Ultra High Definition Research Articles

Short-Range Coulomb Interaction Is a Key to Switch the Utilization of Higher Triplet Excitons in Multiresonance Thermally Activated Delayed Fluorescence Doped Film.

Multiresonance thermally activated delayed fluorescence (MR-TADF) materials have attracted widespread attention due to ultrahigh definition display standards. However, many MR-TADF materials lack TADF properties in both the solution and solid states. Interestingly, the TADF characteristics appear once these MR-TADF compounds are doped in a suitable host film, but the precise mechanism involved in the host-guest interaction remains uncertain. Herein, we systematically investigated the role of host-guest interactions employing doped films (DABNA-1@mCBP and DABNA-1@DPEPO) with opposite phenomena. The results indicate that mCBP with a V-shape and enhanced rigidity could facilitate the formation of an energy spacing layer by employing short-range Coulomb energy through the MR luminescent core, which could offset the sensitivity of the stacking distance, enhancing the coupling between T1 and T2, and thus switch the reverse internal conversion and the higher T2 reverse intersystem crossing process. This work is a further development of luminescence mechanisms and an update of the host-guest interaction criteria for the targeted design of doped films.

Elevating Red Emission: Yb3+ Doped CsPbBr1I2 Nanocrystal Glass for Exceptional PLQY and Stability in Wide Color Gamut Backlit Displays

As display technologies progress towards enhanced visual performance, there was a growing demand for color conversion materials to meet increasingly stringent performance requirements. In recent years, all-inorganic perovskite CsPbX3 (X = Cl, Br, I) nanocrystals (NCs) had gained attention in the field of wide color gamut backlight displays due to their tunable full spectrum, narrow-band emission, high color purity, and ease of processing. However, red-emitting perovskite NCs were still facing challenges with photoluminescence quantum yield (PLQY) and stability. In this study, Yb3+ doped CsPbBr1I2 perovskite nanocrystal glass (Yb3+@CsPbBr1I2 PNG) was synthesized using a melt-quenching technique. Samples emitting at a narrow band centered at 630nm, with a full width at half maximum(FWHM) of 31.8nm, exhibited a high PLQY of 80.36% and retained fluorescence intensity at 98.2% of the initial value in water, maintaining to 92.6% after continuous blue light irradiation for 60 days. Moreover, the optical conversion film Yb3+@CsPbBr1I2@CsPbBr3@PET developed in our study showcased color gamut coverage surpassing 132.9% of the NTSC 1953 standard and 98.8% of the Rec.2020 standard. In summary, the highly efficient and remarkably stable red-emitting Yb3+@CsPbBr1I2 PNG composite material introduced in this study paved the way for future advancements in wide color gamut and ultra-high definition backlight display technologies.

High-speed 0.22 THz communication system with 84 Gbps for real-time uncompressed 8K video transmission of live events

Terahertz frequency band (0.1 ~ 10 THz) can provide ultra-high transmitting rates for massive emerging applications. However, due to the radio frequency impairments and limited signal processing property of baseband devices, it is difficult for the current terahertz communication systems to be used in real applications. In this work, we report a 0.22 THz communication system, which can provide 84 Gbps air interface rate over 1.26 kilometers distance. Such a system is underpinned by three breakthroughs, namely, integrating the reported terahertz wireless communication system, redesigning the signal processing in baseband devices, and increasing the cut-off frequency and transmitting power of radio frequency component by parameter optimizing. Being applied to realize the real-time transmission of the uncompressed 8K ultra high-definition video in the 31st International University Sports Federation in Chengdu, China, this work is a leaping advancement to enable terahertz systems from experiment to promising applications.

An ultra-high-definition multi-exposure image fusion method based on multi-scale feature extraction

Multiple exposure image fusion is a technique used to obtain high dynamic range images. Due to its low cost and high efficiency, it has received a lot of attention from researchers in recent years. Currently, most deep learning-based multiple exposure image fusion methods extract features from different exposure images using a single feature extraction method. Some methods simply rely on two different modules to directly extract features. However, this approach inevitably leads to the loss of some feature information during the feature extraction process, thus further affecting the performance of the model. To minimize the loss of feature information as much as possible, we propose an ultra-high-definition (UHD) multiple exposure image fusion method based on multi-scale feature extraction. The method adopts a U-shaped structure to construct the overall network model, which can fully exploit the feature information at different levels. Additionally, we construct a novel hybrid stacking paradigm to combine convolutional neural networks and Transformer modules. This combined module can extract both local texture features and global color features simultaneously. To more efficiently fuse and extract features, we also design a cross-layer feature fusion module, which can adaptively learn the correlation between features at different layers. Numerous quantitative and qualitative results demonstrate that our proposed method performs well in UHD multiple exposure image fusion.

Optoelectronics Interfaces for a VLC System for UHD Audio-Visual Content Transmission in a Passenger Van: HW Design.

This work aims to provide the hardware (HW) design of the optoelectronics interfaces for a visible-light communication (VLC) system that can be employed for several use cases. Potential applications include the transmission of ultra-high-definition (UHD) streaming video through existing reading lamps installed in passenger vans. In this use case, visible light is employed for the downlink, while infrared light is used for the uplink channel, acting as a remote controller. Two primary components -a Light Fidelity (LiFi) router and a USB dongle-were designed and implemented. The 'LiFi Router', handling the downlink channel, comprises components such as a visible Light-Emitting Diode (LED) and an infrared receiver. Operating at a supply voltage of 12 V and consuming current at 920 mA, it is compatible with standard voltage buses found in transport vehicles. The 'USB dongle', responsible for the uplink, incorporates an infrared LED and a receiver optimized for visible light. The USB dongle works at a supply voltage of 5 V and shows a current consumption of 1.12 A, making it well suited for direct connection to a universal serial bus (USB) port. The bandwidth achieved for the downlink is 11.66 MHz, while the uplink's bandwidth is 12.27 MHz. A system competent at streaming UHD video with the feature of being single-input multiple-output (SIMO) was successfully implemented via the custom hardware design of the optical transceivers and optoelectronics interfaces. To ensure the system's correct performance at a distance of 110 cm, the minimum signal-to-noise ratio (SNRmin) for both optical links was maintained at 10.74 dB. We conducted a proof-of-concept test of the VLC system in a passenger van and verified its optimal operation, effectively illustrating its performance in a real operating environment. Exemplifying potential implementations possible with the hardware system designed in this work, a bit rate of 15.2 Mbps was reached with On-Off Keying (OOK), and 11.25 Mbps was obtained with Quadrature Phase Shift Keying (QPSK) using Orthogonal Frequency-Division Multiplexing (OFDM) obtaining a bit-error rate (BER) of 3.3259 × 10-5 in a passenger van at a distance of 72.5 cm between the LiFi router and the USB dongle. As a final addition, a solar panel was installed on the passenger van's roof to power the user's laptop and the USB dongle via a power bank battery. It took 13.4 h to charge the battery, yielding a battery life of 22.3 h. This characteristic renders the user's side of the system entirely self-powered.

The impact of introducing deep learning based [18F]FDG PET denoising on EORTC and PERCIST therapeutic response assessments in digital PET/CT

Background[18F]FDG PET denoising by SubtlePET™ using deep learning artificial intelligence (AI) was previously found to induce slight modifications in lesion and reference organs’ quantification and in lesion detection. As a next step, we aimed to evaluate its clinical impact on [18F]FDG PET solid tumour treatment response assessments, while comparing “standard PET” to “AI denoised half-duration PET” (“AI PET”) during follow-up.Results110 patients referred for baseline and follow-up standard digital [18F]FDG PET/CT were prospectively included. “Standard” EORTC and, if applicable, PERCIST response classifications by 2 readers between baseline standard PET1 and follow-up standard PET2 as a “gold standard” were compared to “mixed” classifications between standard PET1 and AI PET2 (group 1; n = 64), or between AI PET1 and standard PET2 (group 2; n = 46). Separate classifications were established using either standardized uptake values from ultra-high definition PET with or without AI denoising (simplified to “UHD”) or EANM research limited v2 (EARL2)-compliant values (by Gaussian filtering in standard PET and using the same filter in AI PET). Overall, pooling both study groups, in 11/110 (10%) patients at least one EORTCUHD or EARL2 or PERCISTUHD or EARL2 mixed vs. standard classification was discordant, with 369/397 (93%) concordant classifications, unweighted Cohen’s kappa = 0.86 (95% CI: 0.78–0.94). These modified mixed vs. standard classifications could have impacted management in 2% of patients.ConclusionsAlthough comparing similar PET images is preferable for therapy response assessment, the comparison between a standard [18F]FDG PET and an AI denoised half-duration PET is feasible and seems clinically satisfactory.

Manufacture of the Invar Film Using a Continuous Electrodeposition Technique for the OLED FMM

In the manufacturing processes of red-green-blue (RGB)-type organic light emitting diode (OLED) displays, Invar is used as a material for the fine metal mask (FMM), which guides the evaporated diode materials through its small holes onto the correct positions of the substrate glass. The electroforming of Invar is an immediate hot issue in the RGB-type OLED display industries. Contrary to the conventional top-down method of pyrometallurgically producing Invar, a bottom-up approach of electroforming is a promising technology for producing very thin FMMs to achieve high-quality images in such displays. The present authors have recently presented that the electroformed Invar via sophisticated heat treatment exhibits the coefficient of thermal expansion (CTE) as low as that of the conventional Invar. The current work has been aimed at investigating the effect of the microstructure evolution on the CTE during heat treatment of the electroformed Invar. Finally, we propose optimal process conditions to manufacture the Invar FMM applied for an ultra high definition (UHD) grade of the OLED display.

A Deep Learning-Based Emergency Alert Wake-Up Signal Detection Method for the UHD Broadcasting System.

With the increasing frequency and severity of disasters and accidents, there is a growing need for efficient emergency alert systems. The ultra-high definition (UHD) broadcasting service based on Advanced Television Systems Committee (ATSC) 3.0, a leading terrestrial digital broadcasting system, offers such capabilities, including a wake-up function for minimizing damage through early alerts. In case of a disaster situation, the emergency alert wake-up signal is transmitted, allowing UHD TVs to be activated, enabling individuals to receive emergency alerts and access emergency broadcasting content. However, conventional methods for detecting the bootstrap signal, essential for this function, typically require an ATSC 3.0 demodulator. In this paper, we propose a novel deep learning-based method capable of detecting an emergency wake-up signal without the need for an ATSC 3.0. The proposed method leverages deep learning techniques, specifically a deep neural network (DNN) structure for bootstrap detection and a convolutional neural network (CNN) structure for wake-up signal demodulation and to detect the bootstrap and 2 bit emergency alert wake-up signal. Specifically, our method eliminates the need for Fast Fourier Transform (FFT), frequency synchronization, and interleaving processes typically required by a demodulator. By applying a deep learning in the time domain, we simplify the detection process, allowing for the detection of an emergency alert signal without the full suite of demodulator components required for ATSC 3.0. Furthermore, we have verified the performance of the deep learning-based method using ATSC 3.0-based RF signals and a commercial Software-Defined Radio (SDR) platform in a real environment.

MSTG: Multi-Scale Transformer with Gradient for joint spatio-temporal enhancement

Despite the widespread availability of HDR (High Dynamic Range) display devices, the majority of video sources are still stored in SDR (Standard Dynamic Range) format, leading to an urgent need for UHD (Ultra High Definition) video reconstruction. A simple solution is to divide the task into two sub-tasks: video super-resolution (SR) and video Bit-Depth Enhancement (BDE). While some joint enhancement tasks led by SR have been explored in multiple dimensions, the quantitative dimension has often been overlooked. In this paper, we address the first joint task of BDE and SR and propose a succinct network called MSTG. We conduct a systematic analysis to explain how this effect can be achieved and overcome the core challenge arising from distortions exacerbated by the differing optimization directions of BDE and SR. This challenge is attributed to indistinguishable contours and detailed textures. To tackle this, we employ a dual-branch module that leverages both gradient and image information to discern between BDE and SR. Furthermore, we propose a novel model named Cross-scale Transformer (CTF) embedded with Cross-scale Attention (CSA). This model facilitates information aggregation while adaptively utilizing structural similarity features. It jointly extracts, aligns, and fuses frame features at multiple scales, forming the MST module. Experiments demonstrate that the proposed algorithm outperforms the cascaded video enhancement methods by large margins on both synthetic and realistic datasets.

P‐98: High‐Quality OLED Display Panel Using Optimized IGZO Deposition Process through Eliminating Sputter Target Mura

Target Mura occurs in sputter using separated target at IGZO deposition. This study was evaluated to improve the problem of poor panel quality and reliability for Target Mura. To improve OLED panel quality of a‐IGZO TFT we have been optimized the sputter deposition method and the annealing process of IGZO/Gate Insulator layer. Improved panel quality by completely eliminating sputter target mura and we achieved to highly reliability in Image Sticking. Also, we achieved that the uniformity of threshold voltages of a‐IGZO TFTs on Gen. 8.5 glass is approximately 0.77V by using optimized TFT process. In addition, improved top gate IGZO TFTs backplane of the 55 inch 4K UHD OLED TV could be demonstration.

Local Light Field Control Enables Efficient Quantum Dot Color Conversion Films for Mini‐LED Backlit Displays

AbstractQuantum dot (QD) color conversion films (QD‐CCFs) are used in ultrahigh definition (UHD) Mini‐LED backlit displays due to their narrow band emission and high color purity. However, their poor light conversion efficiency (LCE) leads to greater film thickness and thus high cost. Herein, this study proposes a local light field control strategy to enhance the LCE of QD‐CCFs using high‐refractive BN nanosheets. The LCEs of the green and red QD‐CCFs can increase from 29.9% and 34.8% to 67.4% and 91.0% after adding BN nanosheets, respectively. Combining the experimental data with the finite‐difference time domain (FDTD) simulations, the enhanced LCEs can be attributed to observably increased blue light absorption and improved light extraction efficiency of the light emitted from QDs. The white Mini‐LEDs, prepared by integrating the optimized QD‐CCFs containing BN nanosheets with blue Mini‐LED chips, show a high luminous efficiency of 70.2 lm W−1, a high external quantum efficiency of 36.2%, and a wide color gamut of 96.3% Rec. 2020 standards. This work provides an interesting idea for designing high‐efficiency QD‐CCFs toward low‐cost UHD Mini‐LED backlit displays.

Excited-State Engineering Enables Efficient Deep-Blue Light-Emitting Diodes Exhibiting BT.2020 Color Gamut.

Organic luminescent materials that exhibit thermally activated delayed fluorescence (TADF) can convert non-emissive triplet excitons into emissive singlet states through a reverse intersystem crossing (RISC) process. Therefore, they have tremendous potential for applications in organic light-emitting diodes (OLEDs). However, with the development of ultra-high definition 4K/8K display technologies, designing efficient deep-blue TADF materials to achieve the Commission Internationale de l'Éclairage (CIE) coordinates fulfilling BT.2020 remains a significant challenge. Here, an effective approach is proposed to design deep-blue TADF molecules based on hybrid long- and short-range charge-transfer by incorporation of multiple donor moieties into organoboron multiple resonance acceptors. The resulting TADF molecule exhibits deep-blue emission at 414nm with a full width at half maximum (FWHM) of 29nm, together with a thousand-fold increase in RISC rate. OLEDs based on the champion material achieve a record maximum external quantum efficiency (EQE) of 22.8% with CIE coordinates of (0.163, 0.046), approaching the coordinates of the BT.2020 blue standard. Moreover, TADF-assisted fluorescence devices employing the designed material as a sensitizer exhibit an exceptional EQE of 33.1%. This work thus provides a blueprint for future development of efficient deep-blue TADF emitters, representing an important milestone towards meeting the blue color gamut standard of BT.2020.

Hardware Efficient Integrated In-loop Filter for HEVC Encoder

The deblocking filter (DF) and the sample adaptive offset (SAO) filter, which aids in enhancing the subjective quality of the image, make up the in-loop filter of the high-efficiency video coding (HEVC) encoder and decoder. The in-loop filter significantly increases the computational load on the HEVC encoder. It is challenging to design an in-loop filter on hardware that can handle intensive computations while using the least amount of on-chip memory, taking external memory traffic and dependencies simultaneously delivering high throughput to support Ultra HD video applications. The proposed design employs the following strategies to address these issues. This work proposes an address generation technique for pipelined horizontal and vertical filtering in DF, that avoids a transpose buffer which otherwise is required. This enables easy pipelining and parallelization thus improving throughput while reducing the on-chip memory utilization. A simplified SAO filter with parallel-pipelined processing is included in the design. These features enable the design to support ultra-HD 7680 × 4320 @ 40 fps video applications. The proposed hardware architecture has a total gate count of 7.73 K LUTs and 2.8 K slice registers, and it is implemented on a 28 nm field programmable gate array (FPGA) platform.

The Effect of Resolution and Watermark Strength on Multi-level DWT Image Watermarking

With the widespread use of the Internet and the decrease in storage costs, many media have been transferred to digital media. This situation reduces the security and reliability of digital media. Media producers use watermarking methods for copyright protection. This study focuses on wavelet transform, which is one of the frequency conversion methods for watermarking. The wavelet transform for the watermarking process is usually applied on one of the four subbands obtained in a single level. In this study, the watermarking process is carried out in a total of 12 sub-bands, including the 1st, 2nd and 3rd levels. In addition, a performance evaluation metric for digital image watermarking is presented. The evaluation is applied using 10 different watermark strength factors on 46 different image resolutions. The Ultra High-definition Demoiréing Dataset is used for testing. The mathematical results obtained as a result of 22080 iterations are shown with tables and graphics, and the performance relations between the cover image resolution value and the sub-band selection are interpreted.

Stepwise Progression of Dye-Induced In Situ Photoalignment and Subsequent Stabilization for Noncontact Alignment of Liquid Crystals.

Noncontact alignment of liquid crystals (LCs) is crucial for large-area and ultrahigh definition (UHD) display manufacturing. This research presents an innovative approach to the photoalignment of LCs, aiming to overcome challenges associated with uniformity and assembly in large-sized and UHD displays. Using homogeneously dissolved, nonionic azobenzene chromophores sensitive to both visible and UV light, we demonstrate an in situ stepwise progression of dye-induced LC alignment and subsequent stabilization using reactive mesogen (RM). Both dual-wavelength and single-wavelength approaches enable stepwise interfacial modifications for LC alignment and stabilization. The dye-induced LC alignment is rewritable, allowing for the creation of various patterns and gray-level alignments. The stability of the alignment is ensured through cross-linked RM layers, providing a robust and permanent solution for LC alignment without the need for delicate mechanical treatments. Importantly, this method addresses the challenges associated with conventional photoalignments, including various dye-induced approaches and high-energy photoalignment. The proposed method exhibits high-quality electro-optical switching, azimuthal anchoring strength, and stability against thermal, radiation, and ac-field stresses, making it a promising candidate for commercial mass production, especially in the fabrication of large-sized and UHD LC displays.

TADF deep-blue OLED with color index approaching Rec.2020 standard blue gamut achieved by a boron-based D–A type molecular skeleton

The ultra-high definition (UHD) displays in the market require organic light-emitting diodes (OLEDs) to meet the Rec.2020 color gamut standard, while it is still challenging to exploit ultra-deep-blue emitters that satisfy such blue standard. Herein, by substituting tCz (3,6-di(tert-butyl)carbazole) donor unit on the robust BOC (8,8,12,12-tetramethyl-8,12-dihydro-4-oxa-3a2-boradibenzo[cd,mn]pyrene) acceptor unit with high structural rigidity and appropriate energy levels, a novel boron-based D–A type thermally activated delayed fluorescence (TADF) ultra-deep-blue emitter, namely, tCz-BOC, was readily designed and synthesized. Inheriting from its rigid organoboron core, the emitter presents ultra-deep-blue emission with narrow spectral full width at half maximum (FWHM) of 54 nm and high photoluminescence quantum yield up to 73.7%. Additionally, tCz-BOC can effectively harvest T1 excitons through high-lying T2 mediated reverse intersystem crossing process. As a consequence, the fabricated OLED based on tCz-BOC exhibits a satisfactory ultra-deep-blue emission with λEL of 424 nm, narrow emission bandwidth with FWHM of 47 nm, excellent maximum external quantum efficiency (EQEmax) of 13.9%, as well as the CIEy coordinate of 0.047, which is fairly close to the Rec.2020 blue color gamut.

Constructing multifunctional deep-blue emitters with weak charge transfer excited state for high-performance non-doped blue OLEDs and single-emissive-layer hybrid white OLEDs

Deep-blue emitter with high photoluminescence efficiency (PLQY) is highly desirable in ultra-high definition displays and white solid-state lightings. In this work, two deep-blue phenanthro[9,10]imidazole derivatives, PPIS and PPPIS, with hot exciton property are successfully developed. Compared to PPIS, the embedded phenyl bridge in PPPIS is able to effectively increase the overlap of frontier molecular orbitals. In consequence, PPPIS shows higher oscillator strength and significantly enhanced PLQY. PPPIS also achieves better electroluminescence performance in non-doped device, showing deep-blue emission with Commission International de l'Eclairage (CIE) coordinates of (0.153, 0.087) and the maximum external quantum efficiency (EQEmax) of 8.5% with minuscule efficiency roll-off. Meanwhile, when PPPIS serves as the host for phosphor PO-01, high-efficiency orange phosphorescent device is obtained with high EQEmax of 29.8% and negligible efficiency roll-off at 1000 cd/m2. Further, efficient single-emissive-layer white device is assembled via utilizing PPPIS as a blue emitter as well as the host for PO-01 simultaneously, providing warm-white emission with CIE coordinates of (0.429, 0.433) at 1000 cd/m2, the forward-viewing EQEmax of 27.2% and maximum power efficiency (PEmax) of 80.1 lm/W, respectively. Our studies can establish a viable design strategy for deep-blue emitters in high-performance non-doped blue OLEDs and hybrid WOLEDs.

DRL Empowered On-policy and Off-policy ABR for 5G Mobile Ultra-HD Video Delivery

DRL Empowered On-policy and Off-policy ABR for 5G Mobile Ultra-HD Video Delivery

Correlation Matching Transformation Transformers for UHD Image Restoration

This paper proposes UHDformer, a general Transformer for Ultra-High-Definition (UHD) image restoration. UHDformer contains two learning spaces: (a) learning in high-resolution space and (b) learning in low-resolution space. The former learns multi-level high-resolution features and fuses low-high features and reconstructs the residual images, while the latter explores more representative features learning from the high-resolution ones to facilitate better restoration. To better improve feature representation in low-resolution space, we propose to build feature transformation from the high-resolution space to the low-resolution one. To that end, we propose two new modules: Dual-path Correlation Matching Transformation module (DualCMT) and Adaptive Channel Modulator (ACM). The DualCMT selects top C/r (r is greater or equal to 1 which controls the squeezing level) correlation channels from the max-pooling/mean-pooling high-resolution features to replace low-resolution ones in Transformers, which can effectively squeeze useless content to improve the feature representation in low-resolution space to facilitate better recovery. The ACM is exploited to adaptively modulate multi-level high-resolution features, enabling to provide more useful features to low-resolution space for better learning. Experimental results show that our UHDformer reduces about ninety-seven percent model sizes compared with most state-of-the-art methods while significantly improving performance under different training sets on 3 UHD image restoration tasks, including low-light image enhancement, image dehazing, and image deblurring. The source codes will be made available at https://github.com/supersupercong/UHDformer.

Transferability of laparoscopic skills acquired from three-dimensional high-definition and ultra-high definition endovision system to two-dimensional high-definition endovision system: an ex-vivo randomized study.

Three-dimensional high-definition (3D HD) and ultra-high-definition (4K HD) endovision systems are rapidly adopted in academic setting. However, transferability of laparoscopic skills acquired from these systems to two-dimensional high-definition (2D HD) endovision system is not known. Forty stereo-enabled surgical residents were randomized into two groups. They performed three standardized surgical tasks, Task 1(Peg transfer), Task 2(Precision touch on uneven surface) and Task 3(Surgical knotting on rubber tube) for 15 repetitions using either 3D HD or 4K HD. Both groups then performed the same tasks using 2D HD for 5 repetitions. Their performances were evaluated for execution time (speed) and error scores (safety). The residents in 3D HD group performed all three tasks significantly faster than residents in 4K HD group with comparable error scores. The time taken to complete the tasks on 2D HD were comparable between residents trained in 3D HD and 4K HD in two out of three tasks (p = 0.027, P = 0.115, p = 0.368 in task 1, 2 and 3 respectively). However, in two out of three tasks, residents trained on 3D HD committed significantly more errors than residents trained on 4K HD (p < 0.0001, p < 0.001 in task 1 and task 2 respectively). Skill acquired on 4K HD seems transferable to 2D HD environment. Participants trained in 3D HD made more errors while performing the tasks in 2D HD. It may be prudent to offer additional training on 2D HD to residents trained on 3D HD for safer laparoscopic surgical practice.