Evaluation of the Coherence of Polish Texts Using Neural Network Models

Coherence evaluation of texts falls into a category of natural language processing tasks. The evaluation of texts’ coherence implies the estimation of their semantic and logical integrity; such a feature of a text can be utilized during the solving of multidisciplinary tasks (SEO analysis, medicine area, detection of fake texts, etc.). In this paper, different state-of-the-art coherence evaluation methods based on machine learning models have been analyzed. The investigation of the effectiveness of different methods for the coherence estimation of Polish texts has been performed. The impact of text’s features on the output coherence value has been analyzed using different approaches of a semantic similarity graph. Two neural networks based on LSTM layers and a pre-trained BERT model correspondingly have been designed and trained for the coherence estimation of input texts. The results obtained may indicate that both lexical and semantic components should be taken into account during the coherence evaluation of Polish documents; moreover, it is advisable to analyze corresponding documents in a sentence-by-sentence manner taking into account word order. According to the retrieved accuracy of the proposed neural networks, it can be concluded that suggested models may be used in order to solve typical coherence estimation tasks for a Polish corpus.

DeepInSAR—A Deep Learning Framework for SAR Interferometric Phase Restoration and Coherence Estimation

Over the past decade, using Interferometric Synthetic Aperture Radar (InSAR) remote sensing technology for ground displacement detection has become very successful. However, during the acquisition stage, microwave signals reflected from the ground and received by the satellite are contaminated, for example, due to undesirable material reflectance and atmospheric factors, and there is no clean ground truth to discriminate these noises, which adversely affect InSAR phase computation. Accurate InSAR phase filtering and coherence estimation are crucial for subsequent processing steps. Current methods require expert supervision and expensive runtime to evaluate the quality of intermediate outputs, limiting the usability and scalability in practical applications, such as wide area ground displacement monitoring and predication. We propose a deep convolutional neural network based model DeepInSAR to intelligently solve both phase filtering and coherence estimation problems. We demonstrate our model’s performance using simulated and real data. A teacher-student framework is introduced to handle the issue of missing clean InSAR ground truth. Quantitative and qualitative evaluations show that our teacher-student approach requires less input but can achieve better results than its stack-based teacher method even on new unseen data. The proposed DeepInSAR also outperforms three other top non-stack based methods in time efficiency without human supervision.

Interaural Coherence Estimation for Speech Processing in Reverberant Environment

Interaural coherence is used to quantify the effects of reverberation on speech, and previous studies applied the conventional method using all previous time data in the form of an infinite impulse response filter to estimate interaural coherence. To consider a characteristic of speech that continuously changes over time, this paper proposes a new method of estimating interaural coherence using time data within a finite length of speech, which is called the quasi-steady interval. The length of the quasi-steady interval is determined with various frequency bands, reverberation times, and short-time Fourier transform (STFT) variables through numerical experiment, and it decreased as reverberation time decreased and the frequency increased. In this interval, a diffuse speech, which is an infinite sum of reflected speeches of different propagating paths, is uncorrelated between two microphones apart from each other; thus, the coherence is close to zero. However, a direct speech measured at the two microphones has steady amplitude and phase difference in this internal; thus, the coherence is close to one. Moreover, the new method is the form of a finite impulse response filter that has a linear phase delay or zero phase delay with respect to speech to frequency; thus, the same or zero time delay for each frequency is applied to the power spectral density. Therefore, the coherence estimation of the new method is closer to the ideal value than the conventional one, and the coherence is accurately estimated at the time–frequency bins of direct speech, which is time-varying according to speech variation.

Assessing the Glacier Boundaries in the Qinghai-Tibetan Plateau of China by Multi-Temporal Coherence Estimation with Sentinel-1A InSAR

The sensitivity of synthetic aperture radar (SAR) coherence has been applied in delineating the boundaries of alpine glaciers because it is nearly unaffected by cloud coverage and can collect data day and night. However, very limited work with application of SAR data has been performed for the alpine glaciers in the Qinghai-Tibetan Plateau (QTP) of China. In this study, we attempted to investigate the change of coherence level in alpine glacier zone and access the glacier boundaries in the QTP using time series of Sentinel-1A SAR images. The DaDongkemadi Glacier (DDG) in the central QTP was selected as the study area with land cover mainly classified into wet snow, ice, river outwash and soil land. We utilized 45 Sentinel-1A C-band SAR images collected during October of 2014 through January of 2018 over the DDG to generate time series of interferometric coherence maps, and to further extract the DDG boundaries. Based on the spatiotemporal analysis of coherence values in the selected sampling areas, we first determined the threshold as 0.7 for distinguishing among different ground targets and then extracted the DDG boundaries through threshold-based segmentation and edge detection. The validation was performed by comparing the DDG boundaries extracted from the coherence maps with those extracted from the Sentinel-2B optical image. The testing results show that the wet snow and ice present a relatively low level of coherence (about 0.5), while the river outwash and the soil land present a higher level of coherence (0.8–1.0). It was found that the coherence maps spanning between June and September (i.e., the glacier ablation period) are the most suitable for identifying the snow- and ice-covered areas. When compared with the boundary detected using optical image, the mean value of Jaccard similarity coefficient for the total areas within the DDG boundaries derived from the coherence maps selected around July, August and September reached up to 0.9010. In contrast, the mean value from the coherence maps selected around December was relatively lower (0.8862). However, the coherence maps around December were the most suitable for distinguishing the ice from the river outwash around the DDG terminus, as the river outwash areas could hardly be differentiated from the ice-covered areas from June through September. The correlation analysis performed by using the meteorological data (i.e., air temperature and precipitation records) suggests that the air temperature and precipitation have a more significant influence on the coherence level of the ice and river outwash than the wet snow and soil land. The proposed method, applied efficiently in this study, shows the potential of multi-temporal coherence estimation from the Sentinel-1A mission to access the boundaries of alpine glaciers on a larger scale in the QTP.