...
首页> 外文期刊>Microprocessors and microsystems >Motion image processing system based on multi core FPGA processor and convolutional neural Network
【24h】

Motion image processing system based on multi core FPGA processor and convolutional neural Network

机译:基于多核FPGA处理器和卷积神经网络的运动图像处理系统

获取原文
获取原文并翻译 | 示例
           

摘要

The moving image processing can now be applied to the economic analysis of the imaging target's movement by the high-speed shooting. This method is, consider some of the common approaches. For this reason, the acknowledgement of the movement recognition and movement following are talked about. These projects, not successive autonomous moreover. As of late, Field-Programmable Gate Array (FPGA) is wide, particularly in the versatile and installed gadgets have been utilized in equipment quickening agent for the execution of Convolutional Neural Network (CNN). A CNN dependent on FPGA has been proposed. It is intended to assemble a neural organization convolution with the streamlining and advancement of the memory of the profoundly reusable quickening agent work is low equipment asset utilization. Movement, position, speed, and key data of the camera's objective can catch any ideal data from the caught casing can be shipped off the framework's investigation parts. To follow the moving objects by recognizing movement location is one of these smart frameworks. Even though there is another technique for identifying a moving there, these strategies, there are a few impediments for constant applications. Hence, to give precise outcomes with this strategy, the foundation deduction technique is reasonable for constant applications.
机译:现在可以将运动图像处理应用于通过高速射击对成像目标运动的经济分析。这种方法是,考虑一些常见方法。因此,讨论了对运动识别和移动的确认。这些项目,而不是连续自主。随着晚期,现场可编程门阵列(FPGA)宽,特别是在多功能和安装的小工具中,已经在设备加速剂中用于执行卷积神经网络(CNN)。已经提出了依赖于FPGA的CNN。它旨在通过精简和推进内存的简化和推进来组装神经组织卷积,以至于对深刻可重复使用的加速剂工作的记忆是低设备资产利用率。相机目标的移动,位置,速度和关键数据可以从捕获的外壳中捕获任何理想数据,可以从框架的调查部件上发货。通过识别移动位置遵循移动物体是这些智能框架之一。尽管存在另一种技术来识别那里的移动,但这些策略,恒定应用有一些障碍。因此,为了赋予该策略的精确结果,基础扣除技术对于持续应用是合理的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号