检索结果-内蒙古大学图书馆

Multi-object extraction technology for complex background based on faster regions-cnn algorithm in the context of artificial intelligence

引用

SERVICE ORIENTED COMPUTING AND APPLICATIONS 2025年第1期19卷 15-27页

作者： Peng, Shunsheng Hunan Coll Informat Dept Software Engn Changsha 410200 Peoples R China

With the rapid expansion of artificial intelligence technology, the trajectory of future battlefield operations is increasingly guided by intelligence. The modern combat environment is growing in complexity. An improved Fast Regions-Convolutional Neural Network algorithm is proposed for extracting multiple objects in intricate backgrounds to enhance the precision of striking enemy objects. Furthermore, to reduce the missed detection rate of objects, this research aims to integrate the backbone feature extraction network of the Residual Network with the Feature Pyramid Network. This integration seeks to optimize candidate regions and extract more detailed features of the object, thereby improving overall detection accuracy. This study notably enhances object detection algorithms in complex scenarios, particularly in military and security-sensitive fields, by integrating residual and feature pyramid networks into the Fast Regions-Convolutional Neural Network framework. The primary contribution is improving the algorithm's detection capability for small and multi-scale objects while bolstering its robustness against challenges like poor image quality, low contrast, and occlusion scenarios. Additionally, the study enhances the algorithm's generalization through data augmentation technology, ensuring high accuracy in diverse environments. These advancements contribute significantly to object detection techniques and hold practical value for applications like intelligent monitoring and autonomous driving. The data confirm that the enhanced Faster Regions-Convolutional Neural Network model achieves the highest accuracy at 98.21%, surpassing the original model by 13.47%. Following pruning, the optimized model demonstrates a remarkable 33.33% increase in the detection rate, coupled with a 1.58 times acceleration effect. The research on multi-object extraction technology in complex backgrounds, based on the Faster Regions-Convolutional Neural Network algorithm within artifici

关键词： Artificial intelligence fr-cnn algorithm Object extraction FPN ResNet Deep learning Multi-object detection Machine vision

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：