Results 81 to 90 of about 380 (97)
Research on Pedestrian Detection in Static Images [PDF]
, 2010 行人检测作为目标检测的重要分支，是计算机视觉领域的前沿方向和研究热点，在智能视频监控、汽车辅助驾驶、高级人机接口及人体运动分析等众多领域有着广泛的应用前景。由于行人检测涉及多门学科知识，且受穿着、光照、姿态、尺度、视角和复杂背景等多种因素的影响，至今仍然没有取得令人满意的效果。本文围绕静态图像的行人检测展开研究，创新点主要包含两个部分： 1、针对现有基于整体的行人检测方法需要大量标记样本的缺点，提出了一种基于迁移学习的半监督行人分类方法：首先通过稀疏编码，从任意的未标记样本中学习到一个紧凑 ...
谢尧芳
core
Survey of Uyghur Machine Translation Research [PDF]
As one of the important tasks in China's low-resource machine translation research, the development and application of Uyghur machine translation can better promote cultural exchanges and trade between different regions and ethnic groups.However, Uyghur,
Halidanmu ABUDUKELIMU, Yutao HOU, Dengfeng YAO, Abudukelimu ABULIZI, Jishang CHEN
core +1 more source
Program semantic analysis model for code reuse detection [PDF]
Program similarity analysis had a wide range of applications in areas such as code plagiarism and property protection, but it generally suffered from problems such as excessive computational overhead, a code similarity analysis method based on fuzzy ...
GUO Xi, WANG Pan
core +1 more source
Ripeness identification of pitaya fruit based on YOLOv8 and PSP-Ellipse [PDF]
ObjectiveImprove the accuracy and robustness of maturity detection of pitaya fruit.MethodsCombining the YOLOv8 object detection model with the PSP-Ellipse segmentation algorithm, an efficient and accurate automatic identification method for pitaya fruit ...
FENG Chao +3 more
core +1 more source
Gaoya dianqi配电网单相接地故障占比较高，存在多种故障类型。如果对不同类型的接地故障进行适当的处理，能有效提高配电网的可靠性和可用性。然而，如何快速区分不同的故障类型，一直以来是一个研究热点。本研究针对高阻接地故障电流和电压特征不明显导致的高阻接地故障检测困难的问题，提出了一种创新的解决方法，即采用迁移学习和深度残差网络对高阻接地故障进行识别。首先，提取单相接地故障后2个周期内的零序电压数据，利用小波变换将零序电压映射为二维时频图；其次，搭建深度残差网络，通过冻结底层参数实现小样本迁移，微调参数对不同的故障样本分类 ...
周庭栋 +4 more
doaj
短文本情感分析的研究现状——从社交媒体到资源稀缺语言 [PDF]
, 2018 随着社交网络的逐渐成熟,各类语种的文本出现在社交网络上。而这些非规范的短文本蕴藏着人们对事物的褒贬、需求等意见,是国家政府和企业了解公众舆论的重要参考信息,具有重大的研究价值和应用价值。首先,对目前互联网短文本情感分析领域常用的神经网络、跨语言和应用语言学知识等研究方法进行归纳和总结;其次,对当前短文本情感分析研究的热点领域——社交媒体和资源稀缺语言的情感分析进行现状分析;最后,对短文本情感分析研究的趋势进行总结,分析存在的问题,并对未来进行展望。国家自然科学基金项目(61262086 ...
史晓东, 尼玛扎西, 拥措
core
Micro-expression recognition based on self-supervised masked optical flow(基于自监督掩码光流的人脸微表情识别)
Zhejiang Daxue xuebao. Lixue banIn the field of micro-expression recognition, existing feature extraction methods primarily rely on supervised learning techniques, which limits the model's generalizability on feature representation and their application scenarios. To address this issue,
刘晓宇(LIU Xiaoyu) +2 more
doaj +1 more source
Coal gangue image recognition model based on CSPNet-YOLOv7 target detection algorithm [PDF]
The gangue recognition technology is one of the key technologies in the intelligent construction of mines. To address the problem of low accuracy of the gangue recognition model caused by low illumination and high dust environment at the working face and
Chao LIU +4 more
core +1 more source
Research on Food Image Classification Algorithm based on Improved MobileNetV3-Large [PDF]
Food image recognition plays a crucial role in food safety monitoring, nutritional analysis, and dietary recommendation systems. However, the diversity, complexity, and external factors such as lighting conditions pose numerous difficulties and ...
HE Wei-chan, QIN Jing-hui, YANG Zhi-jing
core +1 more source
基于迁移学习和红外热成像的金属氧化物避雷器表面污染状态检测方法
Dianci bileiqi金属氧化物避雷器外壳受灰尘、盐、碱等污染物长期沉积的影响，会导致其过早失效，因此，准确地检测金属氧化物避雷器表面的污染状态是非常重要的。基于此，面向饱和积污提出了一种利用红外热成像技术和迁移学习检测金属氧化物避雷器外壳污染严重程度的新方法。该方法通过拍摄不同污染状态下金属氧化物避雷器的红外热图像，经过适当的预处理后，捕获的红外热图像被送入预训练卷积神经网络“ResNet50”进行自动特征提取。提取的深度特征被输入到4个机器学习分类器，即k-最近邻分类器、支持向量机分类器 ...
沈顺群 +5 more
doaj