{
  "paper_id": "18bf5cee3ee2355cf36ee9c4ef5b4173",
  "paper_title": "Challenges in Sensor-based Human Activity Recognition and a Comparative Analysis of Benchmark Datasets: A Review",
  "paper_authors": [
    "Anindya Das Antar",
    "Masud Ahmed",
    "Md Atiqur Rahman Ahad"
  ],
  "publication_date": "",
  "publication_venue": "2019 Joint 8th International Conference on Informatics, Electronics \\& Vision (ICIEV) and 2019 3rd International Conference on Imaging, Vision \\& Pattern Recognition (icIVPR)",
  "abstract": "Human Activity Recognition using embedded sensors has lately made renowned development and is drawing growing attention in numerous application domains including machine learning, pattern recognition, context awareness, and human-centric sensing. Due to the lacking of a prominent analysis of this topic that can acquaint concomitant communities of the research avant-garde, there are still vital perspectives that, if pleaded, would create a vital turn in the way of interaction among people and mobile devices. In this paper, we have presented a comprehensive survey along with the prevailing state of various challenges of human activity recognition based on wearable, environmental, and smartphone sensors. Firstly, we have shown numerous factors to be considered for the data pre-processing part regarding noise filtering and segmentation methods. Besides, we have made a list of sensing devices, sensors, and applications that can be used for collecting activity data along with a discussion on sensor position and requirements. Moreover, we have made a comprehensive analysis of some benchmark datasets, which includes information about sensors, attributes, activity classes, etc. Finally, we have shown an analysis of activity recognition approaches on some of the benchmark datasets based on existing works.",
  "keywords": [
    "Benchmark datasets",
    "data pre-processing",
    "human activity recognition",
    "smartphone sensors",
    "survey",
    "wearable sensors"
  ],
  "paper_institutions": [
    "Department of Electrical and Electronic Engineering, University of Dhaka"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "ark-code-latest",
  "prompt_template": "simple",
  "extracted_at": "2026-08-16T19:46:21.856055Z",
  "template_tag": "simple",
  "summary": "### 摘要\n本文针对基于传感器的人体活动识别（HAR）领域进行了全面综述，重点分析了该领域面临的主要挑战，并对多个基准数据集进行了比较研究。文章首先介绍了HAR的基本流程，包括原始传感器数据的预处理（噪声滤波和分割）、特征提取和分类，并比较了常用的四种滤波器的性能。随后，文章系统梳理了可用的传感设备和工具，包括可穿戴设备、环境传感器和智能手机传感器。文章还详细列出了多个基准数据集，涵盖了基于可穿戴、医疗和智能手机传感器的数据集，并提供了它们的关键属性（如人数、活动数、传感器类型等）。最后，文章对比了已有研究在不同数据集上采用的分类方法，并讨论了未来研究方向，包括资源消耗优化、相似活动区分以及错误标注数据的处理。该综述旨在为HAR领域的研究人员提供全面的参考，并推动该领域向更实际和普适的场景发展。\n\n### 关键要点\n- 人体活动识别分为静态活动（如坐、站）、动态活动（如走路、跑步）和姿势转换活动，其中坐与站等相似静态活动在特征空间中重叠较大，难以区分。\n- 数据预处理中的噪声过滤通常采用中值滤波器、巴特沃斯低通滤波器、离散小波包收缩和卡尔曼滤波器，其中卡尔曼滤波器提供更高的信噪比（SNR）和更好的相关系数。\n- 分割窗口长度选择至关重要：较小窗口特征质量好但延迟识别结果，较大窗口可能错过短时活动，最优窗口长度应依据具体活动而定，常用范围为0.25秒至6.7秒，1秒和2秒窗口通常精度更优。\n- 特征提取包括时域特征（均值、标准差、信号幅值面积等）和频域特征（如FFT），特征选择方法有随机森林重要性、单变量特征选择、递归特征消除等。\n- 常用分类器包括线性判别分析（LDA）、逻辑回归、K近邻（KNN）、随机森林和支持向量机（SVM），其参数选择（如SVM的核函数、KNN的K值）显著影响精度，深度学习网络则存在处理时间与性能的权衡。\n- 传感器数量和使用位置是重要挑战：过多传感器增加用户负担，而手机位置不同（如口袋或手中）会导致相同活动数据变化，影响识别准确性。\n- 文章列举了多种传感设备，包括Motion Node、LoRaWAN、ECO传感器、μParts、TMote Sky、Smart-Its、Luna Nurse等，以及UCI机器学习库等数据源。\n- 基准数据集涵盖UCI ML仓库（如HAR、HAPT、OPPORTUNITY）、可穿戴数据集（如USC-HAD、WARD、SKODA）和医疗数据集（如Daphnet FoG、护理活动），以及智能手机数据集（如WISDM、HASC 2010）。\n- 已有工作采用不同方法，如基于腰戴加速度计的嵌入式处理（准确率90.8%）、1D Haar-like滤波（准确率93.91%）、核主成分分析加深信念网络（89.61%）等，深度学习与传统方法各有优劣。\n- 半监督和无监督方法（如自训练、K均值聚类）在标注数据少或混合静态动态活动时性能受限。\n- 未来研究应关注资源消耗（内存、CPU、电池）与识别精度的权衡，探索多分类器组合以区分相似活动，并处理错误标注数据及融合视频数据与传感器数据。\n- 该综述提供了关于传感器选择、数据集比较和分类方法的综合视角，为HAR应用（如健康监测、安全监控、智能家居）提供指导。",
  "available_templates": [
    "deep_read",
    "simple"
  ]
}