新工科和工业4.0背景下,许多高校的高级计算机视觉课程开设了深度学习模型教学,迅速提升了研究生的工程能力,其中大数据和开源模型的信息安全和应用价值观是思政教育热点。本文首先简要分析了该课程融入思政教育的三个问题:信息安全教...新工科和工业4.0背景下,许多高校的高级计算机视觉课程开设了深度学习模型教学,迅速提升了研究生的工程能力,其中大数据和开源模型的信息安全和应用价值观是思政教育热点。本文首先简要分析了该课程融入思政教育的三个问题:信息安全教育的忽视,价值观教育的忽视和变化的社会责任,结合视觉数据获取、视觉目标检测/分类、视频/图像生成三个方面的案例给出了可参考的缓解对策,在高级计算机视觉课程中推进人工智能教育的同时,加强研究生的信息安全观、求是创新的价值观和新的社会责任感,确保该课程思政教育的科学管理和有效运行。Under the background of new engineering and industry 4.0, many colleges and universities have set up deep learning model teaching in advanced computer vision courses, which has rapidly improved the engineering ability of graduate students. The information security and application values of big data and open source models are the hot spots of ideological and political education. This paper first briefly analyzes the three problems of integrating the course into ideological and political education: the neglect of information security education, the neglect of value education and the changing social responsibility. Combined with the three aspects of visual data acquisition, visual target detection/classification, and video/image generation, this paper gives a reference mitigation strategy. While promoting artificial intelligence education in advanced computer vision courses, we should strengthen graduate students’ information security concept, realistic and innovative values and new social responsibility, so as to ensure the scientific management and effective operation of ideological and political education in this course.展开更多
近些年来,随着人工智能技术的发展,人工智能逐渐应用于各行各业,因此,中小学开始普及人工智能教育。人工智能教育需要造价昂贵的硬件设备以及专业技能较强的专任教师,然而很多中小学尤其是西部边远中小学无力承担建设专门的人工智能技...近些年来,随着人工智能技术的发展,人工智能逐渐应用于各行各业,因此,中小学开始普及人工智能教育。人工智能教育需要造价昂贵的硬件设备以及专业技能较强的专任教师,然而很多中小学尤其是西部边远中小学无力承担建设专门的人工智能技术实验室的高额预算,并且人工智能教育师资力量也严重滞后。针对这些问题,本文提出利用Tensorflow.js和Tensorflow Lite等技术,开发基于移动终端的“人工智能技术实验平台”,构建面向应用的交互式人工智能技术教学应用案例,无缝运行于“移动人工智能技术实验平台”上,支持中小学学生在线参与人工智能技术应用实践的全过程,帮助学生提升兴趣,领会人工智能技术的基本原理。实践表明,基于移动人工智能实验平台的中小学教育取得了令人满意的效果。In recent years, with the advancement of artificial intelligence (AI) technology, AI has been increasingly applied across various industries, prompting primary and secondary schools to introduce AI education. However, AI education necessitates expensive hardware facilities and a teaching force proficient in specialized skills, posing challenges for many primary and secondary schools, particularly those in remote western regions, which struggle to afford the high costs associated with establishing dedicated AI technology laboratories. Furthermore, there is a significant shortfall in qualified AI educators. To address these issues, this paper proposes leveraging technologies such as Tensorflow.js and Tensorflow Lite to develop an “AI Technology Experiment Platform” based on mobile devices. This platform will facilitate the construction of application-oriented interactive teaching application cases on AI technology that can seamlessly operate on the “Mobile AI Technology Experiment Platform”, enabling students to engage in the full process of practical AI technology applications online. This approach aims to en展开更多
在烟卷生产过程中,水渍、黄斑、褶皱和破损等缺陷是常见问题。传统的方法依赖于人工检测方法,但由于视觉疲劳等因素,导致生产效率不够高效。为了解决这一问题,本文提出了一种基于改进YOLOv8的烟卷污渍检测算法。本文以YOLOv8网络为骨干...在烟卷生产过程中,水渍、黄斑、褶皱和破损等缺陷是常见问题。传统的方法依赖于人工检测方法,但由于视觉疲劳等因素,导致生产效率不够高效。为了解决这一问题,本文提出了一种基于改进YOLOv8的烟卷污渍检测算法。本文以YOLOv8网络为骨干模型,加入了空间注意力机制,突出污渍区域的小目标检测能力,减小漏检和误检,从而减少小目标样本的误检率和漏检率。实验结果表明,与最新的YOLOv8算法相比,改进后的算法在烟卷污渍数据集上的平均精度均值(mAP)和召回率(Recall)分别提高了6.3%和1.8%;精度提高了6.2%。In the cigarette production process, defects such as water stains, yellow spots, wrinkles, and damage are common issues. Traditional methods rely on manual inspection, but due to factors like visual fatigue, production efficiency is not optimal. To address this problem, this paper proposes a cigarette stain detection algorithm based on an improved YOLOv8. We use the YOLOv8 network as the backbone model and incorporate a spatial attention mechanism to reduce the false detection and missed detection rates of small target samples. Experimental results show that, compared to the latest YOLOv8 algorithm, the improved algorithm increases the mean Average Precision (mAP) and Recall on the cigarette stain dataset by 6.3% and 1.8%, respectively, and improves precision by 6.2%.展开更多
文摘新工科和工业4.0背景下,许多高校的高级计算机视觉课程开设了深度学习模型教学,迅速提升了研究生的工程能力,其中大数据和开源模型的信息安全和应用价值观是思政教育热点。本文首先简要分析了该课程融入思政教育的三个问题:信息安全教育的忽视,价值观教育的忽视和变化的社会责任,结合视觉数据获取、视觉目标检测/分类、视频/图像生成三个方面的案例给出了可参考的缓解对策,在高级计算机视觉课程中推进人工智能教育的同时,加强研究生的信息安全观、求是创新的价值观和新的社会责任感,确保该课程思政教育的科学管理和有效运行。Under the background of new engineering and industry 4.0, many colleges and universities have set up deep learning model teaching in advanced computer vision courses, which has rapidly improved the engineering ability of graduate students. The information security and application values of big data and open source models are the hot spots of ideological and political education. This paper first briefly analyzes the three problems of integrating the course into ideological and political education: the neglect of information security education, the neglect of value education and the changing social responsibility. Combined with the three aspects of visual data acquisition, visual target detection/classification, and video/image generation, this paper gives a reference mitigation strategy. While promoting artificial intelligence education in advanced computer vision courses, we should strengthen graduate students’ information security concept, realistic and innovative values and new social responsibility, so as to ensure the scientific management and effective operation of ideological and political education in this course.
文摘近些年来,随着人工智能技术的发展,人工智能逐渐应用于各行各业,因此,中小学开始普及人工智能教育。人工智能教育需要造价昂贵的硬件设备以及专业技能较强的专任教师,然而很多中小学尤其是西部边远中小学无力承担建设专门的人工智能技术实验室的高额预算,并且人工智能教育师资力量也严重滞后。针对这些问题,本文提出利用Tensorflow.js和Tensorflow Lite等技术,开发基于移动终端的“人工智能技术实验平台”,构建面向应用的交互式人工智能技术教学应用案例,无缝运行于“移动人工智能技术实验平台”上,支持中小学学生在线参与人工智能技术应用实践的全过程,帮助学生提升兴趣,领会人工智能技术的基本原理。实践表明,基于移动人工智能实验平台的中小学教育取得了令人满意的效果。In recent years, with the advancement of artificial intelligence (AI) technology, AI has been increasingly applied across various industries, prompting primary and secondary schools to introduce AI education. However, AI education necessitates expensive hardware facilities and a teaching force proficient in specialized skills, posing challenges for many primary and secondary schools, particularly those in remote western regions, which struggle to afford the high costs associated with establishing dedicated AI technology laboratories. Furthermore, there is a significant shortfall in qualified AI educators. To address these issues, this paper proposes leveraging technologies such as Tensorflow.js and Tensorflow Lite to develop an “AI Technology Experiment Platform” based on mobile devices. This platform will facilitate the construction of application-oriented interactive teaching application cases on AI technology that can seamlessly operate on the “Mobile AI Technology Experiment Platform”, enabling students to engage in the full process of practical AI technology applications online. This approach aims to en
文摘在烟卷生产过程中,水渍、黄斑、褶皱和破损等缺陷是常见问题。传统的方法依赖于人工检测方法,但由于视觉疲劳等因素,导致生产效率不够高效。为了解决这一问题,本文提出了一种基于改进YOLOv8的烟卷污渍检测算法。本文以YOLOv8网络为骨干模型,加入了空间注意力机制,突出污渍区域的小目标检测能力,减小漏检和误检,从而减少小目标样本的误检率和漏检率。实验结果表明,与最新的YOLOv8算法相比,改进后的算法在烟卷污渍数据集上的平均精度均值(mAP)和召回率(Recall)分别提高了6.3%和1.8%;精度提高了6.2%。In the cigarette production process, defects such as water stains, yellow spots, wrinkles, and damage are common issues. Traditional methods rely on manual inspection, but due to factors like visual fatigue, production efficiency is not optimal. To address this problem, this paper proposes a cigarette stain detection algorithm based on an improved YOLOv8. We use the YOLOv8 network as the backbone model and incorporate a spatial attention mechanism to reduce the false detection and missed detection rates of small target samples. Experimental results show that, compared to the latest YOLOv8 algorithm, the improved algorithm increases the mean Average Precision (mAP) and Recall on the cigarette stain dataset by 6.3% and 1.8%, respectively, and improves precision by 6.2%.
