Task Of Labelling Research Articles

Farmland Segmentation in Landsat 8 Satellite Images Using Deep Learning and Conditional Generative Adversarial Networks

Leveraging mid-resolution satellite images such as Landsat 8 for accurate farmland segmentation and land change monitoring is crucial for agricultural management, yet is hindered by the scarcity of labelled data for the training of supervised deep learning pipelines. The particular focus of this study is on addressing the scarcity of labelled images. This paper introduces several contributions, including a systematic satellite image data augmentation approach that aims to maintain data population consistency during model training, thus mitigating performance degradation. To alleviate the labour-intensive task of pixel-wise image labelling, we present a novel application of a modified conditional generative adversarial network (CGAN) to generate artificial satellite images and corresponding farm labels. Additionally, we scrutinize the role of spectral bands in satellite image segmentation and compare two prominent semantic segmentation models, U-Net and DeepLabV3+, with diverse backbone structures. Our empirical findings demonstrate that augmenting the dataset with up to 22.85% artificial samples significantly enhances the model performance. Notably, the U-Net model, employing standard convolution, outperforms the DeepLabV3+ models with atrous convolution, achieving a segmentation accuracy of 86.92% on the test data.

SegMix: Co-occurrence Driven Mixup for Semantic Segmentation and Adversarial Robustness

In this paper, we present a strategy for training convolutional neural networks to effectively resolve interference arising from competing hypotheses relating to inter-categorical information throughout the network. In this work, this is accomplished for the task of dense image labelling by blending images based on (i) categorical clustering or (ii) the co-occurrence likelihood of categories. We then train a source separation network which simultaneously segments and separates the blended images. Subsequent feature denoising to suppress noisy activations reveals additional desirable properties and high degrees of successful predictions. Through this process, we reveal a general mechanism, distinct from any prior methods, for boosting the performance of the base segmentation and salient object detection network while simultaneously increasing robustness to adversarial attacks.

Concurrent time-series selections using deep learning and dimension reduction

The objective of this work was to investigate from a user perspective linkage between a 1D time-series view of data and a 2D representation provided by dimension reduction techniques. Our hypothesis is that when such interaction happens seamlessly, the use of these linked views, compared to only interacting with the 1D time-series view, for the ubiquitous task of selection and labelling, is more efficient and effective both in terms of performance and user experience. To this end we examine different dimension reduction techniques (UMAP, t-SNE, PCA and Autoencoder) and evaluate each technique within our experimental setting. Results demonstrate that there is a positive impact on speed and accuracy through augmenting 1D views with a dimension reduction 2D view when these views are linked and linkage is supported through coordinated interaction.

Semantic role labelling using transfer learning model

Semantic role labelling plays a major role in determining the main predicate and related arguments and its relationship to the predicate. These arguments identify the semantic roles in a sentence such as agent, goal, result, etc. Many approaches have been made to perform the task of semantic role labelling. Many recent works incorporated neural networks making use of syntactic features of the given text such as part of speech tags, POS tagged dependency tree, etc. Other works included the use of BERT along with LSTM and MLP to create semantic role labeling models. In this work, BERT model has been fine tuned to be used as a classifier to detect the main predicate, arguments of the predicate and also the relationship arguments maintain with the predicate in the sentence. Using BERT to build a simple classifier creates an efficient semantic role labelling model. Model performance was assessed using metric accuracy, precision, recall and F1 score. This is achieved by comparing the output of each word with the predicted word and labels as well as the arguments in the prediction of the output.

Canonical Segmentation and Syntactic Morpheme Tagging of Four Resource- scarce Nguni Languages

Morphological analysis involves investigating the syntactic class of a word but can also extend to the decomposition and syntactic analysis of its underlying morpheme composition. This is especially relevant to languages with an agglutinative writing system where multiple linguistic words are expressed as a single orthographic word. In this paper, we propose a memory-based approach to canonical segmentation using a windowing approach to recover the uncondensed morphemes that differ from the surface form of a word. Additionally, we propose treating the syntactic labelling of morphemes as a sequence labelling task, similar to part of speech tagging. This approach leverages the internal morpheme composition of a word as local context in much the same way that the surrounding sentence of word serves in the disambiguation of its part-of-speech. Both tasks are modelled separately but performed sequentially by cascading the decomposed morphemes of a word into the task of syntactic labelling. When evaluated on four resource-scarce, conjunctively written Nguni languages, the proposed approach achieves an overall accuracy ranging between 82% and 92% which outperforms previously developed rule-based analysers for the same languages.

Mutual information guided 3D ResNet for self‐supervised video representation learning

In this work, the authors propose a novel self-supervised learning method based on mutual information to learn representations from the videos without manual annotation. Different video clips sampled from the same video usually have coherence in the temporal domain. To guide the network to learn such temporal coherence, they maximise the mutual information between global features extracted from different clips sampled from the same video (Global-MI). However, maximising the Global-MI leads the network to seek shared content from different video clips and may make the network degenerate to focus on the background of the video. Considering the structure of the video, they further maximise the average mutual information between the global feature and local patches of multiple regions of the video clip (multi-region Local-MI). Their approach, which is called Max-GL, learns the temporal coherence by jointly maximising the Global-MI and multi-region Local-MI. Experiments are conducted to validate the effectiveness of the proposed Max-GL. Experimental results show that the Max-GL can serve as an effective pre-training method for the task of action recognition in videos. Additional experiments for the task of action similarity labelling and dynamic scene recognition also validate the generalisation of the learned representations of the Max-GL.

Effects of Spatial Transformer Location on Segmentation Performance of a Dense Transformer Network

Semantic segmentation solves the task of labelling every pixel inan image with its class label, and remains an important unsolvedproblem. While significant work has gone into using deep learningto solve this problem, almost all the existing research uses methodsthat do not make modifications on spatial context considered for thepixel being labelled. Spatial information is an important cue in taskssuch as segmentation, reusing the same spatial span for every pixeland every label may not be the best approach. Spatial TransformerNetworks have shown promising results in improving classificationperformance of existing networks by allowing networks to activelymanipulate their input data to achieve better performance. Our workshows the benefit of incorporating Spatial Transformer Networksand their corresponding decoders into networks tailored to semanticsegmentation. Our experiments show an improvement in performanceover baseline networks when using networks augmentedwith Spatial Transformers.

Unsupervised deep semantic and logical analysis for identification of solution posts from community answers

These days' discussion forums provide dependable solutions to the problems related to multiple domains and areas. However, due to the presence of huge amount of less-informative/inappropriate posts, the identification of the appropriate problem-solution pairs has become a challenging task. The emergence of a variety of topics, domains and areas has made the task of manual labelling of the problem solution-post pairs a very costly and time consuming task. To solve these issues, we concentrate on deep semantic and logical relation between terms. For this, we introduce a novel semantic correlation graph to represent the text. The proposed representation helps us in the identification of topical and semantic relation between terms at a fine grain level. Next, we apply the improved version of personalised pagerank using random walk with restarts. The main aim is to improve the rank score of terms having direct or indirect relation with terms in the given question. Finally, we introduce the use of the node overlapping version of GAAC to find the actual span of answer text. Our experimental results show that the devised system performs better than the existing unsupervised systems.

CONTEXTUAL CLASSIFICATION OF POINT CLOUDS USING A TWO-STAGE CRF

Abstract. In this investigation, we address the task of airborne LiDAR point cloud labelling for urban areas by presenting a contextual classification methodology based on a Conditional Random Field (CRF). A two-stage CRF is set up: in a first step, a point-based CRF is applied. The resulting labellings are then used to generate a segmentation of the classified points using a Conditional Euclidean Clustering algorithm. This algorithm combines neighbouring points with the same object label into one segment. The second step comprises the classification of these segments, again with a CRF. As the number of the segments is much smaller than the number of points, it is computationally feasible to integrate long range interactions into this framework. Additionally, two different types of interactions are introduced: one for the local neighbourhood and another one operating on a coarser scale. This paper presents the entire processing chain. We show preliminary results achieved using the Vaihingen LiDAR dataset from the ISPRS Benchmark on Urban Classification and 3D Reconstruction, which consists of three test areas characterised by different and challenging conditions. The utilised classification features are described, and the advantages and remaining problems of our approach are discussed. We also compare our results to those generated by a point-based classification and show that a slight improvement is obtained with this first implementation.

ChemInform Abstract: MRI Paramagnetic Probes for Cellular Labeling

AbstractReview: ca. 80 refs.

MRI Paramagnetic Probes for Cellular Labeling

AbstractThis review discusses the main procedures that have been developed for pursuing the task of an efficient cellular labelling with paramagnetic metal complexes for MRI detection. Although most of the work deals with GdIII‐based relaxation enhancers it is shown that the emerging class of paraCEST agents can allow the visualization of more cell types in the same MR image. To improve the detection of labelled cells, it is often necessary to deal with the use of nano‐sized carriers able to deliver a large number of paramagnetic metal complexes. Finally, selected examples are reported to show how proper design of the paramagnetic complex can offer the way to acquire key information on underlying processes in the labelled cells.

Are high‐coherent concept maps better for prior knowledge activation? Differential effects of concept mapping tasks on high school vs. university students

Abstract We investigated whether and how prior knowledge activation improves learning outcomes for high school (less experienced learners) and university students (experienced learners) in a hypertext environment. Map coherence was defined as the extent to which relationships between the concepts in the map were made explicit. Therefore, we classified the mapping task of creating and labelling lines as low‐coherent, and the mapping task of labelling provided lines as high‐coherent. Learners were randomly assigned to the conditions of (1) high‐coherent knowledge activation; (2) low‐coherent knowledge activation; and (3) a baseline condition without prior knowledge activation. We found an overall effect for prior knowledge activation, learning experience, and an interaction between learning experience and the coherence of the prior knowledge activation task on learning outcomes. High school students benefited most from labelling provided lines, while university physics majors benefited most from creating and labelling lines. This interaction effect and effects of the specific mapping tasks on process measures support the claim that different prior knowledge activation tasks are suited for different groups of learners.

Application of statistical models to the efficient selection of endmembers by Field Radiometry

An efficient selection by means of Field Radiometry, of a targets set of homogenous spectral response, named endmembers, is going to optimize the methodology of the mixed spectral response modelling, obtained in the registration process of an image from remote sensors in satellites. Additionally, the information obtained through the analysis of its spectral profiles can contribute to facilitate the task of labelling in a thematic classification process in remote sensing. There is little bibliography about the selection process of endmembers from Field Radiometry with the application of statistical models to this process. In this work, the main objective is the establishment of an efficient selection protocol that allows a selection of the endmember, based on the different types of land covers placed in a selected geographical area. This established methodology will be sustained by the application of statistical models basically applied to the phase of sampling. Specifically, a new statistic related to the sampling process and based on t-Student distribution will be shown. Its use will allow to determine the optimal number of samples in each case.