Processing Pipeline Research Articles

Update to ttprocessing: the R-package to handle the TreeTalker monitoring data

Managing forest ecosystems at global scale is a pressing challenge for resource managers and it needs to address important questions on continuous monitoring. Advance in information technology and in the use of multifunctional devices based on Internet of Things (IoT) technology, for real time observation of biological and physical variable of trees is getting more and more common in modern forestry. Among the different smart sensor networks, the Tree Talkers® devices are modular multiparameter devices designed to monitor tree physiological parameters (e.g. sap flux, growth) and surrounding microenvironmental parameters (e.g. temperature, humidity). There is a dedicated R package for handling the data produced by this monitoring system: ttprocessing, that has been recently updated, with several improvements in the data processing pipelines, the implementation of the soil water content equations of the TT-soil monitoring device and the inclusion of some basic outlier detection functionalities. In this Technical note, after a summary of the structure of the ttprocessing R package, we describe the new and update version. The updates include: a refinement of the methodology to compute the sap-flux, code for compatibility TT+ devices of version 2.0 and 3.4, that previously were not covered, code for processing the TT-soil data, a device of the Tree Talker system that monitors soil moisture and temperature. Finally, several utility functions have been added: outlier removal and functions for the technical management of the system, thus increasing its usability and making it a more comprehensive tool.

Unsupervised clustering of bitcoin transactions

Since its inception in 2009, Bitcoin has become and is currently the most successful and widely used cryptocurrency. It introduced blockchain technology, which allows transactions that transfer funds between users to take place online, in an immutable manner. No real-world identities are needed or stored in the blockchain. At the same time, all transactions are publicly available and auditable, making Bitcoin a pseudo-anonymous ledger of transactions. The volume of transactions that are broadcast on a daily basis is considerably large. We propose a set of features that can be extracted from transaction data. Using this, we apply a data processing pipeline to ultimately cluster transactions via a k-means clustering algorithm, according to the transaction properties. Finally, according to these properties, we are able to characterize these clusters and the transactions they include. Our work mainly differentiates from previous studies in that it applies an unsupervised learning method to cluster transactions instead of addresses. Using the novel features we introduce, our work classifies transactions in multiple clusters, while previous studies only attempt binary classification. Results indicate that most transactions fall into a cluster that can be described as common user transactions. Other clusters include transactions made by online exchanges and lending services, those relating to mining activities as well as smaller clusters, one of which contains possibly illicit or fraudulent transactions. We evaluated our results against an online database of addresses that belong to known actors, such as online exchanges, and found that our results generally agree with them, which enhances the validity of our methods.

Automated Prediction of Photographic Wound Assessment Tool in Chronic Wound Images

Many automated approaches have been proposed in literature to quantify clinically relevant wound features based on image processing analysis, aiming at removing human subjectivity and accelerate clinical practice. In this work we present a fully automated image processing pipeline leveraging deep learning and a large wound segmentation dataset to perform wound detection and following prediction of the Photographic Wound Assessment Tool (PWAT), automatizing the clinical judgement of the adequate wound healing. Starting from images acquired by smartphone cameras, a series of textural and morphological features are extracted from the wound areas, aiming to mimic the typical clinical considerations for wound assessment. The resulting extracted features can be easily interpreted by the clinician and allow a quantitative estimation of the PWAT scores. The features extracted from the region-of-interests detected by our pre-trained neural network model correctly predict the PWAT scale values with a Spearman's correlation coefficient of 0.85 on a set of unseen images. The obtained results agree with the current state-of-the-art and provide a benchmark for future artificial intelligence applications in this research field.

Methodology for biomarker discovery with reproducibility in microbiome data using machine learning

BackgroundIn recent years, human microbiome studies have received increasing attention as this field is considered a potential source for clinical applications. With the advancements in omics technologies and AI, research focused on the discovery for potential biomarkers in the human microbiome using machine learning tools has produced positive outcomes. Despite the promising results, several issues can still be found in these studies such as datasets with small number of samples, inconsistent results, lack of uniform processing and methodologies, and other additional factors lead to lack of reproducibility in biomedical research. In this work, we propose a methodology that combines the DADA2 pipeline for 16s rRNA sequences processing and the Recursive Ensemble Feature Selection (REFS) in multiple datasets to increase reproducibility and obtain robust and reliable results in biomedical research.ResultsThree experiments were performed analyzing microbiome data from patients/cases in Inflammatory Bowel Disease (IBD), Autism Spectrum Disorder (ASD), and Type 2 Diabetes (T2D). In each experiment, we found a biomarker signature in one dataset and applied to 2 other as further validation. The effectiveness of the proposed methodology was compared with other feature selection methods such as K-Best with F-score and random selection as a base line. The Area Under the Curve (AUC) was employed as a measure of diagnostic accuracy and used as a metric for comparing the results of the proposed methodology with other feature selection methods. Additionally, we use the Matthews Correlation Coefficient (MCC) as a metric to evaluate the performance of the methodology as well as for comparison with other feature selection methods.ConclusionsWe developed a methodology for reproducible biomarker discovery for 16s rRNA microbiome sequence analysis, addressing the issues related with data dimensionality, inconsistent results and validation across independent datasets. The findings from the three experiments, across 9 different datasets, show that the proposed methodology achieved higher accuracy compared to other feature selection methods. This methodology is a first approach to increase reproducibility, to provide robust and reliable results.

A fast, lightweight deep learning vision pipeline for autonomous UAV landing support with added robustness

Despite massive development in aerial robotics, precise and autonomous landing in various conditions is still challenging. This process is affected by many factors, such as terrain shape, weather conditions, and the presence of obstacles. This paper describes a deep learning-accelerated image processing pipeline for accurate detection and relative pose estimation of the UAV with respect to the landing pad. Moreover, the system provides increased safety and robustness by implementing human presence detection and error estimation for both landing target detection and pose computation. Human presence and landing pad location are performed by estimating the presence probability via segmentation. This is followed by the landing pad keypoints’ location regression algorithm, which, in addition to coordinates, provides the uncertainty of presence for each defined landing pad landmark. To perform the aforementioned tasks, a set of lightweight neural network models was selected and evaluated. The resulting measurements of the system’s performance and accuracy are presented for each component individually and for the whole processing pipeline. The measurements are performed using onboard embedded UAV hardware and confirm that the method can provide accurate, low-latency feedback information for safe landing support.

Change of location class in gas pipelines from a regulatory perspective

AbstractLocation class is an essential factor in the gas pipeline project because it considers factors such as population density and the number of buildings along the gas pipeline route. As these parameters change, the pipeline operator needs to review the location class, also requiring a possible assessment of changes in allowable pipeline operating pressure. Brazilian regulations related to the safety of pipeline processes are defined in the Onshore Pipeline Technical Regulation and this document indicates the application of the requirements of the ASME B31.8 standard for changing the location class. Although there are preventive and mitigating measures in the ASME B31.8 standard, they are seen as overly conservative. Other international standards have different concepts, with evaluation criteria based on risk analysis. The issue has also been a challenge for other countries, which are reviewing their regulations and developing guidelines for gas pipeline operators. The main objective of this study is to carry out a survey of international practices related to changing the location class of gas pipelines and to propose a criterion based on the best international practices based on risk management, allowing a more comprehensive view of this subject.

A uniform data processing pipeline enables harmonized nanoparticle protein corona analysis across proteomics core facilities

Protein corona, a layer of biomolecules primarily comprising proteins, forms dynamically on nanoparticles in biological fluids and is crucial for predicting nanomedicine safety and efficacy. The protein composition of the corona layer is typically analyzed using liquid chromatography-mass spectrometry (LC-MS/MS). Our recent study, involving identical samples analyzed by 17 proteomics facilities, highlighted significant data variability, with only 1.8% of proteins consistently identified across these centers. Here, we implement an aggregated database search unifying parameters such as variable modifications, enzyme specificity, number of allowed missed cleavages and a stringent 1% false discovery rate at the protein and peptide levels. Such uniform search dramatically harmonizes the proteomics data, increasing the reproducibility and the percentage of consistency-identified unique proteins across distinct cores. Specifically, out of the 717 quantified proteins, 253 (35.3%) are shared among the top 5 facilities (and 16.2% among top 11 facilities). Furthermore, we note that reduction and alkylation are important steps in protein corona sample processing and as expected, omitting these steps reduces the number of total quantified peptides by around 20%. These findings underscore the need for standardized procedures in protein corona analysis, which is vital for advancing clinical applications of nanoscale biotechnologies.

Racoon_clip-a complete pipeline for single-nucleotide analyses of iCLIP and eCLIP data.

A vast variety of biological questions connected to RNA-binding proteins can be tackled with UV crosslinking and immunoprecipitation (CLIP) experiments. However, the processing and analysis of CLIP data are rather complex. Moreover, different types of CLIP experiments like iCLIP or eCLIP are often processed in different ways, reducing comparability between multiple experiments. Therefore, we aimed to build an easy-to-use computational tool for the processing of CLIP data that can be used for both iCLIP and eCLIP data, as well as data from other truncation-based CLIP methods. Here, we introduce racoon_clip, a sustainable and fully automated pipeline for the complete processing of iCLIP and eCLIP data to extract RNA binding signal at single-nucleotide resolution. racoon_clip is easy to install and execute, with multiple pre-settings and fully customizable parameters, and outputs a conclusive summary report with visualizations and statistics for all analysis steps. racoon_clip is implemented as a Snakemake-powered command line tool (Snakemake version ≥7.22, Python version ≥3.9). The latest release can be downloaded from GitHub (https://github.com/ZarnackGroup/racoon_clip/tree/main) and installed via pip. A detailed documentation, including installation, usage, and customization, can be found at https://racoon-clip.readthedocs.io/en/latest/. The example datasets can be downloaded from the Short Read Archive (SRA; iCLIP: SRR5646576, SRR5646577, SRR5646578) or the ENCODE Project (eCLIP: ENCSR202BFN).

Finding neural correlates of depersonalisation/derealisation disorder via explainable CNN-based analysis guided by clinical assessment scores

Mental health disorders are typically diagnosed based on subjective reports (e.g., through questionnaires) followed by clinical interviews to evaluate the self-reported symptoms. Therefore, considering the interconnected nature of psychiatric disorders, their accurate diagnosis is a real challenge without indicators of underlying physiological dysfunction. Depersonalisation/derealisation disorder (DPD) is an example of dissociative disorder affecting 1–2 % of the population. DPD is characterised mainly by persistent disembodiment, detachment from surroundings, and feelings of emotional numbness, which can significantly impact patients' quality of life. The underlying neural correlates of DPD have been investigated for years to understand and help with a more accurate and in-time diagnosis of the disorder. However, in terms of EEG studies, which hold great importance due to their convenient and inexpensive nature, the literature has often been based on hypotheses proposed by experts in the field, which require prior knowledge of the disorder. In addition, participants' labelling in research experiments is often derived from the outcome of the Cambridge Depersonalisation Scale (CDS), a subjective assessment to quantify the level of depersonalisation/derealisation, the threshold and reliability of which might be challenged. As a result, we aimed to propose a novel end-to-end EEG processing pipeline based on deep neural networks for DPD biomarker discovery, which requires no prior handcrafted labelled data. Alternatively, it can assimilate knowledge from clinical outcomes like CDS as well as data-driven patterns that differentiate individual brain responses. In addition, the structure of the proposed model targets the uncertainty in CDS scores by using them as prior information only to guide the unsupervised learning task in a multi-task learning scenario. A comprehensive evaluation has been done to confirm the significance of the proposed deep structure, including new ways of network visualisation to investigate spectral, spatial, and temporal information derived in the learning process. We argued that the proposed EEG analytics could also be applied to investigate other psychological and mental disorders currently indicated on the basis of clinical assessment scores. The code to reproduce the results presented in this paper is openly accessible at https://github.com/AbbasSalami/DPD_Analysis.

Inferring neural activity before plasticity as a foundation for learning beyond backpropagation

For both humans and machines, the essence of learning is to pinpoint which components in its information processing pipeline are responsible for an error in its output, a challenge that is known as ‘credit assignment’. It has long been assumed that credit assignment is best solved by backpropagation, which is also the foundation of modern machine learning. Here, we set out a fundamentally different principle on credit assignment called ‘prospective configuration’. In prospective configuration, the network first infers the pattern of neural activity that should result from learning, and then the synaptic weights are modified to consolidate the change in neural activity. We demonstrate that this distinct mechanism, in contrast to backpropagation, (1) underlies learning in a well-established family of models of cortical circuits, (2) enables learning that is more efficient and effective in many contexts faced by biological organisms and (3) reproduces surprising patterns of neural activity and behavior observed in diverse human and rat learning experiments.

Pattern-based serverless data processing pipelines for Function-as-a-Service orchestration systems

The serverless cloud computing execution model presents an opportunity for implementing data processing pipelines in a cost efficient manner with minimum or no consideration about managing their operational aspects. In doing so, however, a major concern that emerges is the adoption of provider-specific languages required for orchestrating serverless functions and supporting services into the higher order logic of pipelines. As a result, the vendor lock-in problem inherent to cloud computing becomes further intensified. In this work we propose an approach to mitigate this issue through the adoption and adaptation of well-known patterns from the literature as the means for modeling the pipeline logic. By providing mappings from these patterns to vendor-specific orchestration language constructs we allow for efficient realization of provider-agnostic pipeline models as provider-specific executable workflows. An industrial case study provides evidence towards the suitability of our proposal for practical purposes.

ЦИФРОВЕ МОДЕЛЮВАННЯ РУЙНУВАННЯ ДЕГАЗАЦІЙНИХ ТРУБОПРОВОДІВ У ПІДЗЕМНИХ ГІРНИЧИХ ВИРОБКАХ

Purpose. The purpose of the work is to simulate the process of corrosion formation on a degassing pipeline using Solid Works and COMSOL Multiphysics 5.6 software products. Methodology. Development of a mathematical model for establishing corrosion zones of a steel degassing gas pipeline and modeling the corrosion process in a mine environment using the SolidWorks and COMSOL Multiphysics 5.6 software package. Results. It has been established that the degassing network of mines is constantly under the influence of mechanical-electrochemical interaction, which manifests itself during longitudinal deformations of the rocks at the bottom of the mine workings. The high level of corrosion processes in underground pipelines is the result of the interaction of the metal, which acts as an electrode, with groundwater, which acts as an electrolyte; at the same time, the determining factors in the corrosion process are the electrical conductivity of the soil and deformation processes in pipelines. The primary corrosion protection of underground degassing pipelines can be considered as the improvement of methods for monitoring and maintaining the technical condition of mine degassing gas pipelines to eliminate mechanical deformations and the effects of aggressive groundwater. Scientific novelty. It has been established that the high level of corrosion processes in underground pipelines is the result of the interaction of the metal, which acts as an electrode, with groundwater, which acts as an electrolyte. At the same time, the determining factors of the corrosion process are the electrical conductivity of the rocks at the bottom of the mine workings and deformation processes in pipelines. Practical implication. The reasons for the low service life of underground degassing pipelines under the constant influence of mechanical and electrochemical corrosion have been determined. Keywords: degassing, vacuum underground gas pipeline, methane-air mixture, monitoring, corrosion, throughput.

Analysis of sensitivity to hydrate blockage risk in natural gas gathering pipeline

During the operational process of natural gas gathering and transmission pipelines, the formation of hydrates is highly probable, leading to uncontrolled movement and aggregation of hydrates. The continuous migration and accumulation of hydrates further contribute to the obstruction of natural gas pipelines, resulting in production reduction, shutdowns, and pressure build-ups. Consequently, a cascade of risks is prone to occur. To address this issue, this study focuses on the operational process of natural gas gathering and transmission pipelines, where a comprehensive framework is established. This framework includes theoretical models for pipeline temperature distribution, pipeline pressure distribution, multiphase flow within the pipeline, hydrate blockage, and numerical solution methods. By analyzing the influence of inlet temperature, inlet pressure, and terminal pressure on hydrate formation within the pipeline, the sensitivity patterns of hydrate blockage risks are derived. The research indicates that reducing inlet pressure and terminal pressure could lead to a decreased maximum hydrate formation rate, potentially mitigating pipeline blockage during natural gas transportation. Furthermore, an increase in inlet temperature and terminal pressure, and a decrease in inlet pressure, results in a displacement of the most probable location for hydrate blockage towards the terminal station. However, it is crucial to note that operating under low-pressure conditions significantly elevates energy consumption within the gathering system, contradicting the operational goal of energy efficiency and reduction of energy consumption. Consequently, for high-pressure gathering pipelines, measures such as raising the inlet temperature or employing inhibitors, electrical heat tracing, and thermal insulation should be adopted to prevent hydrate formation during natural gas transportation. Moreover, considering abnormal conditions such as gas well production and pipeline network shutdowns, which could potentially trigger hydrate formation, the installation of methanol injection connectors remains necessary to ensure production safety.

Adaptive data reduction workflows for astronomy: The ESO Data Processing System (EDPS)

Context. Astronomical data reduction is usually done via processing pipelines that consist of a series of individual processing steps that can be executed one at a time. These processing steps are then strung together into workflows and fed with data to address a particular processing goal. Examples of such pipeline processing goals are the quality control of incoming data from telescopes, unsupervised production of science and calibration products for an archive, and supervised data reduction to serve the specific science goals of a scientist. For each of these goals, individual workflows need to be developed. These workflows need to evolve when the pipeline, observing strategies, or calibration plans change. Writing and maintaining such a collection of workflows is therefore a complex and expensive task. Aims. In this paper we propose a data processing system that automatically derives processing workflows for different use cases from a single specification of a cascade of processing steps. Methods. The system works by using formalised descriptions of data processing pipelines that specify the input and output of each processing step. Inputs can be existing data or the output of a previous step. The rules for selecting the most appropriate input data are directly attached to the description. Results. A version of the proposed system has been implemented as the ESO Data Processing System (EDPS) in the Python language. The specification of processing cascades and data organisation rules use a restrictive set of Python classes, attributes, and functions. Conclusions. The EDPS implementation of the proposed system was used to demonstrate that it is possible to automatically derive from a single specification of a pipeline processing cascade the workflows that the European Southern Observatory uses for quality control, archive production, and specialised science reduction. The EDPS will be used to replace all data reduction systems using different workflow specifications that are currently in use at the European Southern Observatory.

Brain microstructure and connectivity in COVID-19 patients with olfactory or cognitive impairment

IntroductionThe COVID-19 pandemic has affected millions worldwide, causing mortality and multi-organ morbidity. Neurological complications have been recognized. This study aimed to assess brain structural, microstructural, and connectivity alterations in patients with COVID-19-related olfactory or cognitive impairment using post-acute (time from onset: 264[208–313] days) multi-directional diffusion-weighted MRI (DW-MRI). MethodsThe study included 16 COVID-19 patients with cognitive impairment (COVID-CM), 35 COVID-19 patients with olfactory disorder (COVID-OD), and 14 controls. A state-of-the-art processing pipeline was developed for DW-MRI pre-processing, mean diffusivity and fractional anisotropy computation, fiber density and cross-section analysis, and tractography of white-matter bundles. Brain parcellation required for probing network connectivity, region-specific microstructure and volume, and cortical thickness was based on T1-weighted scans and anatomical atlases. ResultsCompared to controls, COVID-CM patients showed overall gray matter atrophy (age and sex corrected p = 0.004), and both COVID-19 patient groups showed regional atrophy and cortical thinning. Both groups presented an increase in gray matter mean diffusivity (corrected p = 0.001), decrease in white matter fiber density and cross-section (corrected p < 0.05), , and COVID-CM patients also displayed an overall increased diffusivity (p = 0.022) and decreased anisotropy (corrected p = 0.038) in white matter. Graph-based analysis revealed reduced network modularity, with an extensive pattern of connectivity increase, in conjunction with a localized reduction in a few connections, mainly located in the left hemisphere. The left cingulate, anterior cingulate, and insula were primarily involved. ConclusionExpanding upon previous findings, this study further investigated significant alterations in brain morphology, microstructure, and connectivity in COVID-19 patients with olfactory or cognitive disfunction. These findings suggest underlying neurodegeneration, neuroinflammation, and concomitant compensatory mechanisms. Future longitudinal studies are required to monitor the alterations over time and assess their transient or permanent nature.

Developing a Sow Vulva Volume Estimation Pipeline Based on LiDAR Imagery and Deep Learning

Highlights An artificial intelligence-enabled imagery data processing pipeline. Sow vulva swelling can be quantified using imagery automatically. Biological signs of sow estrus are quantified. Abstract. Accurate and timely estrus detection is one of the critical factors for improving the conception rate of sows in swine reproduction. Estrus detection in the current industry relies on experienced breeding technicians using the back pressure test method, which is time-consuming, labor-intensive, and subjective. This study aimed to develop an image-processing pipeline that could predict the onset of estrus by monitoring the variation of the vulva volume of sows. A robotic imaging system consisting of a LiDAR camera was used to collect depth and infrared (IR) images of sows multiple times per hour. The developed image processing pipeline consisted of a deep learning model (MobileNet) to exclude images with defective postures (e.g., excreting posture, vulva region not showing) for vulva detection. The selected images were further processed to (1) segment sow vulva regions (3-dimensional surface) using a U-Net model, (2) exclude incorrectly segmented 3D vulva surfaces using a MobileNet model, and (3) calculate the vulva volume using the segmented 3D vulva surface. Results show that all three models achieved high test accuracy of 94.8% (MobileNet 1), 96.7% (U-Net), and 100% (MobileNet 2). The vulva volume was calculated using the developed pipeline for all the test sows. Daily average vulva volume showed a noticeable increase in all sows at 0-1 day prior to the onset of estrus. This study concluded that the developed image processing pipeline could automatically detect vulva swelling, which may be used as a biological sign indicating the onset of post-weaning estrus in sows. Keywords: 3D camera, Digital agriculture, Estrus detection, Robotic imaging system, Swine reproduction.

Structure and principle of operation of the apparatus for introducing chlorine into the process pipeline

The article discusses the study of swirling flow in technological pipelines. It examines the nature of the swirling flow, presenting the experimental velocity distribution and pressure inside the pipe in graphical and tabular form. The scope of the study includes the displacement of liquid and gas, as well as the purification of natural waters using ozone and chlorine. The article analyzes the existing technological processes for the preparation of drinking water from surface water sources in Uzbekistan. Due to the significant volumes of water consumption for economic activities, the issue of drinking water disinfection is crucial. In addition to traditional disinfection methods, modern researchers are focusing on developing innovative technologies for this process. The article also studies the features of water formation and its qualitative composition, with an analysis of the reliability of the existing drinking water treatment technology.

Parallel VLSI Architectures for High-Throughput Image Processing in Medical Imaging

The need for efficient and reliable image processing techniques has never been greater in medical imaging. The proposed work analyzes and designs parallel VLSI (Very Large-Scale Integration) architectures that meet the stringent demands of high-throughput medical image processing. The remainder of the paper is organized as follows, beginning with a discussion of the computational challenges that arise organically in medical imaging and why it is crucial to surmount these obstacles. The section on the proposed work describes the architecture, parallelization strategies, and processing pipeline, providing context for the hardware and Software underlying the system. The evaluation results below demonstrate significant enhancements in processing speed and image quality. Due to their real-time capabilities and high efficiency, parallel VLSI architectures have the potential to revolutionize medical image processing, resulting in more accurate diagnoses and more effective treatments. Research into optimizing these structures and discovering new applications is anticipated to result in further advances in medical imaging.

Triggerless data acquisition pipeline for Machine Learning based statistical anomaly detection

This work describes an online processing pipeline designed to identify anomalies in a continuous stream of data collected without external triggers from a particle detector. The processing pipeline begins with a local reconstruction algorithm, employing neural networks on an FPGA as its first stage. Subsequent data preparation and anomaly detection stages are accelerated using GPGPUs. As a practical demonstration of anomaly detection, we have developed a data quality monitoring application using a cosmic muon detector. Its primary objective is to detect deviations from the expected operational conditions of the detector. This serves as a proof-of-concept for a system that can be adapted for use in large particle physics experiments, enabling anomaly detection on datasets with reduced bias.

High-Speed Phase Structured Light Integrated Architecture on FPGA

Phase-shifting fringe projection profilometry (FPP) in three-dimension (3D) structured light measurement is widely applied in high-accuracy 3D shape measurement scenarios. But motion objects or other dynamic scenarios make the traditional phase-shifting FPP hard to deal with due to its sizable CPU-based execution time. Therefore, an embedded high-speed and high-accuracy phase-shifting FPP system is designed and implemented on FPGAs. The system implements a three-frequency four-step phase-shifting hardware algorithm with RTL-level caching, multi-frame image ping-pong buffering, and data handling, etc. The system exhibits fully-parallel and pipeline processing while behaving with lower resource and power consumption. Measurement experiments indicate that the proposed method can reach high-accuracy real-time 3D shape measurement with a 3D decoding speed of 40.15 fps when the clock frequency is 200MHz, and the image size is 2432×2048.