Improving spoken term detection using complementary information

This research has made contributions to the area of spoken term detection (STD), defined as the process of finding all occurrences of a specified search term in a large collection of speech segments. The use of visual information in the form of lip movements of the speaker in addition to audio and the use of topic of the speech segments, and the expected frequency of words in the target speech domain, are proposed. By using these complementary information, improvement in the performance of STD has been achieved which enables efficient search of key words in large collection of multimedia documents.

A Bayesian Framework for the Assessment of Vision-based Weed and Fruit Detection and Classification Algorithms

This paper proposes new metrics and a performance-assessment framework for vision-based weed and fruit detection and classification algorithms. In order to compare algorithms, and make a decision on which one to use fora particular application, it is necessary to take into account that the performance obtained in a series of tests is subject to uncertainty. Such characterisation of uncertainty seems not to be captured by the performance metrics currently reported in the literature. Therefore, we pose the problem as a general problem of scientific inference, which arises out of incomplete information, and propose as a metric of performance the(posterior) predictive probabilities that the algorithms will provide a correct outcome for target and background detection. We detail the framework through which these predicted probabilities can be obtained, which is Bayesian in nature. As an illustration example, we apply the framework to the assessment of performance of four algorithms that could potentially be used in the detection of capsicums (peppers).

On Visual Detection of Highly-occluded Objects for Harvesting Automation in Horticulture

Developing accurate and reliable crop detection algorithms is an important step for harvesting automation in horticulture. This paper presents a novel approach to visual detection of highly-occluded fruits. We use a conditional random field (CRF) on multi-spectral image data (colour and Near-Infrared Reflectance, NIR) to model two classes: crop and background. To describe these two classes, we explore a range of visual-texture features including local binary pattern, histogram of oriented gradients, and learn auto-encoder features. The pro-posed methods are evaluated using hand-labelled images from a dataset captured on a commercial capsicum farm. Experimental results are presented, and performance is evaluated in terms of the Area Under the Curve (AUC) of the precision-recall curves.Our current results achieve a maximum performance of 0.81AUC when combining all of the texture features in conjunction with colour information.

Nonparametric rank regression for analyzing water quality concentration data with multiple detection limits

Environmental data usually include measurements, such as water quality data, which fall below detection limits, because of limitations of the instruments or of certain analytical methods used. The fact that some responses are not detected needs to be properly taken into account in statistical analysis of such data. However, it is well-known that it is challenging to analyze a data set with detection limits, and we often have to rely on the traditional parametric methods or simple imputation methods. Distributional assumptions can lead to biased inference and justification of distributions is often not possible when the data are correlated and there is a large proportion of data below detection limits. The extent of bias is usually unknown. To draw valid conclusions and hence provide useful advice for environmental management authorities, it is essential to develop and apply an appropriate statistical methodology. This paper proposes rank-based procedures for analyzing non-normally distributed data collected at different sites over a period of time in the presence of multiple detection limits. To take account of temporal correlations within each site, we propose an optimal linear combination of estimating functions and apply the induced smoothing method to reduce the computational burden. Finally, we apply the proposed method to the water quality data collected at Susquehanna River Basin in United States of America, which dearly demonstrates the advantages of the rank regression models.

Statistical power calculation and sample size determination for environmental studies with data below detection limits

Power calculation and sample size determination are critical in designing environmental monitoring programs. The traditional approach based on comparing the mean values may become statistically inappropriate and even invalid when substantial proportions of the response values are below the detection limits or censored because strong distributional assumptions have to be made on the censored observations when implementing the traditional procedures. In this paper, we propose a quantile methodology that is robust to outliers and can also handle data with a substantial proportion of below-detection-limit observations without the need of imputing the censored values. As a demonstration, we applied the methods to a nutrient monitoring project, which is a part of the Perth Long-Term Ocean Outlet Monitoring Program. In this example, the sample size required by our quantile methodology is, in fact, smaller than that by the traditional t-test, illustrating the merit of our method.

Perylene-based profluorescent nitroxides for the rapid monitoring of polyester degradation upon weathering: An assessment

A profluorescent nitroxide possessing an isoindoline nitroxide moiety linked to a perylene fluorophore was developed to monitor radical mediated degradation of melamine-formaldehyde crosslinked polyester coil coatings in an industry standard accelerated weathering tester. Trapping of polyester-derived radicals (most likely C-radicals) that are generated during polymer degradation leads to fluorescent closed-shell alkoxy amines, which was used to obtain time-dependent degradation profiles to assess the relative stability of different polyesters towards weathering. The nitroxide probe couples excellent thermal stability and satisfactory photostability with high sensitivity and enables detection of free radical damage in polyesters under conditions that mimic exposure to the environment on a time scale of hours rather than months or years required by other testing methods. There are indications that the profluorescent nitroxide undergoes partial photo-degradation in the absence of polymer-derived radicals. Unexpectedly, it was also found that UV-induced fragmentation of the NO–C bond in closed-shell alkoxy amines leads to regeneration of the profluorescent nitroxide and the respective C-radical. The maximum fluorescence intensity that could be achieved with a given probe concentration is therefore not only determined by the amount of polyester radicals formed during accelerated weathering, but also by the light-driven side reactions of the profluorescent nitroxide and the corresponding alkoxy amine radical trapping products. Studies to determine the optimum probe concentration in the polymer matrix revealed that aggregation and re-absorption effects lowered the fluorescence intensity at higher concentrations of the profluorescent nitroxide, but too low probe concentrations, where these effects would be avoided, were not sufficient to trap the amount of polyester radicals formed upon weathering. The optimized experimental conditions were used to assess the impact of temperature and UV irradiance on polymer degradation during accelerated weathering.

Learning Temporal Alignment Uncertainty for Efficient Event Detection

In this paper we tackle the problem of efficient video event detection. We argue that linear detection functions should be preferred in this regard due to their scalability and efficiency during estimation and evaluation. A popular approach in this regard is to represent a sequence using a bag of words (BOW) representation due to its: (i) fixed dimensionality irrespective of the sequence length, and (ii) its ability to compactly model the statistics in the sequence. A drawback to the BOW representation, however, is the intrinsic destruction of the temporal ordering information. In this paper we propose a new representation that leverages the uncertainty in relative temporal alignments between pairs of sequences while not destroying temporal ordering. Our representation, like BOW, is of a fixed dimensionality making it easily integrated with a linear detection function. Extensive experiments on CK+, 6DMG, and UvA-NEMO databases show significant performance improvements across both isolated and continuous event detection tasks.

Rapid isolation and detection of erythropoietin in blood plasma by magnetic core gold nanoparticles and portable Raman spectroscopy

Isolating, purifying, and identifying proteins in complex biological matrices is often difficult, time consuming, and unreliable. Herein we describe a rapid screening technique for proteins in biological matrices that combines selective protein isolation with direct surface enhanced Raman spectroscopy (SERS) detection. Magnetic core gold nanoparticles were synthesised, characterised, and subsequently functionalized with recombinant human erythropoietin (rHuEPO)-specific antibody. The functionalized nanoparticles were used to capture rHuEPO from horse blood plasma within 15 minutes. The selective binding between the protein and the functionalized nanoparticles was monitored by SERS. The purified protein was then released from the nanoparticles’ surface and directly spectroscopically identified on a commercial nanopillar SERS substrate. ELISA independently confirmed the SERS identification and quantified the released rHuEPO. Finally, the direct SERS detection of the extracted protein was successfully demonstrated for in-field screening by a handheld Raman spectrometer within 1 minute sample measurement time.

Normative values for corneal nerve morphology assessed using corneal confocal microscopy: A multinational normative data set

OBJECTIVE Corneal confocal microscopy is a novel diagnostic technique for the detection of nerve damage and repair in a range of peripheral neuropathies, in particular diabetic neuropathy. Normative reference values are required to enable clinical translation and wider use of this technique. We have therefore undertaken a multicenter collaboration to provide worldwide age-adjusted normative values of corneal nerve fiber parameters. RESEARCH DESIGN AND METHODS A total of 1,965 corneal nerve images from 343 healthy volunteers were pooled from six clinical academic centers. All subjects underwent examination with the Heidelberg Retina Tomograph corneal confocal microscope. Images of the central corneal subbasal nerve plexus were acquired by each center using a standard protocol and analyzed by three trained examiners using manual tracing and semiautomated software (CCMetrics). Age trends were established using simple linear regression, and normative corneal nerve fiber density (CNFD), corneal nerve fiber branch density (CNBD), corneal nerve fiber length (CNFL), and corneal nerve fiber tortuosity (CNFT) reference values were calculated using quantile regression analysis. RESULTS There was a significant linear age-dependent decrease in CNFD (-0.164 no./mm(2) per year for men, P < 0.01, and -0.161 no./mm(2) per year for women, P < 0.01). There was no change with age in CNBD (0.192 no./mm(2) per year for men, P = 0.26, and -0.050 no./mm(2) per year for women, P = 0.78). CNFL decreased in men (-0.045 mm/mm(2) per year, P = 0.07) and women (-0.060 mm/mm(2) per year, P = 0.02). CNFT increased with age in men (0.044 per year, P < 0.01) and women (0.046 per year, P < 0.01). Height, weight, and BMI did not influence the 5th percentile normative values for any corneal nerve parameter. CONCLUSIONS This study provides robust worldwide normative reference values for corneal nerve parameters to be used in research and clinical practice in the study of diabetic and other peripheral neuropathies.

Teledermatology: Its use in the detection and management of actinic keratosis

Teledermatology can profoundly improve access to medical services for those who may have limited access to dermatology due to workforce shortages, distance to providers, or limitations in their mobility. Two common ways of teledermatology are differentiated: life synchronous, where patient and doctor communicate directly, or store and forward asynchronous methods, where the patient and doctor provide and assess the medical information independently. Teledermatology has been tested for its safety, feasibility and accuracy for a number of dermatological conditions, including the early detection of skin cancer and is usually safe, feasible and accurate. Studies reported somewhat better results for synchronous than asynchronous methods, possibly because of loss of information if no direct patient doctor contact is feasible. However asynchronous methods are easier to organize, require less sophisticated technology and are more widely accessible, and are more convenient for both patients and doctors. No study to date focused solely on teledermatology of actinic keratosis, but such lesions are typically found during teledermatology examinations for other main target lesions. In studies where such results were reported, actinic keratoses seemed to be readily identifiable for teledermatologists and adequate management and treatment can be suggested within remote consultations.

Autonomous UAVs wildlife detection using thermal imaging, predictive navigation and computer vision

There is an increased interest on the use of Unmanned Aerial Vehicles (UAVs) for wildlife and feral animal monitoring around the world. This paper describes a novel system which uses a predictive dynamic application that places the UAV ahead of a user, with a low cost thermal camera, a small onboard computer that identifies heat signatures of a target animal from a predetermined altitude and transmits that target’s GPS coordinates. A map is generated and various data sets and graphs are displayed using a GUI designed for easy use. The paper describes the hardware and software architecture and the probabilistic model for downward facing camera for the detection of an animal. Behavioral dynamics of target movement for the design of a Kalman filter and Markov model based prediction algorithm are used to place the UAV ahead of the user. Geometrical concepts and Haversine formula are applied to the maximum likelihood case in order to make a prediction regarding a future state of the user, thus delivering a new way point for autonomous navigation. Results show that the system is capable of autonomously locating animals from a predetermined height and generate a map showing the location of the animals ahead of the user.

MPC controlled multirotor with suspended slung load: System architecture and visual load detection

There is an increased interest in the use of Unmanned Aerial Vehicles for load transportation from environmental remote sensing to construction and parcel delivery. One of the main challenges is accurate control of the load position and trajectory. This paper presents an assessment of real flight trials for the control of an autonomous multi-rotor with a suspended slung load using only visual feedback to determine the load position. This method uses an onboard camera to take advantage of a common visual marker detection algorithm to robustly detect the load location. The load position is calculated using an onboard processor, and transmitted over a wireless network to a ground station integrating MATLAB/SIMULINK and Robotic Operating System (ROS) and a Model Predictive Controller (MPC) to control both the load and the UAV. To evaluate the system performance, the position of the load determined by the visual detection system in real flight is compared with data received by a motion tracking system. The multi-rotor position tracking performance is also analyzed by conducting flight trials using perfect load position data and data obtained only from the visual system. Results show very accurate estimation of the load position (~5% Offset) using only the visual system and demonstrate that the need for an external motion tracking system is not needed for this task.

Decision trees for detection of activity intensity in youth with cerebral palsy

PURPOSE To develop and test decision tree (DT) models to classify physical activity (PA) intensity from accelerometer output and Gross Motor Function Classification System (GMFCS) classification level in ambulatory youth with cerebral palsy (CP); and 2) compare the classification accuracy of the new DT models to that achieved by previously published cut-points for youth with CP. METHODS Youth with CP (GMFCS Levels I - III) (N=51) completed seven activity trials with increasing PA intensity while wearing a portable metabolic system and ActiGraph GT3X accelerometers. DT models were used to identify vertical axis (VA) and vector magnitude (VM) count thresholds corresponding to sedentary (SED) (<1.5 METs), light PA (LPA) (>/=1.5 and <3 METs) and moderate-to-vigorous PA (MVPA) (>/=3 METs). Models were trained and cross-validated using the 'rpart' and 'caret' packages within R. RESULTS For the VA (VA_DT) and VM decision trees (VM_DT), a single threshold differentiated LPA from SED, while the threshold for differentiating MVPA from LPA decreased as the level of impairment increased. The average cross-validation accuracy for the VC_DT was 81.1%, 76.7%, and 82.9% for GMFCS levels I, II, and III, respectively. The corresponding cross-validation accuracy for the VM_DT was 80.5%, 75.6%, and 84.2%, respectively. Within each GMFCS level, the decision tree models achieved better PA intensity recognition than previously published cut-points. The accuracy differential was greatest among GMFCS level III participants, in whom the previously published cut-points misclassified 40% of the MVPA activity trials. CONCLUSION GMFCS-specific cut-points provide more accurate assessments of MVPA levels in youth with CP across the full spectrum of ambulatory ability.

Quick abnormal-bid-detection method for construction contract auctions

Non-competitive bids have recently become a major concern in both Public and Private sector construction contract auctions. Consequently, several models have been developed to help identify bidders potentially involved in collusive practices. However, most of these models require complex calculations and extensive information that is difficult to obtain. The aim of this paper is to utilize recent developments for detecting abnormal bids in capped auctions (auctions with an upper bid limit set by the auctioner) and extend them to the more conventional uncapped auctions (where no such limits are set). To accomplish this, a new method is developed for estimating the values of bid distribution supports by using the solution to what has become known as the German tank problem. The model is then demonstrated and tested on a sample of real construction bid data and shown to detect cover bids with high accuracy. This work contributes to an improved understanding of abnormal bid behavior as an aid to detecting and monitoring potential collusive bid practices.

Prediction of masonry compressive behaviour using a damage mechanics inspired modelling method

Masonry under compression is affected by the properties of its constituents and their interfaces. In spite of extensive investigations of the behaviour of masonry under compression, the information in the literature cannot be regarded as comprehensive due to ongoing inventions of new generation products – for example, polymer modified thin layer mortared masonry and drystack masonry. As comprehensive experimental studies are very expensive, an analytical model inspired by damage mechanics is developed and applied to the prediction of the compressive behaviour of masonry in this paper. The model incorporates a parabolic progressively softening stress-strain curve for the units and a progressively stiffening stress-strain curve until a threshold strain for the combined mortar and the unit-mortar interfaces is reached. The model simulates the mutual constraints imposed by each of these constituents through their respective tensile and compressive behaviour and volumetric changes. The advantage of the model is that it requires only the properties of the constituents and considers masonry as a continuum and computes the average properties of the composite masonry prisms/wallettes; it does not require discretisation of prism or wallette similar to the finite element methods. The capability of the model in capturing the phenomenological behaviour of masonry with appropriate elastic response, stiffness degradation and post peak softening is presented through numerical examples. The fitting of the experimental data to the model parameters is demonstrated through calibration of some selected test data on units and mortar from the literature; the calibrated model is shown to predict the responses of the experimentally determined masonry built using the corresponding units and mortar quite well. Through a series of sensitivity studies, the model is also shown to predict the masonry strength appropriately for changes to the properties of the units and mortar, the mortar joint thickness and the ratio of the height of unit to mortar joint thickness. The unit strength is shown to affect the masonry strength significantly. Although the mortar strength has only a marginal effect, reduction in mortar joint thickness is shown to have a profound effect on the masonry strength. The results obtained from the model are compared with the various provisions in the Australian Masonry Structures Standard AS3700 (2011) and Eurocode 6.