Detection of rain in acoustic recordings of the environment using machine learning techniques

This thesis is concerned with the detection and prediction of rain in environmental recordings using different machine learning algorithms. The results obtained in this research will help ecologists to efficiently analyse environmental data and monitor biodiversity.

Structural damage alarming and localization of cable-supported bridges using multi-novelty indices: A feasibility study

This paper presents a feasibility study on structural damage alarming and localization of long-span cable-supported bridges using multi-novelty indices formulated by monitoring-derived modal parameters. The proposed method which requires neither structural model nor damage model is applicable to structures of arbitrary complexity. With the intention to enhance the tolerance to measurement noise/uncertainty and the sensitivity to structural damage, an improved novelty index is formulated in terms of auto-associative neural networks (ANNs) where the output vector is designated to differ from the input vector while the training of the ANNs needs only the measured modal properties of the intact structure under in-service conditions. After validating the enhanced capability of the improved novelty index for structural damage alarming over the commonly configured novelty index, the performance of the improved novelty index for damage occurrence detection of large-scale bridges is examined through numerical simulation studies of the suspension Tsing Ma Bridge (TMB) and the cable-stayed Ting Kau Bridge (TKB) incurred with different types of structural damage. Then the improved novelty index is extended to formulate multi-novelty indices in terms of the measured modal frequencies and incomplete modeshape components for damage region identification. The capability of the formulated multi-novelty indices for damage region identification is also examined through numerical simulations of the TMB and TKB.

Can perceptual indices estimate physiological strain across a range of environments and metabolic workloads when wearing explosive ordnance disposal and chemical protective clothing?

Objective Explosive ordnance disposal (EOD) often requires technicians to wear multiple protective garments in challenging environmental conditions. The accumulative effect of increased metabolic cost coupled with decreased heat dissipation associated with these garments predisposes technicians to high levels of physiological strain. It has been proposed that a perceptual strain index (PeSI) using subjective ratings of thermal sensation and perceived exertion as surrogate measures of core body temperature and heart rate, may provide an accurate estimation of physiological strain. Therefore, this study aimed to determine if the PeSI could estimate the physiological strain index (PSI) across a range of metabolic workloads and environments while wearing heavy EOD and chemical protective clothing. Methods Eleven healthy males wore an EOD and chemical protective ensemble while walking on a treadmill at 2.5, 4 and 5.5 km·h− 1 at 1% grade in environmental conditions equivalent to wet bulb globe temperature (WBGT) 21, 30 and 37 °C. WBGT conditions were randomly presented and a maximum of three randomised treadmill walking trials were completed in a single testing day. Trials were ceased at a maximum of 60-min or until the attainment of termination criteria. A Pearson's correlation coefficient, mixed linear model, absolute agreement and receiver operating characteristic (ROC) curves were used to determine the relationship between the PeSI and PSI. Results A significant moderate relationship between the PeSI and the PSI was observed [r = 0.77; p < 0.001; mean difference = 0.8 ± 1.1 a.u. (modified 95% limits of agreement − 1.3 to 3.0)]. The ROC curves indicated that the PeSI had a good predictive power when used with two, single-threshold cut-offs to differentiate between low and high levels of physiological strain (area under curve: PSI three cut-off = 0.936 and seven cut-off = 0.841). Conclusions These findings support the use of the PeSI for monitoring physiological strain while wearing EOD and chemical protective clothing. However, future research is needed to confirm the validity of the PeSI for active EOD technicians operating in the field.

Semi-automated annotation of environmental acoustic recordings

Environmental sensors collect massive amounts of audio data. This thesis investigates computational methods to support human analysts in identifying faunal vocalisations from that audio. A series of experiments was conducted to trial the effectiveness of novel user interfaces. This research examines the rapid scanning of spectrograms, decision support tools for users, and cleaning methods for folksonomies. Together, these investigations demonstrate that providing computational support to human analysts increases their efficiency and accuracy; this allows bioacoustics projects to efficiently utilise their valuable human analysts.

Acoustic sensing: Roles and applications in monitoring avian biodiversity

This thesis examined the use of acoustic sensors for monitoring avian biodiversity. Acoustic sensors have the potential to significantly increase the spatial and temporal scale of ecological observations, however acoustic recordings of the environment can be opaque and complex. This thesis developed methods for analysing large volumes of acoustic data to maximise the detection of bird species, and compared the results of acoustic sensor biodiversity surveys with traditional bird survey techniques.

Diffusion indices on magnetic resonance imaging and neuropsychological performance in amnestic mild cognitive impairment

Background: Magnetic resonance diffusion tensor imaging (DTI) shows promise in the early detection of microstructural pathophysiological changes in the brain. Objectives: To measure microstructural differences in the brains of participants with amnestic mild cognitive impairment (MCI) compared with an age-matched control group using an optimised DTI technique with fully automated image analysis tools and to investigate the correlation between diffusivity measurements and neuropsychological performance scores across groups. Methods: 34 participants (17 participants with MCI, 17 healthy elderly adults) underwent magnetic resonance imaging (MRI)-based DTI. To control for the effects of anatomical variation, diffusion images of all participants were registered to standard anatomical space. Significant statistical differences in diffusivity measurements between the two groups were determined on a pixel-by-pixel basis using gaussian random field theory. Results: Significantly raised mean diffusivity measurements (p<0.001) were observed in the left and right entorhinal cortices (BA28), posterior occipital-parietal cortex (BA18 and BA19), right parietal supramarginal gyrus (BA40) and right frontal precentral gyri (BA4 and BA6) in participants with MCI. With respect to fractional anisotropy, participants with MCI had significantly reduced measurements (p<0.001) in the limbic parahippocampal subgyral white matter, right thalamus and left posterior cingulate. Pearson's correlation coefficients calculated across all participants showed significant correlations between neuropsychological assessment scores and regional measurements of mean diffusivity and fractional anisotropy. Conclusions: DTI-based diffusivity measures may offer a sensitive method of detecting subtle microstructural brain changes associated with preclinical Alzheimer's disease.

Acoustic adaptation in cross database audio visual SHMM training for phonetic spoken term detection

Visual information in the form of lip movements of the speaker has been shown to improve the performance of speech recognition and search applications. In our previous work, we proposed cross database training of synchronous hidden Markov models (SHMMs) to make use of external large and publicly available audio databases in addition to the relatively small given audio visual database. In this work, the cross database training approach is improved by performing an additional audio adaptation step, which enables audio visual SHMMs to benefit from audio observations of the external audio models before adding visual modality to them. The proposed approach outperforms the baseline cross database training approach in clean and noisy environments in terms of phone recognition accuracy as well as spoken term detection (STD) accuracy.

Generalised features for bird vocalisation retrieval in acoustic recordings

Bioacoustic monitoring has become a significant research topic for species diversity conservation. Due to the development of sensing techniques, acoustic sensors are widely deployed in the field to record animal sounds over a large spatial and temporal scale. With large volumes of collected audio data, it is essential to develop semi-automatic or automatic techniques to analyse the data. This can help ecologists make decisions on how to protect and promote the species diversity. This paper presents generic features to characterize a range of bird species for vocalisation retrieval. In the implementation, audio recordings are first converted to spectrograms using short-time Fourier transform, then a ridge detection method is applied to the spectrogram for detecting points of interest. Based on the detected points, a new region representation are explored for describing various bird vocalisations and a local descriptor including temporal entropy, frequency bin entropy and histogram of counts of four ridge directions is calculated for each sub-region. To speed up the retrieval process, indexing is carried out and the retrieved results are ranked according to similarity scores. The experiment results show that our proposed feature set can achieve 0.71 in term of retrieval success rate which outperforms spectral ridge features alone (0.55) and Mel frequency cepstral coefficients (0.36).

Clustering acoustic events in environmental recordings for species richness surveys

Environmental acoustic recordings can be used to perform avian species richness surveys, whereby a trained ornithologist can observe the species present by listening to the recording. This could be made more efficient by using computational methods for iteratively selecting the richest parts of a long recording for the human observer to listen to, a process known as “smart sampling”. This allows scaling up to much larger ecological datasets. In this paper we explore computational approaches based on information and diversity of selected samples. We propose to use an event detection algorithm to estimate the amount of information present in each sample. We further propose to cluster the detected events for a better estimate of this amount of information. Additionally, we present a time dispersal approach to estimating diversity between iteratively selected samples. Combinations of approaches were evaluated on seven 24-hour recordings that have been manually labeled by bird watchers. The results show that on average all the methods we have explored would allow annotators to observe more new species in fewer minutes compared to a baseline of random sampling at dawn.

Birdcall retrieval from environmental acoustic recordings using image processing

Acoustic recordings of the environment provide an effective means to monitor bird species diversity. To facilitate exploration of acoustic recordings, we describe a content-based birdcall retrieval algorithm. A query birdcall is a region of spectrogram bounded by frequency and time. Retrieval depends on a similarity measure derived from the orientation and distribution of spectral ridges. The spectral ridge detection method caters for a broad range of birdcall structures. In this paper, we extend previous work by incorporating a spectrogram scaling step in order to improve the detection of spectral ridges. Compared to an existing approach based on MFCC features, our feature representation achieves better retrieval performance for multiple bird species in noisy recordings.

Refractive indices used by the Haag-Streit Lenstar to calculate axial biometric dimensions

PURPOSE To estimate refractive indices used by the Lenstar biometer to translate measured optical path lengths into geometrical path lengths within the eye. METHODS Axial lengths of model eyes were determined using the IOLMaster and Lenstar biometers; comparing those lengths gave an overall eye refractive index estimate for the Lenstar. Using the Lenstar Graphical User Interface, we noticed that boundaries between media could be manipulated and opposite changes in optical path lengths on either side of the boundary could be introduced. Those ratios were combined with the overall eye refractive index to estimate separate refractive indices. Furthermore, Haag-Streit provided us with a template to obtain 'air thicknesses' to compare with geometrical distances. RESULTS The axial length estimates obtained using the IOLMaster and the Lenstar agreed to within 0.01 mm. Estimates of group refractive indices used in the Lenstar were 1.340, 1.341, 1.415, and 1.354 for cornea, aqueous, lens, and overall eye, respectively. Those refractive indices did not match those of schematic eyes, but were close in the cases of aqueous and lens. Linear equations relating air thicknesses to geometrical thicknesses were consistent with our findings. CONCLUSION The Lenstar uses different refractive indices for different ocular media. Some of the refractive indices, such as that for the cornea, are not physiological; therefore, it is likely that the calibrations in the instrument correspond to instrument-specific corrections and are not the real optical path lengths.

Acoustic classification of Australian anurans using syllable features

Acoustic classification of anurans (frogs) has received increasing attention for its promising application in biological and environment studies. In this study, a novel feature extraction method for frog call classification is presented based on the analysis of spectrograms. The frog calls are first automatically segmented into syllables. Then, spectral peak tracks are extracted to separate desired signal (frog calls) from background noise. The spectral peak tracks are used to extract various syllable features, including: syllable duration, dominant frequency, oscillation rate, frequency modulation, and energy modulation. Finally, a k-nearest neighbor classifier is used for classifying frog calls based on the results of principal component analysis. The experiment results show that syllable features can achieve an average classification accuracy of 90.5% which outperforms Mel-frequency cepstral coefficients features (79.0%).

Acoustic feature extraction using perceptual wavelet packet decomposition for frog call classification

Frog protection has become increasingly essential due to the rapid decline of its biodiversity. Therefore, it is valuable to develop new methods for studying this biodiversity. In this paper, a novel feature extraction method is proposed based on perceptual wavelet packet decomposition for classifying frog calls in noisy environments. Pre-processing and syllable segmentation are first applied to the frog call. Then, a spectral peak track is extracted from each syllable if possible. Track duration, dominant frequency and oscillation rate are directly extracted from the track. With k-means clustering algorithm, the calculated dominant frequency of all frog species is clustered into k parts, which produce a frequency scale for wavelet packet decomposition. Based on the adaptive frequency scale, wavelet packet decomposition is applied to the frog calls. Using the wavelet packet decomposition coefficients, a new feature set named perceptual wavelet packet decomposition sub-band cepstral coefficients is extracted. Finally, a k-nearest neighbour (k-NN) classifier is used for the classification. The experiment results show that the proposed features can achieve an average classification accuracy of 97.45% which outperforms syllable features (86.87%) and Mel-frequency cepstral coefficients (MFCCs) feature (90.80%).

Acoustic features for multi-level classification of Australian frogs

Over past few decades, frog species have been experiencing dramatic decline around the world. The reason for this decline includes habitat loss, invasive species, climate change and so on. To better know the status of frog species, classifying frogs has become increasingly important. In this study, acoustic features are investigated for multi-level classification of Australian frogs: family, genus and species, including three families, eleven genera and eighty five species which are collected from Queensland, Australia. For each frog species, six instances are selected from which ten acoustic features are calculated. Then, the multicollinearity between ten features are studied for selecting non-correlated features for subsequent analysis. A decision tree (DT) classifier is used to visually and explicitly determine which acoustic features are relatively important for classifying family, which for genus, and which for species. Finally, a weighted support vector machines (SVMs) classifier is used for the multi- level classification with three most important acoustic features respectively. Our experiment results indicate that using different acoustic feature sets can successfully classify frogs at different levels and the average classification accuracy can be up to 85.6%, 86.1% and 56.2% for family, genus and species respectively.

Compact features for birdcall retrieval from environmental acoustic recordings

Bioacoustic data can be used for monitoring animal species diversity. The deployment of acoustic sensors enables acoustic monitoring at large temporal and spatial scales. We describe a content-based birdcall retrieval algorithm for the exploration of large data bases of acoustic recordings. In the algorithm, an event-based searching scheme and compact features are developed. In detail, ridge events are detected from audio files using event detection on spectral ridges. Then event alignment is used to search through audio files to locate candidate instances. A similarity measure is then applied to dimension-reduced spectral ridge feature vectors. The event-based searching method processes a smaller list of instances for faster retrieval. The experimental results demonstrate that our features achieve better success rate than existing methods and the feature dimension is greatly reduced.