A novel white-light hyperbranched conjugated polymer utilizing polyfluorene as the branches and Ir(piq)₃ cores has been investigated.

With developments in deep learning, semantic segmentation of remote sensing images has made great progress. Currently, mainstream methods are based on convolutional neural networks (CNNs) or vision transformers. However, these not very effective extracting features from images, which usually high resolution with plenty detail. Operations including downsampling will cause the loss such features. To address this problem, we propose a novel module called Hierarchical Wavelet Feature Enhancement...

Recently, the development of deep learning boosts object detection for remote sensing images. The existing methods can be divided into two types. region-based represented by Faster R-CNN have progressive performance in accuracy. However, their computational cost is massive due to Convolutional Neural Network (CNN) backbones, which limits efficiency. regression-based such as YOLO and Single Shot MultiBox Detector (SSD) are advantageous speed while accuracy not satisfactory. To meet increasing...

At the onset of their flight careers, novice pilots often lack clarity regarding standard attention-allocation pattern. Therefore, to enhance efficiency initial training, it is crucial for students develop a comprehensive understanding control and behavior during learning process. In this study, flight-performance data eye-movement from experienced instructors in no-power stall scenarios were collected create an training course. An experimental group underwent course, while followed...

In this letter, we present a hierarchical connection graph (HCG) algorithm based on self-avoiding polygon (SAP) model for detecting and extracting gable roofs from aerial imagery. The SAP is deformable shape that capable of representing various shapes appearances. composed sequence roof-corner templates are connected into SAP, which serves as flexible prior. An energy function combines features three channels (corner, boundary, interior area) defined over the to quantify variability in...

A novel type of single white‐light‐emitting copolymers, PF‐DTFO x , derived from poly(9,9‐dioctylfluorene) (PF) in which 2,7‐di‐(2‐thienyl)‐9‐fluorenone (DTFO) is introduced as an orange‐emitting unit, reported, and their application white‐light polymer light‐emitting devices (PLEDs) explored. Because the very simple structure DTFO, synthetic routes to are not complicated, raw materials cheap; these advantageous reducing fabrication cost PLEDs. In fluorescence spectra, solid powder shows...

Spiro[3.3]heptane-2,6-dispirofluorene as a three-dimensional, conjugation-uninterrupted branching point helps to enhance the electroluminescent performance of polyfluorene-based white-light-emitting hyperbranched copolymers.

A heteroleptic iridium(<sc>iii</sc>) complex [(CzhBTZ)₂Ir(fpptz)] was synthesized for efficient and stable PhOLEDs by spin-coating. It shows a maximum brightness of 9617 cd m⁻², current efficiency about 9.43 A⁻¹, International Commission on Illumination (CIE) coordinates (0.42, 0.56).

In this paper, we address the problem of multi-class semantic segmentation high resolution aerial imagery with a deep-convolutional-neural-network-based model named D-LinkNet, which was initially designed for task road extraction. D-LinkNet extends LinkNet, is considered as an efficient method and adopts encoder-decoder architecture, by inserting dilated convolution layers between LinkNets encoder decoder to enlarge receptive field kernel without reducing feature maps. Besides, use form...

A potential hyperbranched structure with 2,7-fluorene-<italic>alt</italic>-3,6-carbazole and 4,7-dithienyl-2,1,3-benzothiadiazole branches, a three-dimensional-structured spiro[3.3]heptane-2,6-dispirofluorene core is synthesized for efficient stable WPLEDs.

We prepared organic light-emitting diodes (OLED) using graphene oxide (GO) doped poly(3,4ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) as a hole injection layer (HIL). The PEDOT:PSS: GO HIL was by mixing into PEDOT:PSS aqueous solution and spin-coating onto indiumtin (ITO). luminance of the ITO/PEDOT:PSS:GO (w)/NPB/Alq3/LiF/Al devices were investigated. High luminous efficiency at low turn-on voltage observed when PEDOT:PSS:GO in OLEDs. This can be attributed to its enhanced...

This paper addresses the problem of contextual hyperspectral image (HSI) classification. A novel conditional random fields (CRFs) model, known as higher order support vector (HSVRFs), is proposed for HSI By incorporating potentials into a with Mahalanobis distance boundary constraint (SVRFMC) HSVRFs model not only takes advantage machine (SVM) classifier and constraint, but can also capture level information to depict complicated details in HSI. The are defined on segments, which created by...

In the field of road extraction, an dominant network is D-LinkNet which won first place in DeepGlobe 2018 challenge. Although creatively proposed D-block with progressively enlarged dilated convolution and proved its efficiency, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{N}\times \mathrm{N}$</tex> square kernel it used still has limitations for extraction. Road aerial imagery usually narrow-and-long shape, direction randomly...

In 2021, the Transformer based models have demonstrated extraordinary achievement in field of computer vision. Among which, Maskformer, a model adopting mask classification method, is an outstanding both semantic segmentation and instance segmentation. Considering specific characteristics remote sensing images (RSIs), we design CADA-MaskFormer(a Mask classification-based with Cross-shaped window self-Attention Densely connected feature Aggregation) on Maskformer by improving its encoder...

With the development of remote sensing, semantic segmentation high-resolution sensing images (RSIs) is increasingly essential. At same time, characteristics objects in RSIs, such as large size, variation object scales, and complex details, make it necessary to capture both long-range context local information. There are some methods Fully Convolutional Networks (FCN) Pyramid Scene Parsing Network (PSPNet) lack ability dependencies, due limited receptive field Neural (CNN). However,...

A new idea of nosebleed air jets with strong coupled internal and external flow is put forward using the lateral jet control principle to improve maneuverability fast reaction capabilities hypersonic vehicles. The vehicle’s nose stagnant high-pressure high-temperature gas utilized as drive source for long-term control. significant interaction changes aerodynamic characteristics. As a result, structure basic does not rely on any achieve flight attitude complicated characteristics in...

On object detection of remote sensing images, speed and accuracy are considered equally significant for time-sensitive tasks. Traditional methods images use anchors to regress the bounding boxes position objects which quite limit detection. In this paper, we make an attempt at employing anchor-free method named CenterNet address problem images. This model computes key point location, local offset size without enumerating a nearly exhaustive list potential locations size. Thus is simpler,...

Abstract The reverse design of a tunable terahertz metasurface antenna is proposed based on the deep neural network (DNN). To obtain properties antenna, phase‐changed material vanadium dioxide (VO 2 ) introduced by controlling voltage without changing structures antenna. improve efficiency and accuracy DNN used to establish relationship amplitude phase for unit. prediction errors are below 10%. radiation direction can be changed from 0° 52° VO with different states. error angle only 1°...

Since the D-LinkNet won first place in CVPR2018 Deep-Globe Challenge, deep neural network built on encoder-decoder structure has become dominant pixel-wise road extraction [1], [2]. The primary reason for D-LinkNet's success is creative design of D-Block, which extracts features by dilated convolutions with receptive fields progressively growing sizes and fuses element-wise addition. Although D-Block promotes connectivity significantly’ still struggles detecting regions compact topologies...

The identification of shadow and shading boundaries is a key step towards reducing the imaging effects that are caused by direct illumination light source in scene. Discriminating from images natural scenes has been widely applied field computer vision such as object recognition, intelligent monitoring image understanding. In this paper, we propose method to identify based on multiple kernel learning. We first extract all possible candidate then analyze their properties. Unlike previous...