Search

• We work on multiple wild animals detection and pose estimation. • We constructed a large wild animal surveillance dataset. • We proposed a ratio based data distillation framework. Monitoring wildlife populations and activities have significance for biology and ecology. With the rapid development of computer vision and deep learning techniques, it is possible to employ state-of-the-art convoluntional neural network (CNN) based detectors to process the big data from field surveillance cameras and assist in the following studies. However, data labelling during the training stage is a very time-consuming, labour intensive and expensive task. In this paper, we detect multiple animals (Kangaroo, emu, dingo, bird and wildcat) in the wild in an Omni-supervised learning setting. The unlabeled data from the surveillance cameras will be filtered and used for training via data distillation approach. Moreover, we also perform joint pose estimation and detection for Kangaroo which has the most samples in the dataset. To study the feasibility, we also built a large full high definition (HD) wild animal surveillance image dataset from collected data from several national parks across the State of Queensland in Australia and this dataset will be made publicly available. Extensive experiments show that the detection and pose estimation results can be further improved by using unlabeled data wisely. [ABSTRACT FROM AUTHOR]

Digitalization and the use of Smart Farming Technologies are considered a major opportunity for the future of agriculture. However, realisation of full benefits is constrained by: (1) farmers' interest in and use of big data to improve farm decision making; (2) issues of data sovereignty and trust between providers and users of data and technology; (3) institutional arrangements associated with the governance of data platforms. This paper examines the case of Australia's dairy herd milk recording system, arguably one of agriculture's first cases of 'big data' use, which collects, analyses and uses farm-level data (milk production, lactation and breeding records) to provide individual cow and herd performance information, used by individual farmers for farm management decisions. The aim of this study was to 1) examine the use of big data to add value to farm decision making; and 2) explore factors and processes, including institutional arrangements, which influence farmer engagement with and use of big data. This paper traces the Australian history of the organisation of dairy herd recording (established in 1912 and digitalized in late 1970s) and then uses findings from a longitudinal study of 7 case study dairy farms, which were incentivised to become involved in herd recording in 2015. Applying a conceptual framework linking path dependency in farm decision making and collaborative governance capacity, we find three new important dimensions of the farm user context influencing farmer demand for big data applications: 1) the transition to a new business stage; 2) the additionality farmers seek from data generated in one component of the farm system to other subsystems, and 3) the use of data in long term or strategic decision making. Further, we identified critical attributes of support services in addressing digital literacy, capacity and capability issues at farm level, including diversity in data presentation formats and facilitation of the on-farm transition process through intermediary herd test organisations. The role of farmers as governance actors, or citizens in the decisions of the trajectory of big data applications, adds to understanding of the nature of collaborative governance arrangements that support farm engagement. • Farmer interest in data relates to business stage and the decision's importance. • Cooperative governance structures facilitate farmer trust and use of big data. • Support services reduce the impact of low digital literacy, capacity and capability. • Inter-dependence occurs between existing data processes and optimising data use. [ABSTRACT FROM AUTHOR]

Online behavioural advertising (OBA) comes to consumers at a price. Often unknowingly, people deliver up commercially-valuable personal information as a condition of online user experience, functionality and access. Websites are increasingly tracking user behaviours for commercial purposes and social media derives its income largely from data collection and advertising targeted to the personal disclosures and behavioural attributes which are its data-production mainstay. In this context, consumers face a plethora of information collection practices, all designed to generate data analytics including inferential and predictive profiling to create a ‘digital identity’ for OBA purposes. In this subterranean exchange, consumers are economically redefined as data subjects and advertising targets; a reframing which is perhaps why the OBA industry faces a crisis in consumer concern, both as to privacy and trust. This paper proposes that the regulatory control of OBA in Australia is in disarray. Consumer ignorance of online privacy management and OBA practices is demonstrable. Industry transparency, disclosure, consent processes and compliance practices are questionable. Regulator interest is minimal, industry self-regulation is weak and consumer technical ability and personal responsibility is a last fragile line of defence. Data breaches are ubiquitous in a crowded and poorly-audited supply chain, and entail significant adverse consumer consequences. Yet despite these serious concerns, Australian regulators are failing to respond to OBA issues, either through mandating greater industry disclosure or through regulatory action. The author seeks to expose these weaknesses in calling for consumer and privacy regulators to take more meaningful action to better protect consumers' interests online. [ABSTRACT FROM AUTHOR]

• Developing a novel approach for using trip sequences instead of separate trips. • Developing a unique similarity measure for the trip sequences. • Focusing only on the temporal attributes of the trip sequences. • Running the method on real-world case studies. • Evaluating the performance of the developed method. • Comparing the outcome with the existing literature. • Achieving a considerable improvement in inferring the trip purpose. Smart card transactions are known as a rich and continuous source of public transit data, but they miss some important attributes about trips and passengers. One of these missing attributes is the trip purpose attribute. This paper proposes a novel method to infer the trip purpose attribute from the sequences of trips of passengers instead of separate trips. The proposed method infers the trip purpose attribute (a missing attribute in the smart card data) from the temporal attributes (available attributes in the smart card data). First, the relation between the temporal attributes and the trip purpose attribute is learnt by discovering clusters of passengers in the Household Travel Survey dataset while each passenger is represented by one sequence of trips. Then, the discovered clusters are utilized to infer the trip purpose of smart card transactions by allocating each passenger to the closest clusters. The proposed method is implemented on the smart card and HTS datasets from southeast Queensland, Australia. The evaluation results showed a considerable improvement in inferring the trip purpose compared to the results published in the literature. Notably, the effect of considering the trip sequence was more significant than considering land use variables. [ABSTRACT FROM AUTHOR]