自己写程序利用lenet模型识别手写数字
来源:互联网 发布:淘宝客服催好评话术 编辑:程序博客网 时间:2024/04/28 05:48
自己编写程序,将手写图片送入训练得到的lenet模型,评估识别结果。
code:https://github.com/lhnows/caffeProjects
如果自己的caffe是用CMakeLists编译安装的,这样的话,可以运行如下的CMakeLists来编译自己的调用了caffe库的程序
CMakeLists.txt
cmake_minimum_required (VERSION 2.8)PROJECT (mnistTest)# Requires OpenCV v2.4.1 or later FIND_PACKAGE( OpenCV REQUIRED ) IF (${OpenCV_VERSION} VERSION_LESS 2.4.1) MESSAGE(FATAL_ERROR "OpenCV version is not compatible : ${OpenCV_VERSION}. requires atleast OpenCV v2.4.1") ENDIF() find_package(Caffe) include_directories(${Caffe_INCLUDE_DIRS}) add_definitions(${Caffe_DEFINITIONS}) add_executable(${PROJECT_NAME} mnistTest.cpp)include_directories ( /Users/liuhao/devlibs/deeplearning/caffe/install/include /usr/local/include /usr/local/cuda/include )target_link_libraries(${PROJECT_NAME} ${Caffe_LIBRARIES} ${OpenCV_LIBS} )
mnistTest.cpp
#define USE_OPENCV 1#define CPU_ONLY 1//貌似caffe有3种矩阵计算加速方式 mkl accelerate blas,本人Mac编译的可能是下面这种(其他会报错找不到头文件)//#define USE_ACCELERATE#include <iostream>#include <string>#include <caffe/caffe.hpp>#include <vector>#include <opencv2/core/core.hpp>#include <opencv2/highgui/highgui.hpp>#include <opencv2/imgproc/imgproc.hpp>#include "head.h"#include <algorithm>#include <iosfwd>#include <memory>#include <utility>#include <vector>using namespace caffe;using namespace cv;cv::Point previousPoint(-1, -1), nowPoint(-1, -1);Mat srcimage=Mat::zeros(280,280,CV_8UC1);Mat srcimageori = Mat::zeros(280, 280, CV_8UC1);class Classifier {public: Classifier(const string& model_file, const string& trained_file); int Classify(const cv::Mat& img);private: std::vector<int> Predict(const cv::Mat& img); void WrapInputLayer(std::vector<cv::Mat>* input_channels); void Preprocess(const cv::Mat& img, std::vector<cv::Mat>* input_channels);private: shared_ptr<Net<float> > net_; cv::Size input_geometry_; int num_channels_;};Classifier::Classifier(const string& model_file, const string& trained_file) {#ifdef CPU_ONLY Caffe::set_mode(Caffe::CPU);#else Caffe::set_mode(Caffe::GPU);#endif /* Load the network. */ net_.reset(new Net<float>(model_file, TEST)); net_->CopyTrainedLayersFrom(trained_file); CHECK_EQ(net_->num_inputs(), 1) << "Network should have exactly one input."; CHECK_EQ(net_->num_outputs(), 1) << "Network should have exactly one output."; Blob<float>* input_layer = net_->input_blobs()[0]; num_channels_ = input_layer->channels(); CHECK(num_channels_ == 3 || num_channels_ == 1) << "Input layer should have 1 or 3 channels."; input_geometry_ = cv::Size(input_layer->width(), input_layer->height());}/* Return the top N predictions. */int Classifier::Classify(const cv::Mat& img) { std::vector<int> output = Predict(img); std::vector<int>::iterator iter=find(output.begin(), output.end(), 1); int prediction = distance(output.begin(), iter); return prediction<10 ? prediction:0;}std::vector<int> Classifier::Predict(const cv::Mat& img) { Blob<float>* input_layer = net_->input_blobs()[0]; input_layer->Reshape(1, num_channels_, input_geometry_.height, input_geometry_.width); /* Forward dimension change to all layers. */ net_->Reshape(); std::vector<cv::Mat> input_channels; WrapInputLayer(&input_channels); Preprocess(img, &input_channels); net_->Forward(); /* Copy the output layer to a std::vector */ Blob<float>* output_layer = net_->output_blobs()[0]; const float* begin = output_layer->cpu_data(); const float* end = begin + output_layer->channels(); return std::vector<int>(begin, end);}void Classifier::WrapInputLayer(std::vector<cv::Mat>* input_channels) { Blob<float>* input_layer = net_->input_blobs()[0]; int width = input_layer->width(); int height = input_layer->height(); float* input_data = input_layer->mutable_cpu_data(); for (int i = 0; i < input_layer->channels(); ++i) { cv::Mat channel(height, width, CV_32FC1, input_data); input_channels->push_back(channel); input_data += width * height; }}void Classifier::Preprocess(const cv::Mat& img, std::vector<cv::Mat>* input_channels) { /* Convert the input image to the input image format of the network. */ cv::Mat sample; if (img.channels() == 3 && num_channels_ == 1) cv::cvtColor(img, sample, cv::COLOR_BGR2GRAY); else if (img.channels() == 4 && num_channels_ == 1) cv::cvtColor(img, sample, cv::COLOR_BGRA2GRAY); else if (img.channels() == 4 && num_channels_ == 3) cv::cvtColor(img, sample, cv::COLOR_BGRA2BGR); else if (img.channels() == 1 && num_channels_ == 3) cv::cvtColor(img, sample, cv::COLOR_GRAY2BGR); else sample = img; cv::Mat sample_resized; if (sample.size() != input_geometry_) cv::resize(sample, sample_resized, input_geometry_); else sample_resized = sample; cv::Mat sample_float; if (num_channels_ == 3) sample_resized.convertTo(sample_float, CV_32FC3); else sample_resized.convertTo(sample_float, CV_32FC1); cv::split(sample_float, *input_channels); CHECK(reinterpret_cast<float*>(input_channels->at(0).data) == net_->input_blobs()[0]->cpu_data()) << "Input channels are not wrapping the input layer of the network.";}static void on_Mouse(int event, int x, int y, int flags, void*){ if (event == EVENT_LBUTTONUP || !(flags&EVENT_FLAG_LBUTTON)) { previousPoint = cv::Point(-1,-1); } else if (event == EVENT_LBUTTONDOWN) { previousPoint = cv::Point(x, y); } else if (event == EVENT_MOUSEMOVE || (flags&EVENT_FLAG_LBUTTON)) { cv::Point pt(x, y); if (previousPoint.x<0) { previousPoint = pt; } line(srcimage, previousPoint, pt, Scalar(255), 16, 8, 0); previousPoint = pt; imshow("result", srcimage); }}int main(int argc, char** argv){ ::google::InitGoogleLogging(argv[0]);#ifdef CPU_ONLY Caffe::set_mode(Caffe::CPU);#else Caffe::set_mode(Caffe::GPU);#endif string model_file = "lenet.prototxt"; string trained_file = "lenet_iter_10000.caffemodel"; Classifier classifier(model_file, trained_file); std::cout << "------directed by watersink------" << std::endl; std::cout << "------------enter:退出-----------" << std::endl; std::cout << "--------------1:还原-------------" << std::endl; std::cout << "-------------2:写数字------------" << std::endl; std::cout << "-----lhnows@qq.com-----" << std::endl; imshow("result", srcimage); setMouseCallback("result", on_Mouse, 0); while (1) { char c = (char)waitKey(); if (c == 27) break; if (c=='1') { srcimageori.copyTo(srcimage); imshow("result", srcimage); } if (c == '2') { cv::Mat img; cv::resize(srcimage, img, cv::Size(28, 28)); CHECK(!img.empty()) << "Unable to decode image " << std::endl; int prediction = classifier.Classify(img); std::cout << "prediction:" << prediction << std::endl; imshow("result", srcimage); } } waitKey(); return 0;}
head.h
#include <caffe/common.hpp>#include <caffe/layer.hpp>#include <caffe/layer_factory.hpp>#include <caffe/layers/input_layer.hpp>#include <caffe/layers/inner_product_layer.hpp>#include <caffe/layers/dropout_layer.hpp>#include <caffe/layers/conv_layer.hpp>#include <caffe/layers/relu_layer.hpp>#include <caffe/layers/pooling_layer.hpp>#include <caffe/layers/softmax_layer.hpp> namespace caffe{ extern INSTANTIATE_CLASS(InputLayer); extern INSTANTIATE_CLASS(InnerProductLayer); extern INSTANTIATE_CLASS(DropoutLayer); extern INSTANTIATE_CLASS(ConvolutionLayer); //REGISTER_LAYER_CLASS(Convolution); extern INSTANTIATE_CLASS(ReLULayer); //REGISTER_LAYER_CLASS(ReLU); extern INSTANTIATE_CLASS(PoolingLayer); //REGISTER_LAYER_CLASS(Pooling); extern INSTANTIATE_CLASS(SoftmaxLayer); //REGISTER_LAYER_CLASS(Softmax);}
cmake.. & make
$ cmake ..-- The C compiler identification is AppleClang 8.1.0.8020042-- The CXX compiler identification is AppleClang 8.1.0.8020042-- Check for working C compiler: /Library/Developer/CommandLineTools/usr/bin/cc-- Check for working C compiler: /Library/Developer/CommandLineTools/usr/bin/cc -- works-- Detecting C compiler ABI info-- Detecting C compiler ABI info - done-- Detecting C compile features-- Detecting C compile features - done-- Check for working CXX compiler: /Library/Developer/CommandLineTools/usr/bin/c++-- Check for working CXX compiler: /Library/Developer/CommandLineTools/usr/bin/c++ -- works-- Detecting CXX compiler ABI info-- Detecting CXX compiler ABI info - done-- Detecting CXX compile features-- Detecting CXX compile features - done-- Found OpenCV: /usr/local (found version "3.2.0") -- Configuring done-- Generating done-- Build files have been written to: /Users/liuhao/projects/caffeProjects/mnistTest/build $ makeScanning dependencies of target mnistTest[ 50%] Building CXX object CMakeFiles/mnistTest.dir/mnistTest.cpp.o[100%] Linking CXX executable mnistTest[100%] Built target mnistTest
输入./mnistTests 执行
阅读全文
0 0
- 自己写程序利用lenet模型识别手写数字
- LeNet识别自己的手写数字
- 深度学习二:自己写java代码,识别手写数字
- 利用神经网络识别手写数字
- 剖析手写数字识别器LeNet-5认识卷积网络
- [TensorFlow实战] 构建LeNet实现手写数字识别
- 【深度学习】笔记3_caffe自带的第一个例子,Mnist手写数字识别所使用的LeNet网络模型的详细解释
- tensorflow 卷积神经网络 LeNet-5模型 MNIST手写体数字识别
- 手写体数字识别例程——LeNet-5模型
- arm9板子上写了一个小程序:手写数字及英文大小写字符的识别
- LeNet-5手写数字识别——可训练参数的求解详解
- [Keras实战] 构建LeNet实现手写数字识别(mnist数据集)
- 手写数字识别(1)---- Softmax回归模型
- 手写数字识别(2)---- CNN网络模型
- Tensorflow小样例-分类模型(识别mnist手写数字)
- caffe中如何训练自己的手写数字识别系统?
- 利用贝叶斯分类器实现手写数字识别
- knn-2 利用knn算法实现手写数字识别
- CSAPP笔记
- 主题七 最终的胜利----41.C语言程序员简历编写及面试技巧
- 简述MVC框架模式以及在你(Android)项目中的应用
- 小狗路程问题
- 为RecyclerView添加item的点击事件(个人备忘录)
- 自己写程序利用lenet模型识别手写数字
- Vue.js官方文档摘记
- 安卓登陆与注册
- 写了半天的C语言,感觉还是太难了,烦请中,
- 数据库系统概念(机械工业出版社,第七版)复习——第六章:数据库设计和E-R模型
- webService 入门
- Java中的知识点—(5)
- 两数之和
- JAVA环境的配置和Android Studio的安装