arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.12541cs.NEcs.AIeess.IV
收藏

基于振荡神经网络的异联想记忆图像边缘检测

Oscillatory Neural Network as Hetero-Associative Memory for Image Edge Detection

LIRMM(蒙彼利埃计算机科学、机器人学与微电子实验室);Univ. of Montpellier(蒙彼利埃大学);CNRS(法国国家科学研究中心)

Madeleine Abernot, Thierry Gil, Aida Todri-Sanial

摘要:

在边缘设备(如相机)上处理的数据量越来越多,这推动了边缘的人工智能(AI)集成。在边缘处执行的典型图像处理方法(例如特征提取或边缘检测)使用卷积滤波器,所述卷积滤波器是消耗能量、计算和存储器的算法。但是,边缘设备和摄像头的计算资源、带宽和功率都很有限,并且由于隐私限制,将数据发送到云端的能力也很有限。因此,需要在边缘处处理图像数据。多年来,这种需求激发了人们对在边缘实现类脑计算的兴趣。神经形态系统旨在模拟生物神经功能,以实现节能计算。近年来,振荡神经网络(Oscillatory Neural Networks,ONN)提出了一种新的脑激励计算方法,通过模拟脑振荡来执行自联想记忆类型的应用。为了加快图像边缘检测速度并降低功耗,我们对ONNs进行了深入的研究。提出了一种新的图像处理方法,利用ONNs作为异联想记忆(HAM)进行图像边缘检测。我们首先使用Matlab仿真器模拟我们的ONN-HAM解决方案,然后使用全数字ONN设计。我们在灰度正方形评估图上显示结果,也在黑白和灰度28 x28 MNIST图像上显示结果,最后在黑白512 x512标准测试图像上显示结果。我们比较我们的解决方案与标准的边缘检测滤波器,如Sobel和Canny。最后,使用全数字设计仿真结果,我们报告的时间和资源特性,并评估其可行性的实时图像处理应用。我们的数字ONN-HAM解决方案可以处理高达120 x120像素(166 MHz系统频率)的图像,满足实时相机的限制。这项工作是第一次探索ONNs作为异联想记忆的图像处理应用。

英文摘要:

The increasing amount of data to be processed on edge devices, such as cameras, has motivated Artificial Intelligence (AI) integration at the edge. Typical image processing methods performed at the edge, such as feature extraction or edge detection, use convolutional filters that are energy, computation, and memory hungry algorithms. But edge devices and cameras have scarce computational resources, bandwidth, and power and are limited due to privacy constraints to send data over to the cloud. Thus, there is a need to process image data at the edge. Over the years, this need has incited a lot of interest in implementing neuromorphic computing at the edge. Neuromorphic systems aim to emulate the biological neural functions to achieve energy-efficient computing. Recently, Oscillatory Neural Networks (ONN) present a novel brain-inspired computing approach by emulating brain oscillations to perform autoassociative memory types of applications. To speed up image edge detection and reduce its power consumption, we perform an in-depth investigation with ONNs. We propose a novel image processing method by using ONNs as a hetero-associative memory (HAM) for image edge detection. We simulate our ONN-HAM solution using first, a Matlab emulator, and then a fully digital ONN design. We show results on gray scale square evaluation maps, also on black and white and gray scale 28x28 MNIST images and finally on black and white 512x512 standard test images. We compare our solution with standard edge detection filters such as Sobel and Canny. Finally, using the fully digital design simulation results, we report on timing and resource characteristics, and evaluate its feasibility for real-time image processing applications. Our digital ONN-HAM solution can process images with up to 120x120 pixels (166 MHz system frequency) respecting real-time camera constraints. This work is the first to explore ONNs as hetero-associative memory for image processing applications.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑