scholarly journals Scene Complexity: A New Perspective on Understanding the Scene Semantics of Remote Sensing and Designing Image-Adaptive Convolutional Neural Networks

2021 ◽  
Vol 13 (4) ◽  
pp. 742
Author(s):  
Jian Peng ◽  
Xiaoming Mei ◽  
Wenbo Li ◽  
Liang Hong ◽  
Bingyu Sun ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Neural Networks ◽  
Fundamental Problem ◽  
Semantic Representation ◽  
Feature Learning ◽  
Essential Elements ◽  
Complex Scene ◽  
Feature Representations ◽  
The Right ◽  
The Relationship

Scene understanding of remote sensing images is of great significance in various applications. Its fundamental problem is how to construct representative features. Various convolutional neural network architectures have been proposed for automatically learning features from images. However, is the current way of configuring the same architecture to learn all the data while ignoring the differences between images the right one? It seems to be contrary to our intuition: it is clear that some images are easier to recognize, and some are harder to recognize. This problem is the gap between the characteristics of the images and the learning features corresponding to specific network structures. Unfortunately, the literature so far lacks an analysis of the two. In this paper, we explore this problem from three aspects: we first build a visual-based evaluation pipeline of scene complexity to characterize the intrinsic differences between images; then, we analyze the relationship between semantic concepts and feature representations, i.e., the scalability and hierarchy of features which the essential elements in CNNs of different architectures, for remote sensing scenes of different complexity; thirdly, we introduce CAM, a visualization method that explains feature learning within neural networks, to analyze the relationship between scenes with different complexity and semantic feature representations. The experimental results show that a complex scene would need deeper and multi-scale features, whereas a simpler scene would need lower and single-scale features. Besides, the complex scene concept is more dependent on the joint semantic representation of multiple objects. Furthermore, we propose the framework of scene complexity prediction for an image and utilize it to design a depth and scale-adaptive model. It achieves higher performance but with fewer parameters than the original model, demonstrating the potential significance of scene complexity.

scholarly journals Convolutional Neural Networks with Deep Supervised Feature Learning for Remote Sensing Scene Classification

2020 ◽  
Author(s):  
Grigorios Tsagkatakis ◽  
Panagiotis Tsakalides
Keyword(s):  
Remote Sensing ◽  
Feature Learning ◽  
Ground Truth ◽  
Classification Performance ◽  
Cross Entropy ◽  
Scene Classification ◽  
Feature Representations ◽  
Benchmark Datasets ◽  
Low Dimensional ◽  
Fully Connected

State-of-the-art remote sensing scene classification methods employ different Convolutional Neural Network architectures for achieving very high classification performance. A trait shared by the majority of these methods is that the class associated with each example is ascertained by examining the activations of the last fully connected layer, and the networks are trained to minimize the cross-entropy between predictions extracted from this layer and ground-truth annotations. In this work, we extend this paradigm by introducing an additional output branch which maps the inputs to low dimensional representations, effectively extracting additional feature representations of the inputs. The proposed model imposes additional distance constrains on these representations with respect to identified class representatives, in addition to the traditional categorical cross-entropy between predictions and ground-truth. By extending the typical cross-entropy loss function with a distance learning function, our proposed approach achieves significant gains across a wide set of benchmark datasets in terms of classification, while providing additional evidence related to class membership and classification confidence.

Chapter 4. Graph Reasoning Networks and Applications

2021 ◽  
Author(s):  
Qingxing Cao ◽  
Wentao Wan ◽  
Xiaodan Liang ◽  
Liang Lin
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Domain Knowledge ◽  
Feature Learning ◽  
External Information ◽  
Feature Representations ◽  
Domain Specific ◽  
The Neural Network ◽  
Structured Knowledge ◽  
Knowledge Graphs

Despite the significant success in various domains, the data-driven deep neural networks compromise the feature interpretability, lack the global reasoning capability, and can’t incorporate external information crucial for complicated real-world tasks. Since the structured knowledge can provide rich cues to record human observations and commonsense, it is thus desirable to bridge symbolic semantics with learned local feature representations. In this chapter, we review works that incorporate different domain knowledge into the intermediate feature representation.These methods firstly construct a domain-specific graph that represents related human knowledge. Then, they characterize node representations with neural network features and perform graph convolution to enhance these symbolic nodes via the graph neural network(GNN).Lastly, they map the enhanced node feature back into the neural network for further propagation or prediction. Through integrating knowledge graphs into neural networks, one can collaborate feature learning and graph reasoning with the same supervised loss function and achieve a more effective and interpretable way to introduce structure constraints.

Laziness and Labor: Variations on a Malevich Theme

2019 ◽  
Vol 29 (1) ◽  
pp. 259-272
Keyword(s):  
Science And Technology ◽  
Essential Elements ◽  
The State ◽  
Point Of View ◽  
John Maynard Keynes ◽  
Economic Problems ◽  
Right To Work ◽  
Starting Point ◽  
The Right ◽  
The Relationship

What idleness, leisure, and free time have in common is that they are the opposite of labor; all three are linked with the cessation or interruption of labor. The article takes Kazimir Malevich’s provocative essay Laziness as the Truth of Mankind (1921) as the starting point for an examination of the complex and fraught issue of the balance between idleness and labor. Malevich redefines idleness as grace, as the point of labor and its peer, and as something that is not only a release from hard labor but that also leads to peace and God. The author proposes a reading of Malevich’s apologetics of idleness in juxtaposition with Marx’s early focus on the issues of human freedom and on alleviating alienation in a newly arranged society, and with Paul Lafargue’s argument that workers would do better to fight for the right to be idle than for the right to work. The comparison with Marx and Lafargue reveals a fundamental flaw in their socialist program of heroic labor, which preserved the exploitation of labor but had the state rather than the capitalists appropriate it. Malevich’s argument comes close to certain insights of John Maynard Keynes in which he envisaged science and technology resolving economic problems by enabling humanity to enter an age of idleness and plenty. Giorgio Agamben’s philosophical deliberations round out the contemporary understanding of the relationship between labor and idleness. From this point of view, laziness and idleness become essential elements of meaningful labor. The option to remain idle, to reject work, to prolong it or to delay its completion are becoming the sine qua non of creative labor worthy of a free person.

scholarly journals Application of Sentinel-1B Polarimetric Observations to Soil Moisture Retrieval Using Neural Networks: Case Study for Bare Siberian Chernozem Soil

2021 ◽  
Vol 13 (17) ◽  
pp. 3480
Author(s):  
Konstantin Muzalevskiy ◽  
Anatoly Zeyliger
Keyword(s):  
Remote Sensing ◽  
Neural Networks ◽  
Surface Roughness ◽  
Soil Moisture ◽  
Soil Surface ◽  
Weather Conditions ◽  
Agricultural Field ◽  
Wide Range ◽  
Soil Surface Roughness ◽  
The Relationship

Sentinel-1 is currently the only synthetic-aperture radar, which radar measurements of the earth’s surface to be carried out, regardless of weather conditions, with high resolution up to 5–40 m and high periodicity from several to 12 days. Sentinel-1 creates a technological platform for the development of new globally remote sensing algorithms of soil moisture, not only for hydrological and climatic model applications, but also on a single field scale for individual farms in precision farming systems used. In this paper, the potential of soil moisture remote sensing using polarimetric Sentinel-1B backscattering observations was studied. As a test site, the fallow agricultural field with bare soil near the Minino village (56.0865°N, 92.6772°E), Krasnoyarsk region, the Russian Federation, was chosen. The relationship between the cross-polarized ratio, reflectivity, and the soil surface roughness established Oh used as a basis for developing the algorithm of soil moisture retrieval with neural networks (NNs) computational model. Two NNs is used as a universal regression technique to establish the relationship between scattering anisotropy, entropy and backscattering coefficients measured by the Sentinel-1B on the one hand and reflectivity on the other. Finally, the soil moisture was found from the soil reflectivity in solving the inverse problem using the Mironov dielectric model. During the field campaign from 21 May to 25 August 2020, it was shown that the proposed approach allows us to predict soil moisture values in the layer thickness of 0.00–0.05 m with the root-mean-square error and determination coefficient not worse than 3% and 0.726, respectively. The validity of the proposed approach needs additional verification on a wider dataset using soils of different textures, a wide range of variations in soil surface roughness, and moisture.

scholarly journals A Discriminative Feature Learning Approach for Remote Sensing Image Retrieval

2019 ◽  
Vol 11 (3) ◽  
pp. 281 ◽  
Author(s):  
Wei Xiong ◽  
Yafei Lv ◽  
Yaqi Cui ◽  
Xiaohan Zhang ◽  
Xiangqi Gu
Keyword(s):  
Remote Sensing ◽  
Image Retrieval ◽  
Network Structure ◽  
Feature Learning ◽  
Decisive Role ◽  
Remote Sensing Image ◽  
Feature Representations ◽  
Discriminative Feature ◽  
Comparison Of The Results ◽  
Significant Effort

Effective feature representations play a decisive role in content-based remote sensing image retrieval (CBRSIR). Recently, learning-based features have been widely used in CBRSIR and they show powerful ability of feature representations. In addition, a significant effort has been made to improve learning-based features from the perspective of the network structure. However, these learning-based features are not sufficiently discriminative for CBRSIR. In this paper, we propose two effective schemes for generating discriminative features for CBRSIR. In the first scheme, the attention mechanism and a new attention module are introduced to the Convolutional Neural Networks (CNNs) structure, causing more attention towards salient features, and the suppression of other features. In the second scheme, a multi-task learning network structure is proposed, to force learning-based features to be more discriminative, with inter-class dispersion and intra-class compaction, through penalizing the distances between the feature representations and their corresponding class centers. Then, a new method for constructing more challenging datasets is first used for remote sensing image retrieval, to better validate our schemes. Extensive experiments on challenging datasets are conducted to evaluate the effectiveness of our two schemes, and the comparison of the results demonstrate that our proposed schemes, especially the fusion of the two schemes, can improve the baseline methods by a significant margin.

The machine training in problems of satellite images’s processing

Metrologiya ◽  
2020 ◽  
pp. 15-37
Author(s):  
L. P. Bass ◽  
Yu. A. Plastinin ◽  
I. Yu. Skryabysheva
Keyword(s):  
Remote Sensing ◽  
Neural Networks ◽  
Computer Vision ◽  
Satellite Images ◽  
Vision Systems ◽  
Earth Remote Sensing ◽  
Practical Applications ◽  
Convolution Neural Networks ◽  
Computer Vision Systems ◽  
Trained Neural Network

Use of the technical (computer) vision systems for Earth remote sensing is considered. An overview of software and hardware used in computer vision systems for processing satellite images is submitted. Algorithmic methods of the data processing with use of the trained neural network are described. Examples of the algorithmic processing of satellite images by means of artificial convolution neural networks are given. Ways of accuracy increase of satellite images recognition are defined. Practical applications of convolution neural networks onboard microsatellites for Earth remote sensing are presented.

Tanggung Jawab Hukum Penilai Publik terhadap Laporan Penilaian Kegiatan Pengadaan Tanah bagi Pembangunan untuk Kepentingan Umum (Studi Atas Kantor Jasa Penilai Publik Toto Suharto)

2020 ◽  
Vol 2 (1) ◽  
Author(s):  
Gagah Yaumiyya Riyoprakoso ◽  
AM Hasan Ali ◽  
Fitriyani Zein
Keyword(s):  
Future Development ◽  
Public Interest ◽  
Legal Responsibility ◽  
Legal Research ◽  
The Public ◽  
The Government ◽  
Conducting Research ◽  
The Right ◽  
The Relationship ◽  
Systematic Explanation

This study is based on the legal responsibility of the assessment of public appraisal reports they make in land procurement activities for development in the public interest. Public assessment is obliged to always be accountable for their assessment. The type of research found in this thesis is a type of normative legal research with the right-hand of the statue approach and case approach. Normative legal research is a study that provides systematic explanation of rules governing a certain legal category, analyzing the relationship between regulations explaining areas of difficulty and possibly predicting future development. . After conducting research, researchers found that one of the causes that made the dispute was a lack of communication conducted between the Government and the landlord. In deliberation which should be the place where the parties find the meeting point between the parties on the magnitude of the damages that will be given, in the field is often used only for the delivery of the assessment of the compensation that has been done.

Rancangan Sistem Pengukuran Kinerja Berbasis Human Resource Scorecard Pada Universitas Muhammadiyah Luwuk

2020 ◽  
Vol 10 (1) ◽  
pp. 63-71
Author(s):  
Nurhaeda Abbas ◽  
Anggraini Sukmawati ◽  
Muhammad Syamsun
Keyword(s):  
Performance Measurement ◽  
Human Resource ◽  
Performance Measurement System ◽  
Group Discussions ◽  
Strategic Objectives ◽  
Depth Interview ◽  
And Performance ◽  
The Right ◽  
Hierarchy Process ◽  
The Relationship

Today the performance measurement of Muhammadiyah Luwuk uUniversity’s performance has not formulated yet based on University’s vision and mission. It will affect the strategic steps needed and performance improvement efforts in the future.  Human resource scorecard is the right system to be applied in Muhammadiyah Luwuk University. The purpose of this study is to designed a performance measurement system at Muhammadiyah Luwuk University using the Human Resource Scorecard with four perspectives: stakeholder, academic management and kemuhammadiyaan, operational and innovation, as well as and learning. Data was analyzed by analytical hierarchy process method. This research was conducted by distributing questionnaires, focus group discussions and in-depth interview with stakeholders at Muhammadiyah Luwuk University. The results showed that there were 14 strategic objectives and 33 key performance indicators to be achieved by the priority objectives, which are: empowerment and development of faculty, increased administrative process quality, improved sound budget performance and, improvement of the relationship with stakeholders.

Penanganan Issu Terkait Keselamatan Pasien di Rumah Sakit

2020 ◽  
Author(s):  
Kristin Natal Riang Gea
Keyword(s):  
Patient Safety ◽  
Health Workers ◽  
Sampling Technique ◽  
Good Health ◽  
Bivariate Analysis ◽  
P Value ◽  
Patient Identification ◽  
Cross Sectional ◽  
The Right ◽  
The Relationship

AbstrakKeselamatan pasien merupakan dasar dari pelayanan kesehatan yang baik. Pengetahuan tenaga kesehatan dalam sasaran keselamatan pasien terdiri dari ketepatan identifikasi pasien, peningkatan komunikasi yang efektif, peningkatan keamanan obat yang perlu diwaspadai, kepastian tepat lokasi, prosedur, dan tepat pasien operasi, pengurangan risiko infeksi, pengurangan risiko pasien jatuh. Tujuan penelitian untuk mengetahui hubungan antara pengetahuan dengan penerapan keselamatan pasien pada petugas kesehatan di Puskesmas Kedaung Wetan Kota Tangerang. Metode Penelitian menggunakan deskriptif korelasi menggunakan pendekatan cross sectional. Populasi sebanyak 50 responden. Teknik pengambilan sampel menggunakan total sampling. Instrumen yang digunakan berupa lembar kuesioner. Teknik analisa diatas menggunakan analisa Univariat dan Bivariat. Hasil Penelitian ada Hubungan Pengetahuan dengan Penerapan Keselamatan Pasien pada Petugas Kesehatan, dengan hasil, p value sebesar 0,013 < 0,05 maka dapat disimpulkan bahwa ada Hubungan Pengetahuan dengan Penerapa Keselamatan Pasien pada Petugas Kesehatan. Kesimpulan penelitian ada Hubungan Pengetahuan dengan Penerapan Keselamatan Pasien.. AbstrackPatient safety is the basis of good health services. Knowledge of health personnel in patient safety targets consists of accurate patient identification, increased effective communication, increased safety of the drug that needs to be watched, certainty in the right location, procedure, and precise patient surgery, reduction in risk of infection, reduction in risk of falling patients. The purpose of this study was to determine the relationship between knowledge and the application of patient safety to health workers in the Kedaung Wetan Health Center, Tangerang City. The research method uses descriptive correlation using cross sectional approach. The population is 50 respondents. The sampling technique uses total sampling. The instrument used was a questionnaire sheet. The analysis technique above uses Univariate and Bivariate analysis. The results of the study there is a Relationship of Knowledge with the Implementation of Patient Safety in Health Officers, with the result, p value of 0.013 <0.05, it can be concluded that there is a Relationship between Knowledge and Patient Safety Implementation in Health Officers. The conclusion of the study is the Relationship between Knowledge and the Implementation of Patient Safety.Keywords Knowledge, Patient safety, Health workers

Sign in / Sign up

Export Citation Format

Share Document