to identify tactical information is broadly an unexplored field, and the existing analytical approaches usually consist of simply providing general statistical information. In this paper, we present a tennis model based on Markov decision processes (MDPs), which describes the dynamic interaction between the players and we introduce a novel Monte Carlo-based method with an aim to extract optimal strategic information. In order to test the approach with real tennis data, we also present a system that transforms broadcast video tennis sequences into discrete temporal data that is fed into the model. We show that this framework, based on states, actions and rewards, allows for the identification of optimal strategies.

Conventional algorithms for fingerprint recognition are mainly based on minutiae information. However, the small number of minutiae in partial fingerprints is still a challenge in fingerprint matching. In this paper, a novel algorithm is proposed to improve the performance of partial fingerprint matching. A simulation scheme was firstly proposed to construct a serial of partial fingerprints with different area. Then, the influence of the fingerprint area in partial fingerprint recognition is studied. By comparing the performance of partial fingerprint recognition with different fingerprint area, some useful conclusions can be drawn: (1) The decrease of the fingerprint area degrades the performance of partial fingerprint recognition; (2) When the fingerprint area decreases, the genuine matching scores will decrease, whereas the imposter matching scores will increase. Based on these observations, we proposed a fusion scheme based on modified support vector machine (SVM) to combine the area information for fingerprint recognition. Experimental result illustrates the effectiveness of the proposed method.

A new alternative way named modified improved Runge-Kutta fifth-order five stage (MIRK(5,5)) technique is employed to compute a second-order robot arm problem and also for variations in angles of a robot arm joints model, which needs less number of function evaluations per time step, is the main focus of this paper. Numerically computed outputs match with the exact solution of the industrial robot arm problem which in turn shows that MIRK(5,5) algorithm are better to solve real-time problems practically.