Search

Proper management of the earth's natural resources is imperative to combat further degradation of the natural environment. However, the environmental datasets necessary for informed resource planning and conservation can be costly to collect and annotate. Consequently, there is a lack of publicly available datasets, particularly annotated image datasets relevant for environmental conservation, that can be used for the evaluation of machine learning algorithms to determine their applicability in real-world scenarios. To address this, the Time-evolving Data Science and Artificial Intelligence for Advanced Open Environmental Science (TAIAO) project in New Zealand aims to provide a collection of datasets and accompanying example notebooks for their analysis. This paper showcases three New Zealand-based annotated image datasets that form part of the collection. The first dataset contains annotated images of various predator species, mainly small invasive mammals, taken using low-light camera traps predominantly at night. The second provides aerial photography of the Waikato region in New Zealand, in which stands of Kahikatea (a native New Zealand tree) have been marked up using manual segmentation. The third is a dataset containing orthorectified high-resolution aerial photography, paired with satellite imagery taken by Sentinel-2. Additionally, the TAIAO web platform also contains a collated list of other datasets provided and licensed by our data partners that may be of interest to other researchers. [ABSTRACT FROM AUTHOR]

Labeling mistakes occur in the many real training sets of hyperspectral image (HSI) classification, due to mistakes in the collection of labeled training data samples phase. Label noise is an essential problem in classification, with many possible negative consequences. In this paper, a new method is proposed to detect and delete noisy label samples and therefore remove the effect of errors in the classification process. For this purpose, first, a modified mean shift (MMS) method originated from minimum Bayesian risk is proposed and used to improve the separability of the mislabeled samples from training sets. Then denoised training samples are given to the classification algorithms such as SVM, KNN, MLR, and KOMP to classify the HSI data sets. Then, the effect of several types of loss functions is investigated for the proposed MMS method. Also, the analysis of bounded differences inequality, mean square error (MSE), asymptotic mean square error (AMSE), and asymptotic normality are obtained for our method. The performance of the mentioned compared methods greatly improves at high levels of noise, when the MMS method are combined to them. The obtained experimental results show that the proposed MMS method improved the performance of these classification methods for real HSI data sets. • Labeling mistakes occur in many real training sets of hyperspectral image (HSI) classification. • A generalized mean shift method originated from minimum Bayesian risk can solve this problem. • Denoised training samples are given to the classification algorithms to classify the HSI data sets. • The effect of several types of loss functions are investigated for GMS method. • Analysis of bounded differences inequality, MSE, AMSE, and asymptotic normality are obtained. [ABSTRACT FROM AUTHOR]

Distributed sensing systems for studying scientific phenomena are critical applications of information technologies. By embedding computational intelligence in the environment of study, sensing systems allow researchers to study phenomena at spatial and temporal scales that were previously impossible to achieve. We present an ethnographic study of field research practices among researchers in the Center for Embedded Networked Sensing (CENS), a National Science Foundation Science & Technology Center devoted to developing wireless sensing systems for scientific and social applications. Using the concepts of boundary objects and trading zones, we trace the processes of collaborative research around sensor technology development and adoption within CENS. Over the 10-year lifespan of CENS, sensor technologies, sensor data, field research methods, and statistical expertise each emerged as boundary objects that were understood differently by the science and technology partners. We illustrate how sensing technologies were incompatible with field-based environmental research until researchers 'unearthed' their infrastructures, explicitly reintroducing human skill and expertise into the data collection process and developing new collaborative languages that emphasized building dynamic sensing systems that addressed human needs. In collaborating around a dynamic sensing model, the sensing systems became embedded not in the environment of study, but in the practices of the scientists. [ABSTRACT FROM AUTHOR]

The automatic detection of passive tracers on a moving wavy surface has numerous applications in mathematics and engineering. In oceanography, among many examples, down-looking conventional optical cameras can be employed in wave tanks to investigate the mechanisms of particle dispersion induced by wave turbulence. In this context, we present a computational system to automatically track down the trajectories of several identical passive floaters under the influence of mechanically generated water waves with distinct spectral features. Such functionality is not yet publicly available and a free/open source code within the Python programming environment is described and evaluated. The code is robust and fast, able to automatically identify drifters at 30 fps during approximately 40 seconds, while they remained in the camera's field of view. The algorithm is an efficient tool with many operational and scientific applications. • A computational system to track down the trajectories of identical passive floaters. • The code is robust and fast, able to automatically identify drifters at 30 fps. • The algorithm has several applications, specially in wave tanks. • An efficient way to simultaneously track down several drifters on the water surface. • The results are robust, with high correlation and small errors. [ABSTRACT FROM AUTHOR]