Akinori Hidaka

Selection of Histograms of Oriented Gradients Features for Pedestrian Detection

Histograms of Oriented Gradients (HOG) is one of the well-known features for object recognition. HOG features are calculated by taking orientation histograms of edge intensity in a local region. N.Dalal et al.proposed an object detection algorithm in which HOG features were extracted from all locations of a dense grid on a image region and the combined features are classified by using linear Support Vector Machine (SVM). In this paper, we employ HOG features extracted from all locations of a grid on the image as candidates of the feature vectors. Principal Component Analysis (PCA) is applied to these HOG feature vectors to obtain the score (PCA-HOG) vectors. Then a proper subset of PCA-HOG feature vectors is selected by using Stepwise Forward Selection (SFS) algorithm or Stepwise Backward Selection (SBS) algorithm to improve the generalization performance. The selected PCA-HOG feature vectors are used as an input of linear SVM to classify the given input into pedestrian/non-pedestrian. The improvement of the recognition rates are confirmed through experiments using MIT pedestrian dataset.

Face Tracking by Maximizing Classification Score of Face Detector Based on Rectangle Features

Face tracking continues to be an important topic in computer vision. We describe a tracking algorithm based on a static face detector. Our face detector is a rectanglefeature- based boosted classifier, which outputs the confidence whether an input image is a face. The function that outputs this confidence, called a score function, contains important information about the location of a moving target. A target that has moved will be located in the gradient direction of a score function from the location before moving. Therefore, our tracker will go to the region where the score is maximum using gradient information of this function. We show that this algorithm works by the combination of jumping to the gradient direction and precise search at the local region.

Fast training algorithm by Particle Swarm Optimization and random candidate selection for rectangular feature based boosted detector

Adaboost is an ensemble learning algorithm that combines many base-classifiers to improve their performance. Starting with Viola and Jonespsila researches, Adaboost has often been used to local feature selection for object detection. Adaboost by Viola-Jones consists of following two optimization schemes: (1) training of the local features to make base-classifiers, and (2) selection of the best local feature. Because the number of local features becomes usually more than tens of thousands, the learning algorithm is time consuming if the two optimizations are completely performed. To omit the unnecessary redundancy of the learning, we propose fast boosting algorithms by using Particle Swarm Optimization (PSO) and random candidate selection (RCS). Proposed learning algorithm is 50 times faster than the usual Adaboost while keeping comparable classification accuracy.

Non-Neighboring Rectangular Feature selection using Particle Swarm Optimization

Recently, Viola proposed a rectangular features (RFs) based classifier with high accuracy and rapid processing speed for object detection tasks. In this paper, we propose non-neighboring RFs (NNRFs) as an extension of RFs, and a particle swarm optimization (PSO) based feature selection algorithm for NNRFs. NNRFs are the pairs of arbitrary rectangular sub-regions in images, giving us huge number of candidate NNRFs for feature selection (e.g. 1.3 billion NNRFs in 19×19 pixel image). We show that PSO can select the powerful subset of NNRFs efficiently from the various candidates, and the classification accuracy is improved with the same computational cost as compared with that of Violas method.

Discriminant kernels based support vector machine

Recently the kernel discriminant analysis (KDA) has been successfully applied in many applications. KDA is one of the nonlinear extensions of Linear Discriminant Analysis (LDA). But the kernel function is usually defined a priori and it is not known what the optimum kernel function for nonlinear discriminant analysis is.

Automatic Analysis of Composite Physical Signals Using Non-Negative Factorization and Information Criterion

In time-resolved spectroscopy, composite signal sequences representing energy transfer in fluorescence materials are measured, and the physical characteristics of the materials are analyzed. Each signal sequence is represented by a sum of non-negative signal components, which are expressed by model functions. For analyzing the physical characteristics of a measured signal sequence, the parameters of the model functions are estimated. Furthermore, in order to quantitatively analyze real measurement data and to reduce the risk of improper decisions, it is necessary to obtain the statistical characteristics from several sequences rather than just a single sequence. In the present paper, we propose an automatic method by which to analyze composite signals using non-negative factorization and an information criterion. The proposed method decomposes the composite signal sequences using non-negative factorization subjected to parametric base functions. The number of components (i.e., rank) is also estimated using Akaikes information criterion. Experiments using simulated and real data reveal that the proposed method automatically estimates the acceptable ranks and parameters.

Object Tracking by Maximizing Classification Score of Detector Based on Rectangle Features

In this paper, we propose a novel classifier-based object tracker. Our tracker is the combination of Rectangle Feature (RF) based detector [17], [18] and optical-flow based tracking method [1]. We show that the gradient of extended RFs can be calculated rapidly by using Integral Image method. The proposed tracker was tested on real video sequences. We applied our tracker for face tracking and car tracking experiments. Our tracker worked over 100 fps while maintaining comparable accuracy to RF based detector. Our tracking routine that does not contain image I/O processing can be performed about 500 to 2,500 fps with sufficient tracking accuracy.

Automatic factorization of biological signals by using Boltzmann non-negative matrix factorization

We propose an automatic factorization method for time series signals that follow Boltzmann distribution. Generally time series signals are fitted by using a model function for each sample. To analyze many samples automatically, we have to apply a factorization method. When the energy dynamics are measured in thermal equilibrium, the energy distribution can be modeled by Boltzmann distribution law. The measured signals are factorized as the non-negative sum of the probability density function of Boltzmann distribution. If these signals are composed from several components, then they can be decomposed by using the idea of non-negative matrix factorization (NMF). In this paper, we modify the original NMF to introduce the probability density function modeled by Boltzmann distribution. Also the number of components in samples is estimated by using model selection method. We applied our proposed method to actual data that was measured by fluorescence correlation spectroscopy (FCS). The experimental results show that our method can automatically factorize the signals into the correct components.

Nonlinear Discriminant Analysis Based on Probability Estimation by Gaussian Mixture Model

The Bayesian a posterior probability is a very important element in pattern recognition. In classification problems, the posterior probabilities reflect the uncertainty of assessing an example to particular class. Such residual information will be useful for more deep understanding or analysis of examples. In this paper, we propose a nonlinear discriminant analysis based on the probabilistic estimation of the Gaussian mixture model GMM. We use GMM to estimate the Bayesian a posterior probabilities of any classification problems. Then we use posterior probabilities estimated by GMM to construct discriminative kernel function. The performance of the proposed kernel function is confirmed by several experiments using UCI machine learning repository.

Sparse Logistic Discriminant Analysis

Linear discriminant analysis (LDA) is a well-known method to extract efficient features for multi-class classification. Otsu derived the optimal (ultimate) non-linear discriminant analysis (ONDA) by supposing underlying probabilities and showed that ONDA was closely related to Bayesian decision theory (posterior probabilities). Also Otsu pointed out that the usual LDA could be regarded as the linear approximation of this ultimate ONDA through the linear approximations of the Bayesian posterior probabilities. This theory of ONDA suggests that we can construct a novel nonlinear discriminant mapping by utilizing the estimates of the posterior probabilities. Based on this theory, logistic discriminant analysis (LgDA) was proposed by one of the authors as the approximation of ONDA. In LgDA, the posterior probabilities are estimated by logistic regression. In this paper, we propose the sparse logistic discriminant analysis in which the posterior probabilities are estimated by the sparse logistic regression with L2-or L1-regularizer to improve the generalization performance of LgDA further. Experiments using the standard datasets for classification reveal that the discriminant spaces by our proposed method (LgDA-L2 and LgDA-L1) are better than those by LDA and LgDA in terms of the recognition rates for test samples.