经过局部对比度标准化后的特征输入下采样层,为了保证不变性的同时获取较高的识别率,因此对池化层也进行多尺度的的变换,根据输出特征的维度相应调整池化层的采样间隔。其采样间隔越大输出特征映射图就越模糊,特征不变性就越强。经过下采样后将所有特征图传递给全连接层,用于下一步的目标分类识别。
2021-12-31 08:02:29 3KB 实测代码
1
基于HLS工具的CNN加速器的设计与优化方法研究
2021-12-30 22:41:49 600KB C/C++ HLS Xilinx Vivado
1
一.数据集准备 数据集共1400张机场或湖泊的图片,因此此分类为简单的二分类问题,通过CNN对数据集进行模型训练,得出相关指标。 数据集如下: 机场 湖泊 二.读取数据集 数据集路径 导入相关模块 import tensorflow as tf import numpy as np import matplotlib.pyplot as plt import pathlib #使用pathlib对路径对象进行管理 import random 构造路径对象,获取所有图片路径,并打乱数据集 pic_dir = 'D:/tensorflowDataSet/2_class' pic_ro
2021-12-30 16:45:14 301KB ens low ns
1
Tensorflow实现一个完整的CNN例子-附件资源
2021-12-30 16:30:00 23B
1
CNN可视化工具:用于深度卷积神经网络中的单元可视化的工具包 介绍 该存储库包含用于可视化深CNN内的单元(或称为神经元或卷积过滤器)的代码和结果。 一些代码已用于深度场景CNN中出现的物体检测器》。 您可以将此工具包与天真的一起使用,并可以将其编译为matcaffe和pycaffe或 。 该工具包包含以下功能: (新!)PyTorch脚本: pytorch_extract_feature.py :用于为任何给定图像在CNN模型的选定层提取CNN特征的代码。 pytorch_generate_unitsegments.py :用于生成所选层上所有单元的可视化的代码。 Matlab脚本: extract_features.m :用于为任何给定图像提取所选层上的CNN激活的代码。 generate_unitsegments.m :用于生成选定层中所有单元的可视化的代码。 unit
2021-12-30 15:51:53 11.67MB visualization cnn cnn-features C
1
R-CNN原理:   R-CNN遵循传统目标检测的思路,同样采取提取框,对每个框提取特征,图像分类,非极大值抑制等四个步骤,只不过在提取特征这一步将传统的特征换成了深度卷积网络提取的特征。  对于原始图像, 首先使用Selective Search 搜寻可能存在物体的区域。Selective Search 可以从图像中启发式地搜索出可能包含物体的区域。相比穷举而言, Selective Search 可以减少一部分计算量。下一步,将取出的可能含高物体的区域送入CNN 中提取特征。CNN 通常是接受一个固定大小的图像,而取出的区域大小却各有不同。对此, R-CNN的做法是将区域缩放到统一大小,
2021-12-30 12:46:37 289KB AS c cnn
1
1. Introduction about what is the Deep Learning 2. discss about Convolutional Neural Network 3. something about . ImageNet and ILSVRC
2021-12-30 11:00:06 4.53MB 深度学习 Deep Learning CNN
1
CNN图像分类 这个基于CNN的模型将图像分为9类(“飞机”,“汽车”,“鸟”,“猫”,“鹿”,“狗”,“青蛙”,“马”,“船”,“卡车”)使用tensorflow,keras,numpy,scikit-learn,matplotlib
2021-12-29 12:52:30 11KB JupyterNotebook
1
The first CNN appeared in the work of Fukushima in 1980 and was called Neocognitron. The basic architectural ideas behind the CNN (local receptive fields,shared weights, and spatial or temporal subsampling) allow such networks to achieve some degree of shift and deformation invariance and at the same time reduce the number of training parameters. Since 1989, Yann LeCun and co-workers have introduced a series of CNNs with the general name LeNet, which contrary to the Neocognitron use supervised training. In this case, the major advantage is that the whole network is optimized for the given task, making this approach useable for real-world applications. LeNet has been successfully applied to character recognition, generic object recognition, face detection and pose estimation, obstacle avoidance in an autonomous robot etc. myCNN class allows to create, train and test generic convolutional networks (e.g., LeNet) as well as more general networks with features: - any directed acyclic graph can be used for connecting the layers of the network; - the network can have any number of arbitrarily sized input and output layers; - the neuron’s receptive field (RF) can have an arbitrary stride (step of local RF tiling), which means that in the S-layer, RFs can overlap and in the C-layer the stride can differ from 1; - any layer or feature map of the network can be switched from trainable to nontrainable (and vice versa) mode even during the training; - a new layer type: softmax-like M-layer. The archive contains the myCNN class source (with comments) and a simple example of LeNet5 creation and training. All updates and new releases can be found here: http://sites.google.com/site/chumerin/projects/mycnn
2021-12-28 17:21:22 1.07MB CNN 卷积神经网络
1
听说pytorch使用比TensorFlow简单,加之pytorch现已支持windows,所以今天装了pytorch玩玩,第一件事还是写了个简单的CNN在MNIST上实验,初步体验的确比TensorFlow方便。 参考代码(在莫烦python的教程代码基础上修改)如下: import torch import torch.nn as nn from torch.autograd import Variable import torch.utils.data as Data import torchvision import time #import matplotlib.pyplot a
2021-12-28 17:16:58 47KB c IS mnist
1