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:崔丽群[1] 李万欣 CUI Li-qun;LI Wan-xin(Liaoning University of Engineering and Technology,School of Software,Huludao Liaoning 125105,China)
机构地区:[1]辽宁工程技术大学软件学院,辽宁葫芦岛125105
出 处:《计算机仿真》2025年第3期299-303,共5页Computer Simulation
基 金:辽宁省高等学校基本科研项目(LJKMZ20220699)。
摘 要:为使公共场所的安检效率达到更高水平,基于Cascade R-CNN,设计出X射线图像违禁品检测方法。以Cascade R-CNN网络为基础架构,针对该网络检测X射线图像违禁品时,构建由五个卷积层构成的Res Net骨架网络,在特征金字塔网络中添加平衡特征金字塔,整合各分辨率特征。面向候选区域生成网络的池化方式,提出区域特征整合策略,去除取整量化阶段。基于交并比,结合实际边缘与预测边缘的间距、重合度、尺度等因素,取得新的边缘回归损失函数,定位X射线图像中的违禁品,运用改进后的Cascade R-CNN实现X射线图像违禁品检测。仿真测试从视觉与量化数据两方面展开,结果表明所提方法具有较强的噪声免疫力,能清晰、有效地检测出所有违禁品,且检测精度与速率有明显的优越性,每秒能检测59张X射线图像,可投入实际的安检场景中应用。To achieve higher efficiency of security checks in public places,this paper put forward a method of detecting contraband under X-ray based on Cascade R-CNN.The Cascade R-CNN network served as the basic architecture at first,a Res Net backbone network consisting of five convolutional layers was constructed for detecting contrabands in X-ray images.Then,the balanced feature pyramid was added to the feature pyramid network to integrate features at various resolutions.Moreover,a regional feature integration strategy was proposed for the pooling method of the candidate region generation network,thus removing the stage of rounding quantization.Furthermore,considering the factors such as the distance between the actual edge and the predicted edge,contact ratio and scale,a new edge regression loss function was obtained based on the intersection over union,to locate contrabands in X-ray images.Finally,the improved Cascade R-CNN was used to detect the contrabands in X-ray images.Simulation was conducted from the perspectives of visual and quantitative data.Simulation results show that the proposed method has strong noise immunity and can detect all contrabands clearly and effectively.Meanwhile,the method has significant superiority in detection accuracy and rate,which is capable of detecting 59 X-ray images per second.Therefore,this method can be applied in practical security checks.
分 类 号:TP391[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:216.73.216.49