拥有生物激励神经感知引擎的实时多目标识别处理器
《拥有生物激励神经感知引擎的实时多目标识别处理器》由会员分享,可在线阅读,更多相关《拥有生物激励神经感知引擎的实时多目标识别处理器(14页珍藏版)》请在装配图网上搜索。
1、拥 有 生 物 激 励 神 经 感 知 引 擎 的 实 时 多 目 标 识 别 处 理 器 object recognitionFirst, various scale spaces are generated by a cascaded filtering for input video stream.Then, key-points are extracted among neighbor scale spaces by local maxima/ minima search, and each of them is converted to a descriptor vector tha
2、t describes the magnitude and orientation of it.Last, the final recognition is made by nearest neighbor matching with pre-defined object database that generally includes over ten thousands of object descriptor vectors. Disadvantage of SI MD processorstheir identical operations are not suitable for k
3、ey-point or object level operations such as descriptor vector generation and database matching.the multi-core processor of exploits coarse-grained PEs and memory-centric network-on-chip (NoC) for task-level parallelism over data-level parallelism cannot provide enough computing power for real-time o
4、bject recognition due to its data synchronization overhead. The papers workSection I I describes a visual perception based multi-object recognition algorithm in detail.Section I I I explains system architecture of the proposed processor.Detailed designs of each building block are explained in Sectio
5、n I V.Section V describes the architecture. proposed NoC communication The chip implementation and evaluation results follow in Section VI . I I . visual perception based multi-object recognition A. Visual Perception Based Object Recognition Model B. Overall Algorithm(注 : (ROI s)regions-of-interest) I I I . system architecture I V. building block designA. Neural Perception Engine B. SI MD Processor Unit C. Decision Processor V. proposed NoC communication VI . Low-power techniques Thanks for your attention
- 温馨提示:
1: 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
2: 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
3.本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
5. 装配图网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。