Handwritten Annotation Spotting in Printed Documents Using Top-Down Visual Saliency Models

In this article, we address the problem of localizing text and symbolic annotations on the scanned image of a printed document. Previous approaches have considered the task of annotation extraction as binary classification into printed and handwritten text. In this work, we further subcategorize the annotations as underlines, encirclements, inline text, and marginal text. We have collected a new dataset of 300 documents constituting all classes of annotations marked around or in-between printed text. Using the dataset as a benchmark, we report the results of two saliency formulations—CRF Saliency and Discriminant Saliency, for predicting salient patches, which can correspond to different types of annotations. We also compare our work with recent semantic segmentation techniques using deep models. Our analysis shows that Discriminant Saliency can be considered as the preferred approach for fast localization of patches containing different types of annotations. The saliency models were learned on a small dataset, but still, give comparable performance to the deep networks for pixel-level semantic segmentation. We show that saliency-based methods give better outcomes with limited annotated data compared to more sophisticated segmentation techniques that require a large training set to learn the model.

Developing Deep Learning Models for the Classification of Pediatric Elbow Radiographic Abnormalities of Comparable Performance to Physicians: Strategies for Model Optimization With Small Sized Development Sets

BackgroundTo compare the performance of an AI model based on strategies designed to overcome small sized development sets to pediatric ER physicians at a classification triage task of pediatric elbow radiographs. Methods1,314 pediatric elbow lateral radiographs (mean age: 8.2 years) were retrospectively retrieved, binomially classified based on their annotation as normal or abnormal (with pathology), and randomly partitioned into a development set (993 images), tuning set (109 images), second tuning set (100 images) and test set (112 images). The AI model was trained on the development set and utilized the EfficientNet B1 compound scaling network architecture and online augmentations. Its performance on the test set was compared to a group of five physicians (inter-rater agreement: fair). Statistical analysis: AUC of AI model - DeLong method. Performance of AI model and physician groups - McNemar test. ResultsAccuracy of the model on the test set - 0.804 (95% CI, 0.718 - 0.873), AUROC - 0.872 (95% CI, 0.831 - 0.947). AI model performance compared to the physician group on the test set - sensitivity 0.790 (95% CI 0.684 to 0.895) vs 0.649 (95% CI 0.525 to 0.773), p value 0.088; specificity 0.818 (95% CI 0.716 to 0.920) vs 0.873 (95% CI 0.785 to 0.961), p value 0.439.ConclusionsThe AI model for elbow radiograph triage designed with strategies to optimize performance for a small sized development set showed comparable performance to physicians.

Railway efficiency. Metodological aspects

Improving market competitiveness and economic efficiency was the objective behind the demonopolisation and liberalisation of the railway sector in the European Union. Achieving this objective remains important and crucial to the development of a single rail transport market. The transport performance and financial results of the sector under the new, separative organisational structure of railways in the EU is the result of the action of many different actors, private operators and public entities. This significantly complicates the development of uniform and clear comparable performance evaluation indicators for the sector and makes comparative analyses difficult. Moreover, the specific situation of railways in the EU as a tool for implementing environmental and social policy may conflict with the requirements of financial efficiency. The article presents determinants and methods of measuring railway efficiency proposed by researchers and practitioners.

Efficient Image Super-Resolution via Self-Calibrated Feature Fuse

Recently, many super-resolution reconstruction (SR) feedforward networks based on deep learning have been proposed. These networks enable the reconstructed images to achieve convincing results. However, due to a large amount of computation and parameters, SR technology is greatly limited in devices with limited computing power. To trade-off the network performance and network parameters. In this paper, we propose the efficient image super-resolution network via Self-Calibrated Feature Fuse, named SCFFN, by constructing the self-calibrated feature fuse block (SCFFB). Specifically, to recover the high-frequency detail information of the image as much as possible, we propose SCFFB by self-transformation and self-fusion of features. In addition, to accelerate the network training while reducing the computational complexity of the network, we employ an attention mechanism to elaborate the reconstruction part of the network, called U-SCA. Compared with the existing transposed convolution, it can greatly reduce the computation burden of the network without reducing the reconstruction effect. We have conducted full quantitative and qualitative experiments on public datasets, and the experimental results show that the network achieves comparable performance to other networks, while we only need fewer parameters and computational resources.

Energy Consumption Analysis Model of Typical Cogeneration Systems

At present, China’s urban heating system consumes a lot of energy and is seriously polluted. Our government is working hard to develop urban natural gas regional heating systems to replace traditional coal-fired heating to reduce the serious impact of coal combustion emissions on the urban atmospheric environment during the heating season. On this basis, the characteristics of traditional energy efficiency comparison methods and the problems encountered by these traditional methods in the energy efficiency analysis and application of distributed energy cold, hot and power multigeneration systems in China are analyzed, and the comparable performance efficiency analysis methods suitable for the application of cold, hot and hot power multiple production applications of distributed energy are studied.

Study of GeSn (0.524 eV) Single-Junction Thermophotovoltaic Cells Based on Device Transport Model

Based on the transport equation of the semiconductor device model for 0.524 eV GeSn alloy and the experimental parameters of the material, thermal-electricity conversion performance governed by GeSn diode has been systematically studied in its normal and inverted structure. For the normal p+/n (n+/p) structure, it is demonstrated here that an optimal base doping N d(a) = 3 (7)×1018 cm-3 is observed, and the superior p+/n structure can reach the higher performance. To reduce material consumption, an economical active layer can be comprised of 100-300 nm emitter and 3-6 μm base to attain comparable performance as that for the optimal configuration. The results can offer many useful guidelines for the fabrication of economical GeSn thermophotovoltaic devices.

Production of English lexical stress by Mandarin speakers: Acoustics and kinematics

This study aims to understand if Mandarin late learners of English can successfully manipulate acoustic and kinematic cues to deliver English stress contrast in production. Mandarin ( N = 8) and English ( N = 8) speakers were recorded producing English trochaic (initial stress) and iambic (final stress) items during a nonword repetition task. Speakers’ jaw movement for the utterances was tracked and analysed. Acoustic and kinematic cues were measured for each syllable, including acoustic duration, fundamental frequency (F0), and intensity, as well as jaw movement duration, displacement, peak velocity, and stiffness. Stress ratios (syllable 1 / syllable 2) were calculated for each cue and compared between groups. Results showed that English and Mandarin speakers had generally comparable performance in differentiating trochaic from iambic patterns, as well as in the degree of between-syllable contrast within each pattern. Between-group differences were only observed in acoustic duration and jaw movement velocity/stiffness. These results suggest that the experience with Mandarin stress contributes to Mandarin speakers’ overall successful production of English stress but also results in nonnative use of some acoustic/kinematic cues.

Lightweight Multi-Scale Asymmetric Attention Network for Image Super-Resolution

Recently, with the development of convolutional neural networks, single-image super-resolution (SISR) has achieved better performance. However, the practical application of image super-resolution is limited by a large number of parameters and calculations. In this work, we present a lightweight multi-scale asymmetric attention network (MAAN), which consists of a coarse-grained feature block (CFB), fine-grained feature blocks (FFBs), and a reconstruction block (RB). MAAN adopts multiple paths to facilitate information flow and accomplish a better balance of performance and parameters. Specifically, the FFB applies a multi-scale attention residual block (MARB) to capture richer features by exploiting the pixel-to-pixel correlation feature. The asymmetric multi-weights attention blocks (AMABs) in MARB are designed to obtain the attention maps for improving SISR efficiency and readiness. Extensive experimental results show that our method has comparable performance with fewer parameters than the current advanced lightweight SISR.

Approximating Stackelberg Equilibrium in Anti-UAV Jamming Markov Game with Hierarchical Multi-Agent Deep Reinforcement Learning Algorithm

In order to avoid the malicious jamming of the intelligent unmanned aerial vehicle (UAV) to ground users in the downlink communications, a new anti-UAV jamming strategy based on multi-agent deep reinforcement learning is studied in this paper. In this method, ground users aim to learn the best mobile strategies to avoid the jamming of UAV. The problem is modeled as a Stackelberg game to describe the competitive interaction between the UAV jammer (leader) and ground users (followers). To reduce the computational cost of equilibrium solution for the complex game with large state space, a hierarchical multi-agent proximal policy optimization (HMAPPO) algorithm is proposed to decouple the hybrid game into several sub-Markov games, which updates the actor and critic network of the UAV jammer and ground users at different time scales. Simulation results suggest that the hierarchical multi-agent proximal policy optimization -based anti-jamming strategy achieves comparable performance with lower time complexity than the benchmark strategies. The well-trained HMAPPO has the ability to obtain the optimal jamming strategy and the optimal anti-jamming strategies, which can approximate the Stackelberg equilibrium (SE).

Automated identification of chicken distress vocalisations using deep learning models

The annual global production of chickens exceeds 25 billion birds, and they are often housed in very large groups, numbering thousands. Distress calling triggered by various sources of stress has been suggested as an "iceberg indicator" of chicken welfare. However, to date, the identification of distress calls largely relies on manual annotations, which is very labour-intensive and time-consuming. Thus, a novel light-VGG11 was developed to automatically identify chicken distress calls using recordings (3,363 distress calls and 1,973 natural barn sounds) collected on intensive chicken farms. The light-VGG11 was modified from VGG11 with a significantly smaller size in parameters (9.3 million vs 128 million) and 55.88% faster detection speed while displaying comparable performance, i.e., precision (94.58%), recall (94.89%), F1-score (94.73%), and accuracy (95.07%), therefore more useful for model deployment in practice. To further improve the light-VGG11's performance, we investigated the impacts of different data augmentation techniques (i.e., time masking, frequency masking, mixed spectrograms of the same class, and Gaussian noise) and found that they could improve distress calls detection by up to 1.52%. In terms of precision livestock farming, our research opens new opportunities for developing technologies used to monitor the output of distress calls in large, commercial chicken flocks.