基于学习主动中心轮廓模型的场景文本检测

在场景文本检测领域,存在由于文本尺寸波动较大导致的小文本漏检、大文本欠检测和多尺度文本边界检测错误的情况。针对上述问题,提出一种基于学习主动中心轮廓模型的场景文本检测网络。在残差网络ResNet的基础上构建多尺度特征权重融合模型,对输入的场景文本图片进行多尺度特征提取和权重融合,并计算出最终的特征融合图,适应场景文本长宽比变化较大的情况。在此基础上,将融合后的特征图输入到学习主动中心轮廓模型预测文本框的中心点和边界,该模型为场景文本检测提供丰富先验知识,以解决多尺度文本检测框包含过多背景或部分包围文本造成的边界检测错误问题。在MSRA-TD500、IC13、IC15和IC17MLT数据集上的实验结果表明,该网络能够提高多尺度场景文本检测的准确率,其中在MSRA-TD50数据集上F-measure为0.83,相较于MSR方法提升1%,在IC13数据集上F-measure为0.91,相较于PixelLink网络提升2%,在IC15数据集上F-measure值为0.87,相较于PSENet网络提升1%,在IC17MLT数据集上F-measure值为0.74,相较于TridentNet网络提升1%。     

结合栈式监督AE与可变加权ELM的回归预测模型

在现代工业生产过程中,许多关键变量与产品质量或生产效率密切相关,关键变量的实时监测是实现利润最大化及节能降耗的有效途径。针对回归预测任务中目标特征提取不全面、预测精度较低等问题,提出一种基于栈式监督自编码器与可变加权极限学习机的回归预测模型。通过堆叠多层自编码器并在每层自编码器中添加回归网络,同时以有监督方式对栈式自编码器（SAE）进行逐层预训练,得到与输出变量相关的特征表示。利用反向传播算法对网络参数进行微调,优化自编码器模型参数。在分析提取特征与输出变量的相关性基础上,对极限学习机（ELM）的输入权值和偏置进行加权得到预测结果。实验结果表明,与基于ELM和SAE-ELM的回归预测模型相比,该模型在多晶硅铸锭的G6产品数据集上的均方根误差降低0.056 7和0.011 2、决定系数提高0.489 3和0.290 3,具有更高的回归预测准确性及更强的鲁棒性与泛化性能。     

Optimal Deep Learning Model Enabled Secure UAV Classification forIndustry 4.0

Emerging technologies such as edge computing, Internet of Things (IoT), 5G networks, big data, Artificial Intelligence (AI), and Unmanned Aerial Vehicles (UAVs) empower, Industry 4.0, with a progressive production methodology that shows attention to the interaction between machine and human beings. In the literature, various authors have focused on resolving security problems in UAV communication to provide safety for vital applications. The current research article presents a Circle Search Optimization with Deep Learning Enabled Secure UAV Classification (CSODL-SUAVC) model for Industry 4.0 environment. The suggested CSODL-SUAVC methodology is aimed at accomplishing two core objectives such as secure communication via image steganography and image classification. Primarily, the proposed CSODL-SUAVC method involves the following methods such as Multi-Level Discrete Wavelet Transformation (ML-DWT), CSO-related Optimal Pixel Selection (CSO-OPS), and signcryption-based encryption. The proposed model deploys the CSO-OPS technique to select the optimal pixel points in cover images. The secret images, encrypted by signcryption technique, are embedded into cover images. Besides, the image classification process includes three components namely, Super-Resolution using Convolution Neural Network (SRCNN), Adam optimizer, and softmax classifier. The integration of the CSO-OPS algorithm and Adam optimizer helps in achieving the maximum performance upon UAV communication. The proposed CSODL-SUAVC model was experimentally validated using benchmark datasets and the outcomes were evaluated under distinct aspects. The simulation outcomes established the supreme better performance of the CSODL-SUAVC model over recent approaches.     

Learn to survive crises: The role of firm resilience,innovation capabilities and environmental dynamism

Exogenous crises, while disruptive, may also present learning opportunities that could affect a firm's viability and performance. In this study, we examine how exogenous crises can constitute learning opportunities and assess their impact on firm survival. In particular, we investigate the role of learning in response to exogenous crises and how firm resilience, innovation capabilities, and environmental dynamism influence this relationship. Drawing from crisis management and organisational learning literature, we propose that these factors can bolster the connection between learning from crises and firm survival. To test our hypotheses, we conduct a nuanced analysis using both regression analysis and Fuzzy Set Qualitative Comparative Analysis (fsQCA) on data from 249 Italian manufacturing Small and Medium-sized Enterprises (SMEs). This approach allows us to simultaneously examine the impact of firm resilience, innovation capabilities, and environmental dynamism on the relationship between learning from crises and firm survival. Our findings offer theoretical and practical insights into the role of learning from crises in a firm's survival. They also highlight the importance of embracing learning opportunities in crisis situations and suggest that how firms deal with crises could be an opportunity to fine-tune their internal processes and thrive in the long run.     

Japanese educational and learning systems in the post-industrial society: A value chain of educational and learning systems

Japan's educational system has some major problems. The most important among these concerns is the basic concept of the educational process and the goal of education. The old concept of public educational systems has become outdated in today's Japanese society, although this concept had supported social and spiritual faith, economic success and selfless devotion to one's country for more than 100 years. Now, Japanese people need a new concept of the educational process and the goal of education for the twenty-first century. The paper proposes a value chain of educational and learning systems aimed at building a network consisting of multiple fields for fostering future human resources.     

普通本科院校信息与计算科学专业实习实训方案设计

分析当前普通高校信息与计算科学专业学生实习实训模式存在的主要问题,并针对这些问题提出一些解决方案;探讨适合学生实习实训要求的企业实训模式,以安庆师范学院信息与计算科学专业学生实习实训模式为例,对学生实习实训的计划和实施过程进行了阐述。实践表明,新的实训模式能够使学生取得良好的实习实训效果。     

基于抗体选择策略的克隆挖掘算法

针对如何有效地利用大量的原始数据分析现状来预测未来的问题,基于抗体选择策略提出一种克隆选择挖掘算法。通过评估抗体的支持度、可信度和亲和度,求得有效的关联规则。实验结果表明,该算法能较快地获得可理解的规则,并且具有较高的准确率。     

基于多代理的个性化推荐学习系统

针对现有学习系统存在信息过载、缺乏个性化服务能力、不能提供检索服务的问题,提出基于多代理构建个性化推荐学习系统。该系统利用JADE设计学习者Agent与推荐Agent,采用Lucene设计带有个性化能力的搜索引擎支持推荐,并融合3种推荐方法发挥多Agent间协商与协作的优势。实验结果表明,相比单一推荐方法,该系统具有较好的推荐效果和效率。     

Pseudo software: A mediating instrument for modeling software requirements

A new synthesis of software requirements models called pseudo software is proposed with the aim to cut requirements-related errors. Pseudo software achieves this aim by serving as a mediating instrument to empower stakeholders to participate in requirements elicitation and validation through model construction and manipulation, and to provide guidance to the development team to correctly interpret the requirements in the downstream development activities. Pseudo software obtains its traits as a mediating instrument through the choice of requirements information bits and the use of multimodal representations with tool support to integrate the requirements. Using historical data of fifty projects in the enterprise computing domain, pseudo software is shown to effectively cut the requirements-related errors committed by both the customer and the development team.     

Guiding students to the right questions: adaptive navigation support in an E‐Learning system for Java programming

Rapid growth of the volume of interactive questions available to the students of modern E‐Learning courses placed the problem of personalized guidance on the agenda of E‐Learning researchers. Without proper guidance, students frequently select too simple or too complicated problems and ended either bored or discouraged. This paper explores a specific personalized guidance technology known as adaptive navigation support. We developed JavaGuide, a system, which guides students to appropriate questions in a Java programming course, and investigated the effect of personalized guidance a three‐semester long classroom study. The results of this study confirm the educational and motivational effects of adaptive navigation support.