A statistical-driven approach for automatic classification of events in AFL video highlights

Due to the repetitive and lengthy nature, automatic content-based summarization is essential to extract a more compact and interesting representation of sport video. State-of-the art approaches have confirmed that high-level semantic in sport video can be detected based on the occurrences of specific audio and visual features (also known as cinematic). However, most of them still rely heavily on manual investigation to construct the algorithms for highlight detection. Thus, the primary aim of this paper is to demonstrate how the statistics of cinematic features within play-break sequences can be used to less-subjectively construct highlight classification rules. To verify the effectiveness of our algorithms, we will present some experimental results using six AFL (Australian Football League) matches from different broadcasters. At this stage, we have successfully classified each play-break sequence into: goal, behind, mark, tackle, and non-highlight. These events are chosen since they are commonly used for broadcasted AFL highlights. The proposed algorithms have also been tested successfully with soccer video.

Towards universal and statistical-driven heuristics for automatic classification of sports video events

Researchers worldwide have been actively seeking for the most robust and powerful solutions to detect and classify key events (or highlights) in various sports domains. Most approaches have employed manual heuristics that model the typical pattern of audio-visual features within particular sport events To avoid manual observation and knowledge, machine-learning can be used as an alternative approach. To bridge the gaps between these two alternatives, an attempt is made to integrate statistics into heuristic models during highlight detection in our investigation. The models can be designed with a modest amount of domain-knowledge, making them less subjective and more robust for different sports. We have also successfully used a universal scope of detection and a standard set of features that can be applied for different sports that include soccer, basketball and Australian football. An experiment on a large dataset of sport videos, with a total of around 15 hours, has demonstrated the effectiveness and robustness of our
aIlgorithms.

Quantitative experimental comparison of single-beam, sidescan, and multibeam benthic habitat maps

Map comparison is a relatively uncommon practice in acoustic seabed classification to date, contrary to the field of land remote sensing, where it has been developed extensively over recent decades. The aim here is to illustrate the benefits of map comparison in the underwater realm with a case study of three maps independently describing the seabed habitats of the Te Matuku Marine Reserve (Hauraki Gulf, New Zealand). The maps are obtained from a QTC View classification of a single-beam echosounder (SBES) dataset, manual segmentation of a sidescan sonar (SSS) mosaic, and automatic classification of a backscatter dataset from a multibeam echosounder (MBES). The maps are compared using pixel-to-pixel similarity measures derived from the literature in land remote sensing. All measures agree in presenting the MBES and SSS maps as the most similar, and the SBES and SSS maps as the least similar. The results are discussed with reference to the potential of MBES backscatter as an alternative to SSS mosaic for imagery segmentation and to the potential of joint SBES–SSS survey for improved habitat mapping. Other applications of map-similarity measures in acoustic classification of the seabed are suggested.

Quantitative assessment of ADL: a pilot study of upper extremity reaching tasks

Effective telerehabilitation technologies enable patients with certain physiological disabilities to engage in rehabilitative exercises for performing Activities of Daily Living (ADLs). Therefore, training and assessment scenarios for the performance of ADLs are vital for the promotion for telerehabilitation. In this paper we investigate quantitatively and automatically assessing patient's kinematic ability to perform functional upper extremity reaching tasks. The shape of the movement trajectory and the instantaneous acceleration of kinematically crucial body parts, such as wrists, are used to compute the approximate entropy of the motions to represent stability (smoothness) in addition to the duration of the activity. Computer simulations were conducted to illustrate the consistency, sensitivity and robustness of the proposed method. A preliminary experiment with kinematic data captured from healthy subjects mimicking a reaching task with dyskinesia showed a high degree of correlation (Cohen's kappa 0.85 with p < 0.05) between a human observer and the proposed automatic classification tool in terms of assigning the datasets to various levels to represent the subjects' kinematic abilities to perform reaching tasks. This study supported the use of Microsoft Kinect to quantitatively evaluate the ability of individuals with involuntary movements to perform an upper extremity reaching task.

Microarray data classification using automatic SVM kernel selection

Microarray data classification is one of the most important emerging clinical applications in the medical community. Machine learning algorithms are most frequently used to complete this task. We selected one of the state-of-the-art kernel-based algorithms, the support vector machine (SVM), to classify microarray data. As a large number of kernels are available, a significant research question is what is the best kernel for patient diagnosis based on microarray data classification using SVM? We first suggest three solutions based on data visualization and quantitative measures. Different types of microarray problems then test the proposed solutions. Finally, we found that the rule-based approach is most useful for automatic kernel selection for SVM to classify microarray data.

Automatic network traffic classification

The thesis addresses a number of critical problems in regard to fully automating the process of network traffic classification and protocol identification. Several effective solutions based on statistical analysis and machine learning techniques are proposed, which significantly reduce the requirements for human interventions in network traffic classification systems.

Learning from facial aging patterns for automatic age estimation

Age Specific Human-Computer Interaction (ASHCI) has vast potential applications in daily life. However, automatic age estimation technique is still underdeveloped. One of the main reasons is that the aging effects on human faces present several unique characteristics which make age estimation a challenging task that requires non-standard classification approaches. According to the speciality of the facial aging effects, this paper proposes the AGES (AGing pattErn Sub-space) method for automatic age estimation. The basic idea is to model the aging pattern, which is defined as a sequence of personal aging face images, by learning a representative subspace. The proper aging pattern for an unseen face image is then determined by the projection in the subspace that can best reconstruct the face image, while the position of the face image in that aging pattern will indicate its age. The AGES method has shown encouraging performance in the comparative experiments either as an age estimator or as an age range estimator.

Automatic age estimation based on facial aging patterns

While recognition of most facial variations, such as identity, expression, and gender, has been extensively studied, automatic age estimation has rarely been explored. In contrast to other facial variations, aging variation presents several unique characteristics which make age estimation a challenging task. This paper proposes an automatic age estimation method named AGES (AGing pattErn Subspace). The basic idea is to model the aging pattern, which is defined as the sequence of a particular individual's face images sorted in time order, by constructing a representative subspace. The proper aging pattern for a previously unseen face image is determined by the projection in the subspace that can reconstruct the face image with minimum reconstruction error, while the position of the face image in that aging pattern will then indicate its age. In the experiments, AGES and its variants are compared with the limited existing age estimation methods (WAS and AAS) and some well-established classification methods (kNN, BP, C4.5, and SVM). Moreover, a comparison with human perception ability on age is conducted. It is interesting to note that the performance of AGES is not only significantly better than that of all the other algorithms, but also comparable to that of the human observers.

A meta-learning approach to automatic kernel selection for support vector machines

Appropriate choice of a kernel is the most important ingredient of the kernel-based learning methods such as support vector machine (SVM). Automatic kernel selection is a key issue given the number of kernels available, and the current trial-and-error nature of selecting the best kernel for a given problem. This paper introduces a new method for automatic kernel selection, with empirical results based on classification. The empirical study has been conducted among five kernels with 112 different classification problems, using the popular kernel based statistical learning algorithm SVM. We evaluate the kernels’ performance in terms of accuracy measures. We then focus on answering the question: which kernel is best suited to which type of classification problem? Our meta-learning methodology involves measuring the problem characteristics using classical, distance and distribution-based statistical information. We then combine these measures with the empirical results to present a rule-based method to select the most appropriate kernel for a classification problem. The rules are generated by the decision tree algorithm C5.0 and are evaluated with 10 fold cross validation. All generated rules offer high accuracy ratings.

Using decision-tree to automatically construct learned-heuristics for events classification in sports video

Automatic events classification is an essential requirement for constructing an effective sports video summary. It has become a well-known theory that the high-level semantics in sport video can be “computationally interpreted” based on the occurrences of specific audio and visual features which can be extracted automatically. State-of-the-art solutions for features-based event classification have only relied on either manual-knowledge based heuristics or machine learning. To bridge the gaps, we have successfully combined the two approaches by using learning-based heuristics. The heuristics are constructed automatically using decision tree while manual supervision is only required to check the features and highlight contained in each training segment. Thus, fully automated construction of classification system for sports video events has been achieved. A comprehensive experiment on 10 hours video dataset, with five full-match soccer and five full-match basketball videos, has demonstrated the effectiveness/robustness of our algorithms.

Detection and classification of road signs in natural environments

An automatic road sign recognition system first locates road signs within images captured by an imaging sensor on-board of a vehicle, and then identifies the detected road signs. This paper presents an automatic neural-network-based road sign recognition system. First, a study of the existing road sign recognition research is presented. In this study, the issues associated with automatic road sign recognition are described, the existing methods developed to tackle the road sign recognition problem are reviewed, and a comparison of the features of these methods is given. Second, the developed road sign recognition system is described. The system is capable of analysing live colour road scene images, detecting multiple road signs within each image, and classifying the type of road signs detected. The system consists of two modules: detection and classification. The detection module segments the input image in the hue-saturation-intensity colour space, and then detects road signs using a Multi-layer Perceptron neural-network. The classification module determines the type of detected road signs using a series of one to one architectural Multi-layer Perceptron neural networks. Two sets of classifiers are trained using the Resillient-Backpropagation and Scaled-Conjugate-Gradient algorithms. The two modules of the system are evaluated individually first. Then the system is tested as a whole. The experimental results demonstrate that the system is capable of achieving an average recognition hit-rate of 95.96% using the scaled-conjugate-gradient trained classifiers.

Clustering based multi-label classification for image annotation and retrieval

This paper presents a novel multi-label classification framework for domains with large numbers of labels. Automatic image annotation is such a domain, as the available semantic concepts are typically hundreds. The proposed framework comprises an initial clustering phase that breaks the original training set into several disjoint clusters of data. It then trains a multi-label classifier from the data of each cluster. Given a new test instance, the framework first finds the nearest cluster and then applies the corresponding model. Empirical results using two clustering algorithms, four multi-label classification algorithms and three image annotation data sets suggest that the proposed approach can improve the performance and reduce the training time of standard multi-label classification algorithms, particularly in the case of large number of labels.

Automatic extraction of abnormal regions from lung images

A system that could automatically extract abnormal lung regions may assist expert radiologists in verifying lung tissue abnormalities. This paper presents an automated lung nodule detection system consisting of five components: acquisition, pre-processing, background removal, detection, and false positives reduction. The system employs a combination of an ensemble classification and clustering methods. The performance of the developed system is compared against some existing counterparts. Based 011 the experimental results, the proposed system achieved a sensitivity of 100% and a false-positives/slice of 0.67 for 30 tested CT images.

Image to text translation by multi-label classification

This paper presents an image to text translation platform consisting of image segmentation, region features extraction, region blobs clustering, and translation components. A multi-label learning method is suggested for realizing the translation component. Empirical studies show that the predictive performance of the translation component is better than its counterparts when employed a dual-random ensemble multi-label classification algorithm.

Empirical study of multi-label classification methods for image annotation and retrieval

This paper presents an empirical study of multi-label classification methods, and gives suggestions for multi-label classification that are effective for automatic image annotation applications. The study shows that triple random ensemble multi-label classification algorithm (TREMLC) outperforms among its counterparts, especially on scene image dataset. Multi-label k-nearest neighbor (ML-kNN) and binary relevance (BR) learning algorithms perform well on Corel image dataset. Based on the overall evaluation results, examples are given to show label prediction performance for the algorithms using selected image examples. This provides an indication of the suitability of different multi-label classification methods for automatic image annotation under different problem settings.