As a fundamental and challenging task in bridging language vision domains, Image-Text Retrieval (ITR) aims at searching for the target instances that are semantically relevant to given query from other modality, its key challenge is measure semantic similarity across different modalities. Although significant progress has been achieved, existing approaches typically suffer two major limitations: (1) It hurts accuracy of representation by directly exploiting bottom-up attention based...

In smart-logistics ecological chains led by e-commerce platforms, the organisational efficiency has become an important indicator to measure chain’s development. However, no scientific theoretical framework exists regarding factors affecting such efficiency. This paper adopts a multi-case study method research four platforms in China and investigate current situation development of chains. Through interviews analysis, proposes influencing about platform. First, it reveals that under platform...

Achieving rapid and effective object detection in large-scale unmanned aerial vehicle (UAV) images presents a challenge. Existing methods typically split the original large UAV image into overlapping patches perform on each patch. However, extensive object-free background areas imagery reduce efficiency. To address this issue, we propose an efficient approach for via multi-task classification. Specifically, develop lightweight classification (MTC) network to efficiently identify areas. Our...

This study investigates the performance of Global Precipitation Measurement (GPM) satellite in comparison to X-band Phased Array Weather Radar (XPAR) regarding precipitation measurement accuracy, focusing on radar echoes. A comparative analysis was conducted two significant events that occurred 2023 Xiong’ an New Area, Hebei Province, China, utilizing data from XPAR, GPM, and ground-based observations. The results reveal XPAR outperforms GPM quantitative estimation, with a correlation...

Purpose The purpose of this paper is to provide an overview the evolution service supply chain management from a behavioural operations perspective, pointing out future research directions for scholars. Design/methodology/approach This study searched five databases relevant literature published between 2009 and 2018, selecting 64 papers review. selected was categorised according two dimensions: link perspective factor perspective. Comparative analysis used identify gaps in literature,...

The precise location of the tropical cyclone (TC) center is critical for intensity estimation and trajectory prediction. Due to variability TC morphology structure, there are still some challenges in locating its automatically. ability deep convolutional network capture multilevel structural features images exploited. Furthermore, a two-step scheme proposed, which contains object detection TCs with learning comprehensive decision centers. In detection, considering statistical scale...

Radar echo extrapolation is a basic but essential task in meteorological services. It could provide radar prediction results with high spatiotemporal resolution computationally efficient way, and effectively enhance the operational system's forecasting capability for hazards. Traditional methods perform by estimating motions between contiguous data. This strategy difficult to characterize complex nonlinear processes effectively, it benefit from large historical Recently, machine learning...

In order to obtain the feasible condition that both logistics service integrator (LSI) and functional provider (FLSP) have altruistic preferences, this paper establishes four investment decision models with Stackelberg game. It is found supply chain coordination can not be achieved when LSI FLSP then ex-post payment contract 'revenue sharing + franchise fee' are proposed solve problem. Some important conclusions found. First, utilities of higher than only or has an preference. Second,...

Remote sensing image captioning (RSIC) has great significance in understanding, which describes the content natural language. Existing methods are mainly based on deep learning and rely encoder–decoder model to generate sentences. In decoding process, recurrent neural network (RNN) long short-term memory (LSTM) normally applied sequentially captions. this letter, transformer is combined with grid features improve RSIC performance. First, pretrained convolutional (CNN) used extract grid-based...

During an automatic power transmission line inspection, a large number of images are collected by unmanned aerial vehicles (UAVs) to detect existing defects in components, especially insulators. However, with twin insulator strings the inspection images, when umbrella skirts rear string obstructed front string, defect detection becomes difficult. To solve this problem, we propose method self-shattering insulators based on spatial features contained images. Firstly, segmented according...

In this paper we consider the energy-efficient resource allocation that minimizes a general cost function of average user powers for small- or medium-scale wireless sensor networks, where simple time-division multiple-access (TDMA) is adopted as multiple access scheme. A class so-called ß-fair functions derived to balance tradeoff between efficiency and fairness in designs. Based on such functions, optimal channel-adaptive schemes are developed both single-hop multihop TDMA networks. Relying...

In weather forecasting, numerical prediction (NWP) that is based on physical models requires proper post-processing before it can be applied to actual operations. Therefore, research intelligent algorithms has always been an important topic in this field. This paper proposes a model output deep learning (MODL) method for post-processing, which improve the forecast effect of prediction. MODL end-to-end convolutional neural network, directly learns mapping relationship between fields by and...

In energy-constrained wireless networks such as sensor networks, it is important to maximize power efficiency by minimizing consumption for a given quality of service the data rate; equally evenly or fairly distribute all nodes network life. this paper, we develop optimal control methods balance tradeoff between energy and fairness cooperative where several relays assist communication multiple source-destination pairs. Our policy derived in quasi-closed form solving convex optimization...

This paper studies a two-echelon logistics service supply chain (LSSC) composed of one integrator (LSI) and functional provider (FLSP). We develop two-period capacity procurement model where market demand surges in the second period. The LSI, as Stackelberg game leader, tends to be overconfident when surges. results show that period, if purchasing LSI cannot meet demand, overconfidence behaviour will negatively affect LSI's optimal pricing. Besides, leads lowest level FLSP period under...

Abstract A colorless and high temperature resistant fluorinated polyimide (PI) film based on a novel cross‐linkable diamine monomer 4,4′‐diamino‐4″‐(2,3,5,6‐tetrafluoro‐4‐ethylenephenoxy) triphenylmethane (DFPTM) was successfully prepared. The special backbone the bulky tetrafluorostyrol pendant group in DFPTM gave PI material eximious solubility organic solvents, tensile properties, good thermal stability, excellent optical transparency (cutoff wavelength for 331 nm light transmittance...