基于改进Faster R-CNN及YOLO的大角度车牌识别算法

随着大数据时代的到来,智慧城市等一系列概念的提出,人工智能开始在城市各个角落得到广泛应用。其中,车牌识别作为城市道路交通重要的一环,也取得了重大突破。基于深度学习的车牌识别一经提出,各类算法应用于车牌检测及识别上,对比于传统的车牌识别,大大提高了识别的速率和准确率。然而,在非限制条件下,如大角度车牌识别准确率仍有提升的空间。提出了一种改进的FasterR-CNN与YOLO的深度学习车牌识别算法,将对大角度下的车牌识别准确率达到了99.7%。由于是轻量型的框架,为后续的移动设备部署工作提供了便利。     

Towards artificial intelligence at scale in the chemical industry

In the Industry 4.0 era, the chemical industry is embracing broad adoption of artificial intelligence (AI) and machine learning (ML) methods. This article provides a holistic view of how the industry is transforming digitally towards AI at scale. First, a historical perspective on how the industry used AI to aid humans in better decision-making is shown. Then state-of-the-art AI research addressing industrial needs on reliability and safety, process optimization, supply chain, material discovery, and reaction engineering is highlighted. Finally, a vision of the plant of the future is illustrated with critical components of AI-ready culture, model life cycle management, and renewed role of humans in chemical manufacturing.     

基于深度学习的番茄叶部病害识别模型

番茄叶病种类多、成因复杂,其预防和识别难度较大。传统基于机器学习的方法多靠人工识别,需要一定的专家经验,且具有主观性强、识别准确率不高等缺点。为实现番茄叶病特征的自动提取,并提高识别准确率,提出一种基于深度学习的番茄叶病识别模型。该模型基于卷积神经网络对番茄叶部病害特征进行自动提取,获得高维特征后,采用PCA降维算法去除冗余特征;从增大类间距离并减小类内距离的角度改进了softmax分类器,提高了识别准确率。将该模型在CrowdAI提供的数据集上进行了仿真验证,结果表明,该模型能够对番茄叶部常见10种病害进行自动提取特征和识别,综合识别准确率达到95%以上。     

基于主动学习的离群点集成挖掘方法研究

离群点检测任务通常缺少可用的标注数据，且离群数据只占整个数据集的很小一部分，相较于其他的数据挖掘任务，离群点检测的难度较大，尚没有单一的算法适合于所有的场景。因此，结合多样性模型集成和主动学习思想，提出了一种基于主动学习的离群点集成检测方法OMAL（Outlier Mining based on Active Learning）。在主动学习框架指导下，根据各种基学习器的对比分析，选择了基于统计的、基于相似性的、基于子空间划分的三个无监督模型作为基学习器。将各基学习器评判的处于离群和正常边界的数据整合后呈现给人类专家进行标注，以最大化人类专家反馈的信息量；从标注的数据集和各基学习器投票产生的数据集中抽样，基于GBM（Gradient BoostingMachine）训练一个有监督二元分类模型，并将该模型应用于全数据集，得出最终的挖掘结果。实验表明，提出方法的AUC有了较为明显的提升，且具有良好的运行效率，具备较好的实用价值。     

机器人系统有限时间自适应迭代学习控制

研究任意初态下，机器人系统的有限时间自适应迭代学习控制方法。引入初始修正吸引子的概念，构造一个含有初始修正项的误差变量。针对定常机器人系统和时变机器人系统，采用Lyapunov-like方法，分别设计迭代学习控制器处理系统中不确定性。并且，采用未含/含限幅学习机制，保证闭环系统各变量的一致有界性和误差变量在整个作业区间一致收敛性。藉以实现跟踪误差在预先指定区间的完全跟踪。仿真结果验证所设计控制方法的有效性。     

基于“动态学习数据分析”的智慧课堂 评价考核体系构建

This paper firstly describes the current situation and limitations of learning evaluation and assessment in colleges and uni- versities; then, it analyzes the possibility of implementing the learning evaluation and assessment based on "dynamic learning data analysis"; then, it describes the design process of the learning evaluation and assessment system in detail; finally, it determines the various assessment items and carries out quantitative processing to better apply them to actual teaching.     

基于大数据的五年制高职护生在线学习行为要素研究

在线学习已逐渐成为高职院校一种重要学习形式,该领域也成为当前研究热点之一。对在线学习进行介绍,同时对大数据技术在在线学习平台中的应用进行分析,重点对五年制高职护生在线学习行为要素进行研究分析,为更好地开展在线学习评价提供参考,促进高职院校更好地开展在线学习,提高在线学习质量。     

Transfer learning between crop types for semantic segmentation of crops versus weeds in precision agriculture

Agricultural robots rely on semantic segmentation for distinguishing between crops and weeds to perform selective treatments and increase yield and crop health while reducing the amount of chemicals used. Deep‐learning approaches have recently achieved both excellent classification performance and real‐time execution. However, these techniques also rely on a large amount of training data, requiring a substantial labeling effort, both of which are scarce in precision agriculture. Additional design efforts are required to achieve commercially viable performance levels under varying environmental conditions and crop growth stages. In this paper, we explore the role of knowledge transfer between deep‐learning‐based classifiers for different crop types, with the goal of reducing the retraining time and labeling efforts required for a new crop. We examine the classification performance on three datasets with different crop types and containing a variety of weeds and compare the performance and retraining efforts required when using data labeled at pixel level with partially labeled data obtained through a less time‐consuming procedure of annotating the segmentation output. We show that transfer learning between different crop types is possible and reduces training times for up to 80%. Furthermore, we show that even when the data used for retraining are imperfectly annotated, the classification performance is within 2% of that of networks trained with laboriously annotated pixel‐precision data.     

视频侦查中人像智能分析应用及算法优化

人像智能分析指的是对视频或录像中的人像进行结构化和可视化分析,对目标人物进行性别、年龄、发型等特征的智能识别,这项技术在视频侦查中有极高的应用价值。人像识别早期的算法是通过人工提取特征,通过学习低级视觉特征来针对不同属性进行分类学习,这种基于传统方法的模型表现常常不尽如人意。在计算机视觉领域,通过海量图像数据学习的神经网络比传统方法有更丰富的信息量和特征可以被提取。文章尝试通过深度学习技术训练神经网络模型对行人进行检测和识别,对于衣着不同的行人进行智能识别,具有更好的鲁棒性,提升了视频人像识别的准确率,拓展了人工智能技术在身份识别领域的应用。     

Steganography using a 3-player game

Image steganography aims to securely embed secret information into cover images. Until now, adaptive embedding algorithms such as S-UNIWARD or Mi-POD, were among the most secure and most often used methods for image steganography. With the arrival of deep learning and more specifically, Generative Adversarial Networks (GAN), new steganography techniques have appeared. Among them is the 3-player game approach, where three networks compete against each other. In this paper, we propose three different architectures based on the 3-player game. The first architecture is proposed as a rigorous alternative to two recent publications. The second takes into account stego noise power. Finally, our third architecture enriches the second one with a better interaction between embedding and extracting networks. Our method achieves better results compared to existing works Hayes and Danezis (2017), Zhu et al. (2018), and paves the way for future research on this topic.