检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:谢尚志 陈维娜[2] 唐云祁[2] XIE Shang-zhi;CHEN Wei-na;TANG Yun-qi(Graduate School,People’s Public Security University of China,Beijing 102623,China;School of Investigation,People’s Public Security University of China,Beijing 102623,China)
机构地区:[1]中国人民公安大学研究生院,北京102623 [2]中国人民公安大学侦查学院,北京102623
出 处:《印刷与数字媒体技术研究》2025年第1期110-117,共8页Printing and Digital Media Technology Study
基 金:中央高校基本科研业务费项目资助——结合深度学习与反射变换成像技术的形态类物证智能识别(No.2024JKF09)。
摘 要:为了实现法庭科学文件打印方式的自动化识别,本研究探究了网络模型对识别准确率的影响,确定可实现检验目标的文件扫描分辨率。本研究利用不同品牌、型号的激光打印机和喷墨打印机,制备了不同打印方式的文件共100份。这100份文件依次通过1200dpi、800dpi、600dpi、400dpi、300dpi、200dpi分辨率扫描后,裁切分割,制作出152976份样本。将样本随机打乱后,按照训练集∶验证集∶测试集=0.64∶0.16∶0.2的比例送入LeNet-5、AlexNet、VGG16、ResNet18这4种经典卷积神经网络模型中。4种网络模型对300dpi及以上分辨率的训练和测试准确率均能得到90%以上;经过训练后,ResNet18模型对随机寻找的10份未知样本的识别准确率可达100%。实验结果表明,卷积神经网络可以作为文件打印方式识别的一种有效手段。综合分析实验所用的4种网络模型,ResNet18识别速度快、准确率高,是其中的最优选择。In order to achieve automated recognition of scientific document printing methods in the court,in this study,the impact of network models on recognition accuracy was explored,and the scanning resolution of documents that can achieve inspection objectives was determined.A total of 100 documents with different printing methods were prepared using laser printers and inkjet printers of different brands and models in the study.These 100 files were scanned at resolutions of 1200dpi,800dpi,600dpi,400dpi,300dpi,and 200dpi,and then cut and segmented to produce 152976 samples.After randomly shuffling the samples,they were fed into four classic convolutional neural network models of LeNet-5,AlexNet,VGG16,and ResNet18 in a ratio of train∶val∶test=0.64∶0.16∶0.2.Four network models could achieve training and testing accuracy of over 90%for resolutions of 300dpi and above.After training,the ResNet18 model achieved a recognition accuracy of 100%for 10 randomly searched unknown samples.The experimental results indicated that convolutional neural networks can be an effective means of recognizing file printing methods.After comprehensive analysis of the four network models used in the experiment,ResNet18 has fast recognition speed and high accuracy,making it the optimal choice.
分 类 号:TS8[轻工技术与工程] TP399[自动化与计算机技术—计算机应用技术]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.33