Format Conversion Research Articles

An efficient Analysis based on the Internet of Things, SVM and KNN for Operative Diabetic Retinopathy Classification

These days, diabetes is an incurable disease, with millions of people suffering from it worldwide. Several variables namely lack of education, crowded living conditions, obesity and improper diet are among the causes of this recent upsurge in diabetes cases. They are identified by the name of infections induced by bacteria or viruses, harmful compounds in food, autoimmune reactions, obesity, unhealthy lifestyles, and pollution in the environment. Excessive and sight-threatening diabetic retinopathy (DR) is the most common retinal micro-vascular dysfunction that is characterized by the occurrence of a disorder of retinal blood vessels resulting in impaired vision. The IoT-based work is conducted in this work on the machine learning (ML) techniques, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM). The classification of diabetic retinopathy is a topic that is under research. The range of activities of the processes of downsampling, labelling, image flattening, and format conversion is all within the dataset preparation process. An advanced prognosis model is designed which follows a combination of two machine learning techniques such as SVM and KNN. This approach classifies the images of diabetic retinopathy into five segments (subclasses), thus facilitating in-depth analysis. Our solution proposal in this case is a superior one because of its higher classification accuracy and faster processing speed as the findings showed. The robustness and accuracy that the SVM is known for are ensured by the convergence of the KNN to the SVM. The paper also proves a close linkage of clinical symptoms and blood sugar readings to an algorithmic DM prediction system that is based on IoT and ML approaches. This is another advantage of this method that it outperforms the existing classification methods. Amongst all the classifiers that we used in this project, the KNN ML classifier turned out to be the most accurate one with an accuracy rate of 93%. It was found that the algorithm performed with a 79% accuracy rate after tough testing and training and it was consistently providing number one quality DM predictions.

Determination of Groundwater Potential Areas Using Remote Sensing (RS) and GIS Techniques in Mubi North Local Government Area, Adamawa State, North Eastern Nigeria.

This study utilizes Remote Sensing (RS) and Geographic Information Systems (GIS) techniques to determine groundwater potential areas in Mubi North Local Government Area, Adamawa State, Nigeria. The research aim to provide insights for water resource management and sustainable development. Data collected includes Landsat imagery, Digital Elevation Model (DEM), soil data, and vector layers. Preprocessing involved data format conversion, spatial reference alignment, and resolution standardization for compatibility and consistency. Spatial analysis extracted topographic parameters from the DEM and supervised classified Landsat imagery for Land Use/Land Cover (LULC) maps. Six factors, Drainage density, Slope, Soil, Elevation, LULC, and Aspect, were reclassified into low, average, and high groundwater potential indexes. ArcGIS overlay analysis integrated the factors, prioritizing Drainage density, Slope, Soil, and Elevation, to create a composite groundwater potential index. The index was calibrated and classified into five categories: Very Low, Low, Average, Marginally High, and High potential areas. Results reveal diverse groundwater potentials. Very Low areas cover 17,168.23 hectares, Average areas cover 18,534.86 hectares, Marginally High areas cover 19,362.08 hectares, Low areas cover 13,838.41 hectares, and High areas cover 12,757.44 hectares. Spatial distribution shows High potential in settlements like Sabon Gari and Bukkaji, while Very Low potential is found in Wuro Alhaji and Venda. This study contributes to water resource management and sustainable development. Findings aid targeted strategies, benefiting socio-economic development. The research enhances knowledge and informs decision-making for water resource management, fostering development and improving livelihoods in Mubi North LGA.

Application of Intelligent Optimization Algorithms to the Design of Automatic Generation of Software Tests for Data Anomaly Identification

Abstract This paper firstly constructs a system for monitoring precursor observation data and automatic identification of anomalies, designs the functions of downloading and format conversion of precursor data files and downloading and decompression of compressed files, and adopts the interpolation method to pre-process the precursor observation data. Secondly, a method based on the SURF intelligent optimization algorithm is adopted to automatically recognize and classify anomalous data from seismic big data. Finally, experiments were carried out to predict short-acuity earthquakes, and the experimental results were analyzed and contrasted. The results show that the accuracy of 2-day prediction and 4-day prediction gradually stabilizes at 0.728~0.785 when the training reaches 100 steps and gradually decreases and reaches about 0.65 with the increase of the prediction time period, which verifies the feasibility and practicality of the system. The research in this paper provides a new idea and method for earthquake precursor data anomaly identification, which has certain research and application value.

Star-8QAM to PAM4 format conversion in highly nonlinear silicon-PTS waveguides

Star-8QAM to PAM4 format conversion in highly nonlinear silicon-PTS waveguides

Efficient FPGA-Based Sparse Matrix–Vector Multiplication With Data Reuse-Aware Compression

Sparse matrix-vector multiplication (SpMV) on FPGAs has gained much attention. The performance of SpMV is mainly determined by the number of multiplications between non-zero matrix elements and the corresponding vector values per cycle. On the one side, the off-chip memory bandwidth limits the number of non-zero matrix elements transferred from the off-chip DDR to the FPGA chip per cycle. On the other side, the irregular vector access pattern poses challenges to fetch the corresponding vector values. Besides, the read-after-write (RAW) dependency in the accumulation process shall be solved to enable a fully pipelined design. In this work, we propose an efficient FPGA-based sparse matrix-vector multiplication accelerator with data reuse-aware compression. The key observation is that repeated accesses to a vector value can be omitted by reusing the fetched data. Based on the observation, we propose a reordering algorithm to manually exploit the data reuse of fetched vector values. Further, we propose a novel compressed format called data reuse-aware compressed (DRC) to take full advantage of the data reuse and a fast format conversion algorithm to shorten the preprocessing time. Meanwhile, we propose an HLSfriendly accumulator to solve the RAW dependency. Finally, we implement and evaluate our proposed design on the Xilinx Zynq-UltraScale ZCU106 platform with a set of sparse matrices from the SuiteSparse matrix collection. Our proposed design achieves an average 1.18x performance speedup without the DRC format and an average 1.57x performance speedup with the DRC format w.r.t. the state-of-the-art work respectively.

Numerical Simulation and Analysis of Hydraulic Turbines Based on BIM for Sustainable Development

Hydropower is considered to be an important way to achieve the sustainable development goal of human progress. The performance of turbines is very important to the safety and stability of hydropower stations. Most of the hydraulic turbine performance studies only use Computational Fluid Dynamics (CFD) for performance simulation, lacking the integration of Building Information Modeling (BIM) technology and CFD. Therefore, a performance analysis model of a Francis turbine based on BIM was put forward in this paper. The BIM software OpenBuildings Designer CONNECT Edition Update 10 was used to build the hydraulic turbine model, and then the BIM model was transferred to the CFD numerical simulation platform ANSYS through the intermediate format conversion. In the ANSYS environment, the numerical simulation of different working conditions was carried out with the help of Fluent 2021 R1 software. The numerical simulation results show that the fluid velocity gradient in the volute was 2~3 m/s under the three working conditions, which was relatively stable. The water flow could progress the guide vane mechanism at a higher speed, and the drainage effect of the volute was better. There were some negative pressure areas at the back of the runner blades and the inlet of draft tube, and the negative pressure value was as high as −420,000 Pa and −436,842 Pa under maximum head conditions, which were prone to cavitation erosion. It is proven that BIM supported the hydraulic turbine performance analysis and provided a geometric information model for hydraulic turbine CFD numerical simulation, meaning that the performance analysis model based on BIM is feasible. This study can expand the application value of BIM and provide guidance for the study of hydraulic turbine numerical simulation using BIM technology in combination with CFD methods.

Effect of Format Conversion on Source Identification from Audio Recordings: A Study for Forensic Purposes

Effect of Format Conversion on Source Identification from Audio Recordings: A Study for Forensic Purposes

Data management plans as linked open data: exploiting ARGOS FAIR and machine actionable outputs in the OpenAIRE research graph

BackgroundOpen Science Graphs (OSGs) are scientific knowledge graphs representing different entities of the research lifecycle (e.g. projects, people, research outcomes, institutions) and the relationships among them. They present a contextualized view of current research that supports discovery, re-use, reproducibility, monitoring, transparency and omni-comprehensive assessment. A Data Management Plan (DMP) contains information concerning both the research processes and the data collected, generated and/or re-used during a project’s lifetime. Automated solutions and workflows that connect DMPs with the actual data and other contextual information (e.g., publications, fundings) are missing from the landscape. DMPs being submitted as deliverables also limit their findability. In an open and FAIR-enabling research ecosystem information linking between research processes and research outputs is essential. ARGOS tool for FAIR data management contributes to the OpenAIRE Research Graph (RG) and utilises its underlying services and trusted sources to progressively automate validation and automations of Research Data Management (RDM) practices.ResultsA comparative analysis was conducted between the data models of ARGOS and OpenAIRE Research Graph against the DMP Common Standard. Following this, we extended ARGOS with export format converters and semantic tagging, and the OpenAIRE RG with a DMP entity and semantics between existing entities and relationships. This enabled the integration of ARGOS machine actionable DMPs (ma-DMPs) to the OpenAIRE OSG, enriching and exposing DMPs as FAIR outputs.ConclusionsThis paper, to our knowledge, is the first to introduce exposing ma-DMPs in OSGs and making the link between OSGs and DMPs, introducing the latter as entities in the research lifecycle. Further, it provides insight to ARGOS DMP service interoperability practices and integrations to populate the OpenAIRE Research Graph with DMP entities and relationships and strengthen both FAIRness of outputs as well as information exchange in a standard way.

Optical Format Conversion and Interconnection Between 1D and 2D Constellations for Flexible Optical Transmission Networks

Optical format conversion technologies have attracted a lot of attention to bridge the different optical networks employing different formats. In this paper, the advances of optical format conversion technologies are introduced and reviewed. Through analyzing the optical network architecture and modulation technology development, we focus on the format conversions between the signals with one dimensional and two dimensional constellations. The important enabling technologies for flexible constellation manipulation are introduced in detail, including the linear and nonlinear optical effects-based functions. Then the key format conversion researches are reviewed in four categories: one-to-one conversions, optical aggregations, optical de-aggregations, and optical interconversions. Finally, the current issue and future trend of this research field is analyzed and discussed.

All-Optical Modulation Format Conversions from PAM4 to QPSK and 16QAM Using Silicon-Rich Nitride Waveguides

Quadrature phase-shift keying (QPSK) and 16-quadrature amplitude modulation (16QAM) formats are deployed in inter-data center networks where high transmission capacity and spectral efficiency are required. However, in intra-data center networks, a four-level pulse amplitude modulation (PAM4) format is deployed to satisfy the requirements for a simple and low-cost transceiver configuration. For the seamless and effective connection of such heterogeneous networks without an optical-electrical-optical conversion, an all-optical modulation format conversion technique is required. In this paper, we propose all-optical PAM4 to QPSK and 16QAM modulation format conversions using silicon-rich nitride waveguides. The successful conversions from 50-Gbps-class PAM4 signals to 50-Gbps-class QPSK and 100-Gbps-class 16QAM signals are demonstrated via numerical simulations.

Flexible All-Optical 8QAM Signal Format Conversion Using Pump Assisted Nonlinear Optical Loop Mirror

A flexible all-optical format interconversion scheme based on a pump assisted nonlinear optical loop mirror (NOLM) is proposed and numerically simulated for the first time to our knowledge. In this scheme, input multi-Gbps 8QAM signals are divided into clockwise (CW) and counter-clockwise (CCW) components by a 3-dB optical coupler (OC), which also couples light from the input pump into the NOLM from the CCW direction. The numerical model of the pump assisted NOLM with CW and CCW optical paths is simplified using a nonlinear Mach-Zehnder interferometer (MZI). Optical signals in the upper MZI arm will be mainly affected by the self-phase modulation (SPM) effect when traversing the highly nonlinear fiber (HNLF) and those in the lower MZI arm are impacted by cross-phase modulation (XPM) in addition to SPM when they experience the HNLF with the input pump light. When the upper-arm optical signal with SPM phase shift and the lower arm optical signal with SPM and XPM phase shifts are coherently mixed, a new converted 8QAM signal can be obtained. The power transfer function (PTF) of the pump assisted NOLM and the relative phase shift (RPS) between the input and the output optical signals are theoretically provided and verified. By only changing the input power of the 8QAM signal and the pump light, all-optical format interconversion of square-, standard- and star-shaped 8QAM signals can be achieved. Furthermore, the proposed scheme can achieve format conversion from the 30 Gbps square-shaped 8QAM signal to a 20 Gbps quadrature phase shift keying (QPSK) signal. The scheme performance is analyzed via constellation diagrams, eye diagrams, the error vector magnitude (EVM) and the bit error rate (BER) of the optical signals. The scheme developed can be deployed in optical gateways to connect different optical networks by dynamically selecting their appropriate modulation formats.

Design of Digital Twin System for Harbor Cranes Based on Unity 3D

Port container crane (shore bridge) is an important equipment for container loading and unloading in the port, and its working mechanism includes four parts: lifting mechanism, operating mechanism, operating system and equipment structure. In this paper, digital twin technology is integrated to realize the digital mapping of quay bridge operation, which is the basis for realizing predictive maintenance. As a link between the physical space and the digital space, the digital twin is based on a new generation of information technology, which can complete the mapping of the real space in the digital space, so as to reflect the whole life cycle process of the corresponding physical equipment, and can derive related other functions on this basis. Firstly, the bridge was constructed using 3D modeling software solid works, and the mapping and model format conversion were carried out using 3ds Max. Finally, the 3D model was imported into Unity and the port virtual environment was built. Combined with C#, the driving engine of the lifting mechanism and operating mechanism of the quayside bridge is developed, and the virtual and real synchronous operation of the quayside bridge is finally realized.

All-optical wavelength and format conversion of DQPSK-to-PAM4 using DLI and pump assisted NOLM

An all-optical wavelength and format conversion scheme of 20 Gbps differential quadrature phase shift keying (DQPSK)-to-4-ary pulse amplitude modulation (PAM4) signals based on the delay line interferometer (DLI) and the pump assisted nonlinear optical loop mirror (NOLM) is proposed and numerically simulated. The input DQPSK signal is converted into two star-16-ary quadrature amplitude modulation (star-16QAM) signals (namely, star-16QAM-I and star-16QAM-Q) using DLI with the time delay of 1-symbol and different linear phase shifts of π/4 and −π/4, respectively. The star-16QAM signals are converted into the on-off keying (OOK) signals (namely, OOK-I and OOK-Q) using the pump assisted NOLM based on the self-phase modulation (SPM) and the cross-phase modulation (XPM) in the highly nonlinear fiber (HNLF). The two OOK signals can be aggregated into different types of PAM4 signals by adjusting their power ratio (PR) and relative phase shift (RPS) flexibly. Error vector magnitude (EVM) and bit error rate (BER) for the input and output optical signals are calculated to verify the scheme performance. Compared to the DQPSK signal with an input optical signal-to-noise ratio (OSNR) of 30 dB, there are about 2.5 dB penalty of receiver OSNR for the converted OOK-I and OOK-Q signals. For the corresponding aggregated equidistant PAM4 and equiamplitude PAM4 signals, they have about 6.3 dB and 12.1 dB penalty of receiver OSNR. The proposed scheme of DQPSK-to-PAM4 can be deployed at the all-optical gateway to bridge the long-hual and short-reach optical networks.

Compression Performance Analysis of Experimental Holographic Data Coding Systems.

It is challenging to find a proper way to compress computer-generated holography (CGH) data owing to their huge data requirements and characteristics. This study proposes CGH data coding systems with high-efficiency video coding (HEVC), three-dimensional extensions of HEVC (3D-HEVC), and video-based point cloud compression (V-PCC) codecs. In the proposed system, we implemented a procedure for codec usage and format conversion and evaluated the objective and subjective results to analyze the performance of the three coding systems. We discuss the relative advantages and disadvantages of the three coding systems with respect to their coding efficiency and reconstruction results. Our analysis concluded that 3D-HEVC and V-PCC are potential solutions for compressing red, green, blue, and depth (RGBD)-sourced CGH data.

Teaching Thai Language Literacy: Proposed of Using Speech Recognition Technology Techniques to Detect Read Aloud in Thai Tonal Conjugation for Primary Education Students

This academic article has objectives were 1) to study the techniques in data analyzing of speech recognition systems techniques to detect reading aloud skills in Thai tonal conjugation for Primary Education Students, and 2) to study the effect of speech recognition systems techniques to detect reading aloud skills in Thai tonal conjugation for Primary Education Students. Sample group in this research were 30 Primary Education Students in Grades 4 – 6 of Wat Na Klang School in Nakhon Sawan Province, Thailand. Research tools was speech recognition systems techniques to detect reading aloud skills in Thai tonal conjugation.Results showed that,1) the three techniques involved in sound analysis consisted of Spectrogram Sound Wave Change Techniques, Wavelet Transform Audio Format Conversion Techniques, and two sets of data similarity comparison techniques named Dynamic Time Warping (DTW). The researchers used three techniques in analysis and comparison for making the more similar to the sound signal than using a single technique to analyze. 2) the effect of using speech recognition system to detect reading aloud skills in Thai tonal conjugation showed results of pronunciation practice and compared the similarity of audio signals that after using the technique in 5 analytical times, the largest average result was 93.52 Percent, the least average result was 87.75 Percent, and the total average result was 90.66 Percent.

Building an Electronic Medical Record System Exchanged in FHIR Format and Its Visual Presentation.

Currently, the Taiwan Electronic Medical Record Exchange Center uses the Clinical Document Architecture (CDA) framework, which is based on the international medical standard. The CDA R2 standard, defined in 2005, is used for cross-institution retrieval of electronic medical records (Ministry of Health and Welfare, Information Department, 2021). However, CDA R2 only supports the exchange of clinical documents and is limited to the XML format. Due to the lack of a standardized framework for medical data exchange in Taiwan, different standards and specifications result in different data interface methods between systems, requiring customization for each system by healthcare institutions or the government. The inconsistency in data formats requires healthcare institutions and the government to spend more time on data parsing and mapping, resulting in slow integration of medical data. In this study, we simulated healthcare institutions using Fast Healthcare Interoperability Resources (FHIR) for medical information exchange and utilized the exchanged medical information to create a dynamic dashboard to assist healthcare professionals in making medical decisions. To ensure information security, we employed Hyper Text Transfer Protocol Secure (HTTPS) for secure transmission, which encrypts the transmitted medical record data using the Transport Layer Security (TLS) protocol, preventing deliberate interception and tampering of medical record data between the two systems. Finally, to test the load and performance of static and dynamic resources and web applications, we conducted a system performance evaluation using Apache JMeter. The results of this study demonstrate that replacing the gateway of the Electronic Medical Record Exchange Center with an FHIR server effectively reduces the time and cost spent by developers on data format conversion while also mitigating the information security risks associated with the previous VPN solution. Additionally, by utilizing dynamic charts, healthcare professionals are assisted in making medical decisions.

Automatic fault location method for power communication network based on a machine learning algorithm

The current conventional automatic fault location method of power communication network mainly realizes the identification of faulty branches by calculating node power parameters, which leads to poor location effects due to the lack of effective analysis of fault signal data. In this regard, the automatic fault location method of the store head communication network based on a machine learning algorithm is proposed. Firstly, the fault signal data are collected and pre-processed, specifically including data cleaning and data format conversion, and then the individual fault signal is used as the basis to differentiate the weight parameters of different abnormal degrees and combine with the state threshold to realize the judgment of the fault node interval location. In the experiments, the fault location performance of the proposed method is verified. The experimental results show that when the proposed method is used for fault location, the relative positioning error value of the method is small and has a high fault location accuracy.

Optical channel aggregation based on modulation format conversion by coherent spectral superposition with electro-optic modulators

Spectrally efficient modulation formats become essential for optical network scaling as the demand for routed data streams exceeds the present wavelength-division multiplexing systems’ throughput. However, achieving high spectral efficiency at high data rates requires complex and bandwidth-intensive electronics. In this study, we propose an all-optical aggregation scheme that combines multiple low spectral efficiency optical wavelength channels from an optical frequency comb based transmitter into fewer channels with higher spectral efficiency. Our method utilizes coherent spectral superposition and optical vector summation, eliminating the need for optical nonlinearities and relying on linear signal processing with an electro-optic modulator. By adjusting the phase of the radio frequency signal driving the modulator, we can easily achieve the required optical phase tuning for vector addition in the I-Q plane. Through experimental demonstrations, we show that the proposed approach enables the generation of 10 GBd PAM-4 and 10 GBd quadrature phase shift keying (QPSK) signals by aggregating two 10 GBd binary phase shift keying signals. Similarly, we aggregate two 10 GBd QPSK signals into one 10 GBd quadrature amplitude modulation-16 (QAM-16) signal. The experiments employ both conventional and sinc-shaped Nyquist signals. Our in-line, all-optical aggregation concept significantly enhances operational capacity while reducing complexity. It offers a promising solution for realizing flexible integrated optical transmitters for advanced modulation format signals using lower-quality electronics. Additionally, it aligns with the requirements of future dynamically reconfigurable optical networks that leverage spectral traffic aggregation. Given its reliance on linear signal processing with an electro-optic modulator, the integration of the method into any integrated photonic platform is straightforward.

Rapid extraction of pavement aggregate gradation based on point clouds using deep learning networks

Usage of asphalt mixture with poor gradation will most likely lead to pavement deficiency. There is a growing need for rapid and non-destructive methods to extract pavement aggregate gradation. In this study, a deep learning-based method that utilizes point clouds data for gradation extraction was proposed. Firstly, a data enhancement algorithm along with three data format conversion methods (aligned point cloud, voxel, and depth image) were proposed to preprocess the original collected point clouds. Subsequently, different neural network models were designed for each data format to extract gradation. Finally, a multi-feature fusion network was developed, which using extraction network as the backbone and additional auxiliary information. In the case study, the MAE loss of multi-feature fusion networks with PointNet, Vox-ResNet34 and GoogLeNet-v4 as the backbone respectively achieved 0.202, 0.142 and 0.046 on the test set, which means an estimation accuracy of more than 95% for the pavement aggregate gradation.

An Intermediate Language for General Sparse Format Customization

The inevitable trend of hardware specialization drives an increasing use of custom data formats in processing sparse workloads, which are typically memory-bound. These formats facilitate the automated generation of target-aware data layouts to improve memory access latency and bandwidth utilization. However, existing sparse tensor programming models and compilers offer little or no support for productively customizing the sparse formats. Moreover, since these frameworks adopt an attribute-based approach for format abstraction, they cannot easily be extended to support general format customization. To overcome this deficiency, we propose UniSparse, an intermediate language that provides a unified abstraction for representing and customizing sparse formats. More concretely, we express a sparse format as a map from dense coordinates to a layout tree using a small set of well-defined query and mutation primitives. We also develop a compiler leveraging the MLIR infrastructure, which supports adaptive customization of formats, and automatic code generation of format conversion and compute operations for heterogeneous architectures. We demonstrate the efficacy of our approach through experiments running commonly-used sparse linear algebra operations with hybrid formats on multiple different hardware targets, including an Intel CPU, an NVIDIA GPU, and a simulated processing-in-memory (PIM) device.