Activity analysis in complicated scenes using DFT coefficients of particle trajectories

Modelling activities in crowded scenes is very challenging as object tracking is not robust in complicated scenes and optical flow does not capture long range motion. We propose a novel approach to analyse activities in crowded scenes using a “bag of particle trajectories”. Particle trajectories are extracted from foreground regions within short video clips using particle video, which estimates long range motion in contrast to optical flow which is only concerned with inter-frame motion. Our applications include temporal video segmentation and anomaly detection, and we perform our evaluation on several real-world datasets containing complicated scenes. We show that our approaches achieve state-of-the-art performance for both tasks.

Market orientation in the nonprofit sector

As the nonprofit sector moves into a more competitive environment it is being required by the community to become more efficient and effective. One response is for nonprofit organisations to become market oriented, which is the familiar response in the for-profit sector. Two components of market orientation, that is market segmentation and customer oriented products, fit well within the peculiarities of a nonprofit organisation. This is usually accompanied by the desire to obtain a competitive advantage causes problems for various stakeholders within the organisation. This paper contends that three factors, management, scarcity of resources, and conflict between organisational objectives and market orientation, are major influences on the adoption of a market oriented culture for a nonprofit organisation.

Efficient real-time face detection for high resolution surveillance applications

This paper presents an efficient face detection method suitable for real-time surveillance applications. Improved efficiency is achieved by constraining the search window of an AdaBoost face detector to pre-selected regions. Firstly, the proposed method takes a sparse grid of sample pixels from the image to reduce whole image scan time. A fusion of foreground segmentation and skin colour segmentation is then used to select candidate face regions. Finally, a classifier-based face detector is applied only to selected regions to verify the presence of a face (the Viola-Jones detector is used in this paper). The proposed system is evaluated using 640 x 480 pixels test images and compared with other relevant methods. Experimental results show that the proposed method reduces the detection time to 42 ms, where the Viola-Jones detector alone requires 565 ms (on a desktop processor). This improvement makes the face detector suitable for real-time applications. Furthermore, the proposed method requires 50% of the computation time of the best competing method, while reducing the false positive rate by 3.2% and maintaining the same hit rate.

Cross-Lingual knowledge discovery : Chinese-to-English article linking in Wikipedia

In this paper we examine automated Chinese to English link discovery in Wikipedia and the effects of Chinese segmentation and Chinese to English translation on the hyperlink recommendation. Our experimental results show that the implemented link discovery framework can effectively recommend Chinese-to-English cross-lingual links. The techniques described here can assist bi-lingual users where a particular topic is not covered in Chinese, is not equally covered in both languages, or is biased in one language; as well as for language learning.

Speaker diarization : "who spoke when"

Speaker diarization is the process of annotating an input audio with information that attributes temporal regions of the audio signal to their respective sources, which may include both speech and non-speech events. For speech regions, the diarization system also specifies the locations of speaker boundaries and assign relative speaker labels to each homogeneous segment of speech. In short, speaker diarization systems effectively answer the question of ‘who spoke when’. There are several important applications for speaker diarization technology, such as facilitating speaker indexing systems to allow users to directly access the relevant segments of interest within a given audio, and assisting with other downstream processes such as summarizing and parsing. When combined with automatic speech recognition (ASR) systems, the metadata extracted from a speaker diarization system can provide complementary information for ASR transcripts including the location of speaker turns and relative speaker segment labels, making the transcripts more readable. Speaker diarization output can also be used to localize the instances of specific speakers to pool data for model adaptation, which in turn boosts transcription accuracies. Speaker diarization therefore plays an important role as a preliminary step in automatic transcription of audio data. The aim of this work is to improve the usefulness and practicality of speaker diarization technology, through the reduction of diarization error rates. In particular, this research is focused on the segmentation and clustering stages within a diarization system. Although particular emphasis is placed on the broadcast news audio domain and systems developed throughout this work are also trained and tested on broadcast news data, the techniques proposed in this dissertation are also applicable to other domains including telephone conversations and meetings audio. Three main research themes were pursued: heuristic rules for speaker segmentation, modelling uncertainty in speaker model estimates, and modelling uncertainty in eigenvoice speaker modelling. The use of heuristic approaches for the speaker segmentation task was first investigated, with emphasis placed on minimizing missed boundary detections. A set of heuristic rules was proposed, to govern the detection and heuristic selection of candidate speaker segment boundaries. A second pass, using the same heuristic algorithm with a smaller window, was also proposed with the aim of improving detection of boundaries around short speaker segments. Compared to single threshold based methods, the proposed heuristic approach was shown to provide improved segmentation performance, leading to a reduction in the overall diarization error rate. Methods to model the uncertainty in speaker model estimates were developed, to address the difficulties associated with making segmentation and clustering decisions with limited data in the speaker segments. The Bayes factor, derived specifically for multivariate Gaussian speaker modelling, was introduced to account for the uncertainty of the speaker model estimates. The use of the Bayes factor also enabled the incorporation of prior information regarding the audio to aid segmentation and clustering decisions. The idea of modelling uncertainty in speaker model estimates was also extended to the eigenvoice speaker modelling framework for the speaker clustering task. Building on the application of Bayesian approaches to the speaker diarization problem, the proposed approach takes into account the uncertainty associated with the explicit estimation of the speaker factors. The proposed decision criteria, based on Bayesian theory, was shown to generally outperform their non- Bayesian counterparts.

Vision-based docking using an autonomous surface vehicle

This paper describes the development of a novel vision-based autonomous surface vehicle with the purpose of performing coordinated docking manoeuvres with a target, such as an autonomous underwater vehicle, at the water's surface. The system architecture integrates two small processor units; the first performs vehicle control and implements a virtual force based docking strategy, with the second performing vision-based target segmentation and tracking. Furthermore, the architecture utilises wireless sensor network technology allowing the vehicle to be observed by, and even integrated within an ad-hoc sensor network. Simulated and experimental results are presented demonstrating the autonomous vision- based docking strategy on a proof-of-concept vehicle.

Moving object detection in aerial video based on spatiotemporal saliency

In this paper, the problem of moving object detection in aerial video is addressed. While motion cues have been extensively exploited in the literature, how to use spatial information is still an open problem. To deal with this issue, we propose a novel hierarchical moving target detection method based on spatiotemporal saliency. Temporal saliency is used to get a coarse segmentation, and spatial saliency is extracted to obtain the object’s appearance details in candidate motion regions. Finally, by combining temporal and spatial saliency information, we can get refined detection results. Additionally, in order to give a full description of the object distribution, spatial saliency is detected in both pixel and region levels based on local contrast. Experiments conducted on the VIVID dataset show that the proposed method is efficient and accurate.

Towards coordinated vision-based docking using an Autonomous Surface Vehicle

This paper describes the development and experimental evaluation of a novel vision-based Autonomous Surface Vehicle with the purpose of performing coordinated docking manoeuvres with a target, such as an Autonomous Underwater Vehicle, on the water’s surface. The system architecture integrates two small processor units; the first performs vehicle control and implements a virtual force obstacle avoidance and docking strategy, with the second performing vision-based target segmentation and tracking. Furthermore, the architecture utilises wireless sensor network technology allowing the vehicle to be observed by, and even integrated within an ad-hoc sensor network. The system performance is demonstrated through real-world experiments.

Active fibers: Matching deformable tract templates to diffusion tensor images

Reliable quantitative analysis of white matter connectivity in the brain is an open problem in neuroimaging, with common solutions requiring tools for fiber tracking, tractography segmentation and estimation of intersubject correspondence. This paper proposes a novel, template matching approach to the problem. In the proposed method, a deformable fiber-bundle model is aligned directly with the subject tensor field, skipping the fiber tracking step. Furthermore, the use of a common template eliminates the need for tractography segmentation and defines intersubject shape correspondence. The method is validated using phantom DTI data and applications are presented, including automatic fiber-bundle reconstruction and tract-based morphometry. © 2009 Elsevier Inc. All rights reserved.

Mining Sequential Data – in Search of Segmental Structures

Segmentation is a data mining technique yielding simplified representations of sequences of ordered points. A sequence is divided into some number of homogeneous blocks, and all points within a segment are described by a single value. The focus in this thesis is on piecewise-constant segments, where the most likely description for each segment and the most likely segmentation into some number of blocks can be computed efficiently. Representing sequences as segmentations is useful in, e.g., storage and indexing tasks in sequence databases, and segmentation can be used as a tool in learning about the structure of a given sequence. The discussion in this thesis begins with basic questions related to segmentation analysis, such as choosing the number of segments, and evaluating the obtained segmentations. Standard model selection techniques are shown to perform well for the sequence segmentation task. Segmentation evaluation is proposed with respect to a known segmentation structure. Applying segmentation on certain features of a sequence is shown to yield segmentations that are significantly close to the known underlying structure. Two extensions to the basic segmentation framework are introduced: unimodal segmentation and basis segmentation. The former is concerned with segmentations where the segment descriptions first increase and then decrease, and the latter with the interplay between different dimensions and segments in the sequence. These problems are formally defined and algorithms for solving them are provided and analyzed. Practical applications for segmentation techniques include time series and data stream analysis, text analysis, and biological sequence analysis. In this thesis segmentation applications are demonstrated in analyzing genomic sequences.

Segmenting point-sampled surfaces

Extracting features from point-based representations of geometric surface models is becoming increasingly important for purposes such as model classification, matching, and exploration. In an earlier paper, we proposed a multiphase segmentation process to identify elongated features in point-sampled surface models without the explicit construction of a mesh or other surface representation. The preliminary results demonstrated the strength and potential of the segmentation process, but the resulting segmentations were still of low quality, and the segmentation process could be slow. In this paper, we describe several algorithmic improvements to overcome the shortcomings of the segmentation process. To demonstrate the improved quality of the segmentation and the superior time efficiency of the new segmentation process, we present segmentation results obtained for various point-sampled surface models. We also discuss an application of our segmentation process to extract ridge-separated features in point-sampled surfaces of CAD models.

Machine recognition of printed Kannada text

This paper presents the design of a full fledged OCR system for printed Kannada text. The machine recognition of Kannada characters is difficult due to similarity in the shapes of different characters, script complexity and non-uniqueness in the representation of diacritics. The document image is subject to line segmentation, word segmentation and zone detection. From the zonal information, base characters, vowel modifiers and consonant conjucts are separated. Knowledge based approach is employed for recognizing the base characters. Various features are employed for recognising the characters. These include the coefficients of the Discrete Cosine Transform, Discrete Wavelet Transform and Karhunen-Louve Transform. These features are fed to different classifiers. Structural features are used in the subsequent levels to discriminate confused characters. Use of structural features, increases recognition rate from 93% to 98%. Apart from the classical pattern classification technique of nearest neighbour, Artificial Neural Network (ANN) based classifiers like Back Propogation and Radial Basis Function (RBF) Networks have also been studied. The ANN classifiers are trained in supervised mode using the transform features. Highest recognition rate of 99% is obtained with RBF using second level approximation coefficients of Haar wavelets as the features on presegmented base characters.

OTCYMIST: Otsu-Canny minimal spanning tree for born-digital images

Text segmentation and localization algorithms are proposed for the born-digital image dataset. Binarization and edge detection are separately carried out on the three colour planes of the image. Connected components (CC's) obtained from the binarized image are thresholded based on their area and aspect ratio. CC's which contain sufficient edge pixels are retained. A novel approach is presented, where the text components are represented as nodes of a graph. Nodes correspond to the centroids of the individual CC's. Long edges are broken from the minimum spanning tree of the graph. Pair wise height ratio is also used to remove likely non-text components. A new minimum spanning tree is created from the remaining nodes. Horizontal grouping is performed on the CC's to generate bounding boxes of text strings. Overlapping bounding boxes are removed using an overlap area threshold. Non-overlapping and minimally overlapping bounding boxes are used for text segmentation. Vertical splitting is applied to generate bounding boxes at the word level. The proposed method is applied on all the images of the test dataset and values of precision, recall and H-mean are obtained using different approaches.

Benchmarking recognition results on camera captured word image data sets

We have benchmarked the maximum obtainable recognition accuracy on five publicly available standard word image data sets using semi-automated segmentation and a commercial OCR. These images have been cropped from camera captured scene images, born digital images (BDI) and street view images. Using the Matlab based tool developed by us, we have annotated at the pixel level more than 3600 word images from the five data sets. The word images binarized by the tool, as well as by our own midline analysis and propagation of segmentation (MAPS) algorithm are recognized using the trial version of Nuance Omnipage OCR and these two results are compared with the best reported in the literature. The benchmark word recognition rates obtained on ICDAR 2003, Sign evaluation, Street view, Born-digital and ICDAR 2011 data sets are 83.9%, 89.3%, 79.6%, 88.5% and 86.7%, respectively. The results obtained from MAPS binarized word images without the use of any lexicon are 64.5% and 71.7% for ICDAR 2003 and 2011 respectively, and these values are higher than the best reported values in the literature of 61.1% and 41.2%, respectively. MAPS results of 82.8% for BDI 2011 dataset matches the performance of the state of the art method based on power law transform.

GPU-based normalized cuts for road extraction using satellite imagery

This paper presents a GPU implementation of normalized cuts for road extraction problem using panchromatic satellite imagery. The roads have been extracted in three stages namely pre-processing, image segmentation and post-processing. Initially, the image is pre-processed to improve the tolerance by reducing the clutter (that mostly represents the buildings, vegetation,. and fallow regions). The road regions are then extracted using the normalized cuts algorithm. Normalized cuts algorithm is a graph-based partitioning `approach whose focus lies in extracting the global impression (perceptual grouping) of an image rather than local features. For the segmented image, post-processing is carried out using morphological operations - erosion and dilation. Finally, the road extracted image is overlaid on the original image. Here, a GPGPU (General Purpose Graphical Processing Unit) approach has been adopted to implement the same algorithm on the GPU for fast processing. A performance comparison of this proposed GPU implementation of normalized cuts algorithm with the earlier algorithm (CPU implementation) is presented. From the results, we conclude that the computational improvement in terms of time as the size of image increases for the proposed GPU implementation of normalized cuts. Also, a qualitative and quantitative assessment of the segmentation results has been projected.